Sensing systems and methods for providing kidney health and / or diabetes decision support - Patents.com

Abstract

Certain aspects of the present disclosure relate to methods and systems for providing decision support for renal disease. In certain aspects, the method includes monitoring one or more analytes of a patient during a plurality of time periods to obtain analyte data, the one or more analytes including at least potassium, and the analyte data including potassium data, processing the analyte data from the plurality of time periods to determine at least one rate of change of potassium for the patient based on the potassium data, and generating a disease prediction using the analyte data for the one or more analytes including the potassium data and the at least one rate of change of potassium for the patient.

Description

[Technical field]

[0001] (CROSS REFERENCE TO RELATED APPLICATIONS) This application claims priority to and the benefit of U.S. Provisional Patent Application No. 63 / 268,417, filed February 23, 2022, U.S. Provisional Patent Application No. 63 / 377,332, filed September 27, 2022, U.S. Provisional Patent Application No. 63 / 403,568, filed September 2, 2022, and U.S. Provisional Patent Application No. 63 / 403,582, filed September 2, 2022, which are assigned to the assignee hereof and are expressly incorporated by reference in their entireties for all applicable purposes as if fully set forth below. [Background technology]

[0002] The kidneys perform many important functions in the human body, including filtering waste products and excess fluids that are excreted in urine, and removing acids produced by the body's cells to maintain a healthy balance of water, salts, and minerals (e.g., sodium, calcium, phosphorus, and potassium, etc.) in the blood. In other words, the kidneys play a major role in homeostasis by transporting water, salts, and minerals and regulating their secretion, reabsorption, and excretion through the renal mechanisms. In addition, the kidneys secrete renin (e.g., angiotensinogenase), which forms part of the renin-angiotensin-aldosterone system (RAAS), which mediates extracellular fluid and arterial vasoconstriction (e.g., blood pressure). More specifically, high blood pressure (e.g., hypertension) can be regulated through RAAS inhibitors, such as angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs). When the kidneys become diseased or damaged, the impairment or loss of these functions can cause significant damage to the human body.

[0003] Kidney disease occurs when the kidneys become diseased or damaged. Kidney disease is generally classified as either acute or chronic based on the duration of the disease. Acute kidney injury (AKI) (also called "acute kidney failure") is usually caused by an event that leads to kidney failure, such as dehydration, blood loss due to major surgery or injury, and / or the use of medicines. Chronic kidney disease (CKD), on the other hand, is usually caused by a long-term disease, such as hypertension or diabetes, that gradually damages the kidneys and reduces their function over time.

[0004] Symptoms of reduced kidney function, such as fluid accumulation or electrolyte imbalance, are more likely to occur with AKI, regardless of how long the kidneys have been incapacitated. Such symptoms may reflect the actual cause of the kidney problem. For example, dehydration may cause extreme dryness, drowsiness or fainting, and / or a weak and rapid pulse. On the other hand, symptoms of CKD may develop slowly and may not become apparent until there is little kidney function left. Depending on the severity of kidney disease in a person, loss of kidney function may cause numbness, tingling, shortness of breath, muscle weakness, and / or arrhythmia, to name a few. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] U.S. Patent No. 6,015,572 [Patent Document 2] U.S. Patent No. 5,964,745 [Patent Document 3] U.S. Patent No. 6,083,523 [Patent Document 4] U.S. Provisional Patent Application No. 63 / 321340 [Patent Document 5] U.S. Patent No. 7,310,544 [Patent Document 6] U.S. Patent No. 6,931,327; [Patent Document 7] US Patent Application Publication No. 2005 / 0043598 [Patent Document 8] US Patent Application Publication No. 2007 / 0032706 [Patent Document 9] US Patent Application Publication No. 2007 / 0016381 [Patent Document 10] US Patent Application Publication No. 2008 / 0033254 [Patent Document 11] US Patent Application Publication No. 2005 / 0203360 [Patent Document 12] US Patent Application Publication No. 2005 / 0154271 [Patent Document 13] US Patent Application Publication No. 2005 / 0192557 [Patent Document 14] US Patent Application Publication No. 2006 / 0222566 [Patent Document 15] US Patent Application Publication No. 2007 / 0203966 [Patent Document 16] US Patent Application Publication No. 2007 / 0208245 Summary of the Invention [Means for solving the problem]

[0006] In some cases, elevated potassium levels in patients with CKD can lead to hyperkalemia, while lower potassium levels in patients with CKD can lead to hypokalemia. In particular, hyperkalemia is a medical term that describes blood potassium levels that are higher than normal (e.g., higher than the normal blood potassium levels of 3.6 to 5.2 millimoles per liter (mmol / L)). Hyperkalemia increases the risk of cardiac arrhythmia episodes and sudden death. Symptoms of consistent or mild hyperkalemia can be muscle weakness, numbness, tingling, nausea, or other abnormal sensations, or the patient may not notice any symptoms at all. Prominent symptoms of highly elevated potassium levels include palpitations, shortness of breath, chest pain, nausea, or vomiting. Hypokalemia, on the other hand, is a medical term that describes blood potassium levels that are lower than normal. In particular, CKD patients may develop hypokalemia due to gastrointestinal potassium loss from diarrhea or vomiting, or renal potassium loss from non-potassium sparing diuretics (e.g., diuretics used to increase the amount of fluid passing from the body into the urine, regardless of the amount of potassium lost from the body into the urine). The lack of easily noticeable symptoms in many patients is why hyperkalemia and hypokalemia are often referred to as silent killers, especially when patients are hypersensitive to irregular potassium levels. Severe hypokalemia and hyperkalemia can lead to severe symptoms of respiratory failure, sudden cardiac death, or other high mortality events, for example, the diagnosis of hypokalemia or hyperkalemia, as the mediating mechanism, may not be immediately apparent by the time the patient is evaluated by medical personnel.

[0007] Furthermore, CKD may alter the glucose homeostasis of patients, thereby making CKD an independent risk factor for hypoglycemia and hyperglycemia. In particular, the kidney also plays an important role in regulating blood glucose (e.g., blood glucose). With regard to the kidney's involvement in glucose homeostasis, the main mechanisms include the release of glucose into circulation via gluconeogenesis, the uptake of glucose from circulation to meet the energy demand of the kidney, and the reabsorption of glucose at the level of the proximal tubule of the kidney. For example, gluconeogenesis is the formation of glucose from precursors (e.g., lactate, glycerol, and / or amino acids). Glucose is formed and released into the circulation by the kidney. Gluconeogenesis occurs to maintain glucose homeostasis by preventing low blood glucose.

[0008] However, as renal function declines, so does glucose formation, thus limiting the kidney's ability to respond to a drop in blood glucose. In some cases, the reduced ability of the kidney to respond to a drop in blood glucose levels can lead to hypoglycemia. Hypoglycemia is a medical term that describes blood glucose (e.g., glucose) levels lower than normal (e.g., blood glucose levels below 70 milligrams per deciliter (mg / dL), or 3.9 millimoles per liter (mmol / L)). 1.7% of annual hospitalizations are due to hypoglycemia for early stage CKD (CKD<3), and for CKD stages 3 and 4, the annual hospitalization rate is 2.3%. Furthermore, for end stage renal disease (ESRD)-related hospitalizations, 4.3% are due to hypoglycemia, with a mortality rate of 30%. In particular, severe hypoglycemia can lead to myocardial damage, neurocognitive dysfunction, and in certain cases, even stroke or death.

[0009] Thus, renal dysfunction may result in lower and / or longer low glucose levels. Impaired kidneys may be slower and in some cases less effective at countering the drop in glucose levels, resulting in poor glucose control (e.g., considering that the kidneys filter insulin and generate glucose by gluconeogenesis). Hypoglycemic events in renal disease are due to reduced insulin clearance by the kidney and impaired gluconeogenesis. For ESRD-related hospitalization, dialysis is also an aggravating factor.

[0010] Furthermore, decreased insulin metabolism and clearance may occur as a result of declining kidney health. Insulin is a hormone that allows the body to use glucose as energy or store glucose as fat. In other words, insulin stimulates potassium and glucose uptake by the patient's cells, reducing serum (e.g., extracellular) potassium and glucose levels. Insulin is removed by the kidney, and therefore, as renal function declines, insulin is removed more slowly. Thus, a typical dose of insulin may have a prolonged and / or pronounced effect on glucose in patients with renal dysfunction. As renal dysfunction progresses, insulin may have an even more prolonged and / or pronounced effect on glucose. Thus, patients with renal disease may be at risk for hypoglycemia, and the risk increases with disease progression. For example, patients with CKD may find that insulin doses that predict glucose clearance at some later time result in.

[0011] Renal dysfunction may also result in higher and / or longer high glucose levels. For example, considering that the kidney is the response to filter, reabsorb, and consume glucose from the blood, impaired kidneys may be slower and in some cases less effective at reducing glucose levels (i.e., removing glucose). Patients suffering from higher and / or longer high glucose levels may be diagnosed with hyperglycemia. Hyperglycemia is a medical term that describes blood sugar (e.g., glucose) levels higher than normal (e.g., significantly elevated blood sugar levels, usually elevated above 180-200 mg / dL, or 10-11.1 mmol / L). Patients with hyperglycemia may suffer from many negative physiological effects associated with microvascular deterioration (e.g., nerve damage (neuropathy), kidney failure, skin ulcers, diabetic ketoacidosis, or bleeding into the vitreous humor of the eye).

[0012] Other problems such as increased phosphate levels in the blood (e.g., hyperphosphatemia) may also develop in CKD. Hyperphosphatemia (e.g., abnormally high serum phosphate levels) may result from increased phosphate intake, decreased phosphate excretion, or disorders that shift intracellular phosphate to the extracellular space. This increase in serum phosphate levels is associated with decreased renal ion excretion and the use of medicines to reduce the progression of CKD or to control related diseases such as diabetes mellitus and heart failure.

[0013] So that the above-listed features of the present disclosure can be understood in detail, a more particular description, briefly summarized above, can be made by reference to embodiments, some of which are illustrated in the drawings. It should be noted, however, that the attached drawings merely illustrate certain exemplary embodiments of the present disclosure and therefore should not be considered as limiting its scope, since the description may be allowed for other equally valid embodiments. [Brief description of the drawings]

[0014] [Figure 1]FIG. 1 illustrates aspects of an exemplary decision support system that may be used in connection with implementing embodiments of the present disclosure. [Diagram 2] FIG. 1 conceptually illustrates an exemplary continuous analyte monitoring system including an exemplary continuous analyte sensor having sensor electronics, in accordance with certain aspects of the present disclosure. [Diagram 3] 2 illustrates example inputs and example metrics calculated based on the inputs for use by the decision support system of FIG. 1 in accordance with some embodiments disclosed herein. [Figure 4A] 1 is an exemplary workflow for determining the likelihood of one or more symptoms commonly associated with potassium imbalance occurring in real time or within a specified time period using one or more models or rules, and retraining or updating the one or more models or rules, according to certain embodiments of the present disclosure. [Figure 4B] 1 is an exemplary workflow for determining the likelihood of one or more symptoms commonly associated with potassium imbalance occurring in real time or within a specified time period and providing one or more recommendations for treatment for a user based on the determined likelihood, in accordance with certain embodiments of the present disclosure. [Diagram 5] FIG. 1 is a flow diagram illustrating an exemplary method for providing decision support using a continuous analyte sensor including at least a continuous potassium sensor, according to some exemplary aspects of the present disclosure. [Figure 6] FIG. 1 is a flow diagram depicting a method for training a machine learning model to provide predictions associated with renal disease, according to certain embodiments of the present disclosure. [Figure 7] 1 is an exemplary workflow for stratifying a user's risk of diseases commonly associated with renal disease, including hyperkalemia and hypokalemia, according to certain embodiments of the present disclosure. [Figure 8A] FIG. 1 illustrates an exemplary method used to provide renal disease decision support to a user determined to be at risk for hyperkalemia, in accordance with certain embodiments of the present disclosure. [Figure 8B] FIG. 1 illustrates an exemplary method used to provide renal disease decision support to a user determined to be at risk for hyperkalemia, in accordance with certain embodiments of the present disclosure. [Figure 9] FIG. 1 illustrates an exemplary method used to provide renal disease and / or glucose homeostasis treatment recommendations to a user determined to be at risk for hyperkalemia, in accordance with certain embodiments of the present disclosure. [Figure 10] FIG. 1 illustrates an exemplary method used to provide dietary recommendations to a user determined to be at risk for hyperkalemia, in accordance with certain embodiments of the present disclosure. [Figure 11] FIG. 1 illustrates an exemplary method used to provide renal disease decision support to a user determined to be at risk for hypokalemia, in accordance with certain embodiments of the present disclosure. [Figure 12] FIG. 1 illustrates an exemplary method used to provide renal disease and / or glucose homeostasis treatment recommendations to a user determined to be at risk for hypokalemia, in accordance with certain embodiments of the present disclosure. [Figure 13] FIG. 1 illustrates an exemplary method used to provide dietary recommendations to a user determined to be at risk for hypokalemia, in accordance with certain embodiments of the present disclosure. [Figure 14] FIG. 14 is a block diagram illustrating a computing device configured to perform the operations of FIGS. 4A-13 in accordance with certain embodiments of the present disclosure. [Figure 15A] FIG. 1 depicts an exemplary enzyme domain configuration for a continuous multi-analyte sensor, according to certain embodiments of the present disclosure. [Figure 15B] FIG. 1 depicts an exemplary enzyme domain configuration for a continuous multi-analyte sensor, according to certain embodiments of the present disclosure. [Figure 15C] FIG. 1 depicts an exemplary enzyme domain configuration for a continuous multi-analyte sensor, according to certain embodiments of the present disclosure. [Figure 15D]FIG. 1 depicts an exemplary enzyme domain configuration for a continuous multi-analyte sensor, according to certain embodiments of the present disclosure. [Figure 15E] FIG. 1 depicts an exemplary enzyme domain configuration for a continuous multi-analyte sensor, according to certain embodiments of the present disclosure. [Figure 16A] FIG. 1 depicts an alternative view of an exemplary dual electrode enzyme domain configuration for a continuous multi-analyte sensor, according to certain embodiments of the present disclosure. [Figure 16B] FIG. 13 is an alternative diagram of an exemplary dual electrode enzyme domain configuration for a continuous multi-analyte sensor, according to certain embodiments of the present disclosure. [Figure 16C] FIG. 13 is an alternative diagram of an exemplary dual electrode enzyme domain configuration for a continuous multi-analyte sensor, according to certain embodiments of the present disclosure. [Figure 16D] FIG. 13 is an alternative diagram of an exemplary dual electrode enzyme domain configuration for a continuous multi-analyte sensor, according to certain embodiments of the present disclosure. [Figure 16E] FIG. 1 depicts an exemplary dual electrode configuration for a continuous multi-analyte sensor, in accordance with certain embodiments of the present disclosure. [Figure 17A] FIG. 1 depicts an exemplary enzyme domain configuration for a continuous multi-analyte sensor, according to certain embodiments of the present disclosure. [Figure 17B] FIG. 1 depicts alternative exemplary enzyme domain configurations for a continuous multi-analyte sensor, according to certain embodiments of the present disclosure. [Figure 17C] FIG. 1 depicts alternative exemplary enzyme domain configurations for a continuous multi-analyte sensor, according to certain embodiments of the present disclosure. [Figure 18] FIG. 1 depicts an exemplary enzyme domain configuration for a continuous multi-analyte sensor, according to certain embodiments of the present disclosure. [Figure 19A] FIG. 13 is an alternative diagram of exemplary dual electrode enzyme domain configurations G1-G4 for a continuous multi-analyte sensor, according to certain embodiments of the present disclosure. [Figure 19B] FIG. 13 is an alternative diagram of exemplary dual electrode enzyme domain configurations G1-G4 for a continuous multi-analyte sensor, according to certain embodiments of the present disclosure. [Figure 19C] FIG. 13 is an alternative diagram of exemplary dual electrode enzyme domain configurations G1-G4 for a continuous multi-analyte sensor, according to certain embodiments of the present disclosure. [Figure 19D] FIG. 13 is an alternative diagram of exemplary dual electrode enzyme domain configurations G1-G4 for a continuous multi-analyte sensor, according to certain embodiments of the present disclosure. [Figure 20A] FIG. 1 is a diagram illustrating a schematic configuration and components of a device for measuring electrophysiological signals and / or concentrations of target ions in a biological fluid in vivo, according to certain embodiments of the present disclosure. [Figure 20B] FIG. 1 is a diagram illustrating a schematic configuration and components of a device for measuring electrophysiological signals and / or concentrations of target ions in a biological fluid in vivo, according to certain embodiments of the present disclosure. [Figure 21] FIG. 13 is a diagram illustrating, in accordance with certain embodiments of the present disclosure, additional exemplary configurations and components of a device for measuring electrophysiological signals and / or concentrations of target ions in biological fluids in vivo. [Figure 22A] FIG. 1 is a diagram illustrating a schematic configuration and components of a device for measuring electrophysiological signals and / or target analyte concentrations in biological fluids in vivo, according to certain embodiments of the present disclosure. [Figure 22B] 1A-1C are schematic illustrations of exemplary configurations and components of a device for measuring electrophysiological signals and / or target analyte concentrations in biological fluids in vivo, according to certain embodiments of the present disclosure; [Figure 22C] FIG. 1 is a diagram illustrating a schematic configuration and components of a device for measuring electrophysiological signals and / or target analyte concentrations in biological fluids in vivo, according to certain embodiments of the present disclosure. [Figure 23] FIG. 1 depicts an exemplary continuous analyte monitoring system configured to measure target ions and / or other analytes as discussed herein, in accordance with certain embodiments of the present disclosure.

[0015] For ease of understanding, wherever possible, the same reference numerals have been used to designate identical elements that are common to each of the figures. It is contemplated that elements disclosed in one embodiment may be beneficially utilized in other embodiments without specific recitation. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] Accurate assessment of renal function (and in some cases cardiac function) is important as a screening tool and for monitoring disease progression and guiding prognosis, at least for renal disease, as well as hyperkalemia, hypokalemia, hyperglycemia, and hypoglycemia. However, conventional disease diagnostic methods and systems for such diseases, including but not limited to electrocardiogram (ECG) monitoring, albumin-to-creatinine ratio (ACR) tests, glomerular filtration rate (GFR) tests, and blood tests to monitor potassium levels in patients, face many challenges in terms of accuracy and reliability. Moreover, conventional disease diagnostic methods and systems generally fail to provide an efficient and complete analysis of factors that may potentially contribute to each of these diseases.

[0017] For example, ECG monitoring has been evaluated as a method used to recognize the proarrhythmic effects of severe hyperkalemia, such as peak T waves, QRS widening, PR shortening (e.g., the PR interval is the time from the onset of the P wave (atrial depolarization) to the onset of the QRS complex (ventricular depolarization); a shortened PR interval may indicate certain disorders), and bradycardia (e.g., a slower than expected heart rate). However, by the time an ECG device is able to recognize the proarrhythmic effects of severe hyperkalemia or hypokalemia, the patient may be in a severe hyperkalemia or hypokalemia situation. Similarly, by the time an ECG device recognizes the proarrhythmic effects of severe hypokalemia (e.g., ST segment depression, U waves, T wave morphology changes), the patient is already in a dangerous situation. Thus, the only treatment that may be available may be treatment provided to the patient in an emergency room environment. Such treatment may not always be feasible or may not be performed in a sufficient time, thereby resulting in undesirable outcomes, in some cases, the death of the patient. A further treatment paradox exists in that ECG devices, or the ability to interpret such ECG devices, are not readily accessible to most patients, for example in their own homes, and thus a patient's only reasonable option in a severe disease state may be to present to emergency personnel.

[0018] Furthermore, evidence is conflicting regarding whether in-clinic ECG findings are reliable, especially in patients with chronic hyperkalemia. For example, high false positives and high false negatives are often seen with the use of ECG monitoring. As used herein, a false positive is a result that indicates a given condition is present when that condition is not present, and a false negative is a result that indicates a given condition is not present when that condition is actually present. Furthermore, such monitoring lacks the capacity for continuous monitoring required to provide a complete picture of a patient's health.

[0019] Additionally, in some cases, T-wave abnormalities detected by ECG monitoring may be due to factors not related to potassium. For example, T-wave abnormalities may be present in subarachnoid hemorrhage, ischemic stroke, subdural hematoma, heart failure, myocardial edema, viral infections (e.g., Covid-19), traumatic brain injury, rare diseases, and certain tumor pathways, such as, but not limited to, pheochromocytoma. The value of using ECG alone to diagnose clinically significant hyperkalemia is further complicated by situations in which T-wave abnormalities are a comorbidity of renal disease. In these cases, it is even more difficult to determine whether the result of the T-wave abnormality is the result of renal disease or another clinical situation. Thus, T-wave abnormalities recognized by ECG monitoring may not always be caused by reduced renal function (e.g., abnormal potassium levels may be due to reduced renal function in some cases). Therefore, ECG alone may not provide sufficient information to assess renal health, nor may it provide sufficient information regarding the extent of renal disease in the patient's body.

[0020] As another example, ACR and GFR tests are commonly used to screen and diagnose kidney disease. In particular, the ACR test measures both albumin and creatinine in a one-time sample, also known as a random urine sample. The urine albumin test is the preferred first method for detecting elevated proteins in urine, specifically albumin. Although a sustained increase in urinary albumin levels, i.e., increased albumin excretion, measured using the ACR test provides a marker of kidney damage, the test is not without flaws. For example, the current use of the ACR test does not take into account the effect of creatinine excretion, which can lead to inaccurate results and, as a result, misleading results. Furthermore, random urine samples for ACR are more vulnerable to transient changes in creatinine and albumin excretion than timed collections that average such changes over a longer period of time, and thus random random urine ACR results are less consistent than timed urine samples.

[0021] GFR, on the other hand, is the flow rate of filtered fluid through the kidney. Creatinine clearance rate is the volume of plasma from which creatinine is removed per unit time and is used to estimate GFR. GFR can be measured using gold standard methods (e.g., measured GFR (mGFR)) or estimated using a formula (e.g., eGFR). Although eGFR provides a more convenient and rapid analysis for assessing renal function, the equation for estimating GFR has been found to have limitations and is not generalizable across all populations. Therefore, more frequent use of measured GFR tests is recommended. However, determining mGFR requires the measurement of exogenous markers such as inulin, which can prove to be time-consuming and tedious. Furthermore, both tests do not take into account the patient's serum potassium levels, as well as insulin levels.

[0022] Another shortcoming of current methods and techniques for diagnosing, staging, and predicting kidney disease symptoms is that the current clinical standard for potassium measurement (for the purpose of kidney health assessment) is a blood test. In some cases, a whole blood sample is obtained by pricking a finger with a lancet. In some cases, a blood sample is obtained using venous blood sampling. In venous blood sampling, a needle is inserted into a vein to collect a blood sample for testing. Venous blood samples are often ordered weekly for patients. However, measuring whole blood carries the risk of hemolysis (e.g., rupture of red blood cells due to external forces), especially common in the finger prick method of blood collection, which can result in false positive measurements due to the high intracellular concentration of potassium released upon cell rupture. Of all routine blood tests, plasma / serum potassium measurement is one of the most sensitive to the effects of hemolysis, since red blood cell potassium concentration is much higher (about 20 times higher) than that of plasma. Thus, even slight hemolysis can cause erroneously high plasma potassium concentrations, which can make the diagnosis, staging, and prediction of symptoms of kidney disease based on potassium measurements unreliable.

[0023] Furthermore, ensuring that patients continue to participate in such potassium monitoring activities, such as daily or weekly blood testing, can present its own problems. Patients who forgo engaging in such potassium monitoring activities to help stabilize their chronic condition may not be able to manage their condition outside of such testing. If the condition is left unmanaged for too long, the patient's condition may deteriorate significantly, additional health challenges may arise, and in some cases, may lead to an increased risk or likelihood of death.

[0024] Besides the problems mentioned above with conventional disease diagnostic methods and systems used to evaluate and / or monitor renal disease, hyperkalemia, hypokalemia, hyperglycemia, and / or hypoglycemia in patients, such diagnostic methods and systems also fail to consider one or more symptoms in the patient when providing a diagnosis. In particular, while the conventional tests mentioned above may provide diagnostic signs used to aid in the recognition of declining renal health for the purpose of diagnosing renal disease (and / or diagnosing hyperkalemia, hypokalemia, hyperglycemia, hypoglycemia), the identification and prediction of one or more symptoms in a personalized manner, in addition to such diagnostic signs, may provide further insight into the diagnosis, treatment, and management of the disease. In particular, a sign is an objective observable phenomenon that can be identified by another person or device. On the other hand, a symptom is a subjective experience of the patient. Typically, a medical condition such as renal disease (and / or hyperkalemia, hypokalemia, hyperglycemia, hypoglycemia) may be more easily diagnosed with knowledge of both the patient's signs and symptoms.

[0025] Therefore, being able to understand and / or predict a patient's symptoms can play an important role in evaluating renal health, including diagnosing renal disease and other renal health issues.For example, knowledge that a patient is experiencing numbness, tingling, shortness of breath, and / or muscle weakness may suggest that the patient's potassium levels are too high or too low, and in some cases, such high and low potassium levels may be due to the patient's declining renal function.However, when potassium levels are too high or too low, many patients are often asymptomatic or mildly symptomatic until they are immediately at risk or at risk of serious medical complications or death.

[0026] Furthermore, being able to understand and / or predict one or more symptoms of a patient in real time or near real time before the one or more symptoms occur may enable the patient to avoid the one or more symptoms before they occur. For example, if a patient is predicted to experience an arrhythmic event within the next hour, the patient may be able to take one or more actions to prevent the arrhythmia from occurring, or at least seek medical help before the event occurs. However, with conventional methods of measuring potassium, predicting one or more symptoms of a patient in real time before they occur may not be feasible unless the patient has blood drawn constantly (e.g., every few minutes) to allow for continuous assessment of the patient's potassium level over time. Such continuous measurement using conventional methods is generally not feasible or desirable.

[0027] Even if a conventional method is capable of continuously measuring a patient's potassium level, the potassium level used to predict different symptoms of the patient may vary from patient to patient. For example, at a potassium level of 5.8 mmol / L, a first patient may experience severe symptoms, while at the same potassium level, a second patient may have no symptoms. Therefore, in order to accurately predict one or more symptoms of a patient using a conventional method for continuously measuring a patient's potassium level, additional techniques may be required to determine a specific, personalized potassium threshold, which may constantly change for a patient. Determining such a dynamic threshold for predicting different symptoms of a patient may not be possible using conventional methods such as blood sampling.

[0028] Overall, existing diagnostic methods suffer from a first technical problem of being unable to continuously monitor the concentration of a changing analyte, such as potassium, to provide a continuous readout. As used herein, the term "continuous" may mean fully continuous, semi-continuous, periodic, etc. Such continuous monitoring of a particular analyte is advantageous in diagnosing and staging a patient's disease, given that continuous measurements provide continuously up-to-date measurements as well as information regarding trends in the analyte's change over successive periods (e.g., rate of change, trend direction, variability, mean, average, etc.). Such information may be used to make more informed decisions in the assessment of renal health and treatment of renal disease.

[0029] Second, existing diagnostic methods suffer from another technical problem of being unable to continuously monitor the concentrations of multiple varying analytes simultaneously. In particular, continuous, semi-continuous, or periodic monitoring of multiple analytes, such as potassium, glucose, creatinine, lactate, blood urea nitrogen (BUN), ammonia, C-peptide, and / or cystatin C, may provide additional insight when assessing the presence and / or severity of renal disease, hyperkalemia, hypokalemia, hyperglycemia, and / or hypoglycemia in a patient. The additional insight gained from using a combination of analytes, rather than just a single analyte such as potassium, may help to increase the accuracy of predictions and make more informed patient-specific treatment decisions and / or recommendations for preventing disease progression (and in some cases, for disease regression).

[0030] Third, existing diagnostic methods suffer from another technical problem of being unable to predict the symptoms of one or more diseases of a patient, more specifically, being unable to predict the symptoms of one or more diseases of a patient, because the symptoms of one or more diseases of a patient vary from patient to patient.Knowing in advance whether a patient will have a certain symptom can have significant implications in terms of health strategies and interventions.For example, a certain intervention strategy can be recommended to the patient to reduce the likelihood of such symptoms.In addition, predictive models of disease symptoms can represent useful inputs for models designed to assess the presence and / or severity of one or more diseases in a patient.

[0031] As a result of these technical problems, diagnosing renal disease, hyperkalemia, hypokalemia, hyperglycemia, and / or hypoglycemia, or the risk thereof, and identifying and predicting symptoms of such diseases using conventional techniques may prove to be not only inaccurate but also impossible, and in some cases, life-threatening for patients with one or more of these diseases. In particular, predicting the progression of renal health and its corresponding symptoms in a personalized manner with reasonable accuracy may be necessary, considering the dynamic and subclinical nature of renal disease in the early stages, and the heterogeneity of patients. Furthermore, predicting hyperkalemia, hypokalemia, hyperglycemia, and / or hypoglycemia, and their corresponding symptoms in a personalized manner with reasonable accuracy may be necessary, considering that such diseases may contribute to the development of severe or life-threatening conditions (e.g., cardiac arrhythmias, etc.), and in some cases, may lead to death. Therefore, improved methods for detecting and predicting symptoms in patients, diagnosing renal disease, hyperkalemia, hypokalemia, hyperglycemia, and / or hypoglycemia, and understanding the interactions between such conditions in patients are desired.

[0032] Accordingly, certain embodiments described herein provide a technical solution to the above-mentioned technical problem by providing decision support regarding renal health, such as renal disease, using continuous analyte measurements including at least continuous potassium measurements using a continuous analyte monitoring system. In some examples, the continuous potassium measurements are obtained using a continuous potassium monitor (CPM). Furthermore, certain embodiments described herein provide a technical solution to the above-mentioned technical problem by providing decision support regarding renal disease and / or other co-occurring conditions, such as diabetes, using continuous potassium and glucose measurements. For example, a continuous analyte monitoring system including at least a CPM and a continuous glucose monitor (CGM) may be provided to enable providing decision support regarding renal disease and / or other conditions, such as diabetes.

[0033] The decision support may be provided in the form of risk assessment, diagnosis, staging, recommendations for the treatment or prevention of renal disease (or related diseases such as hyperkalemia and hypokalemia), and / or (2) achieving or maintaining glucose homeostasis, as described in more detail herein. As used herein, risk assessment may refer to an assessment or estimation of a patient's renal disease reaching a more advanced stage, hyperkalemia, hypokalemia, hyperglycemia, and / or hypoglycemia, risk of mortality, risk of being diagnosed with one or more other diseases, risk of experiencing one or more symptoms, etc.

[0034] In certain embodiments, the continuous analyte monitoring system may provide decision support to the patient based on various collected data including analyte data, patient information, secondary sensor data (e.g., non-analyte data), etc. For example, the analyte data may include continuously monitored potassium data in addition to other continuously monitored analyte data such as glucose, creatinine, blood urea nitrogen (BUN), ammonia, C-peptide, and cystatin C.

[0035] Potassium helps regulate fluid balance, muscle contraction, and nerve signals in the body. A high potassium diet can also help reduce blood pressure and water retention, protect against stroke, and prevent osteoporosis and kidney stones. As mentioned herein, the kidney plays a major role in potassium homeostasis through the renal mechanisms that transport potassium and regulate its secretion, reabsorption, and excretion. Thus, continuously monitored potassium data can indicate or be used to determine the patient's potassium level and / or the rate of change of the patient's potassium level over time to assess renal health and function for the patient.

[0036] Furthermore, potassium imbalance can be improved by continuous measurement and evaluation of glucose. As mentioned, renal disease can result in glucose imbalance, so continuous monitoring of glucose levels in patients with or at risk of poor and / or declining renal health can be appropriate for reducing the risk of renal disease (and / or related diseases) and for early detection. Additionally, insulin reduces both serum glucose and serum potassium levels by stimulating the uptake of glucose and potassium by cells. Exogenous insulin (e.g., exogenous insulin refers to insulin that people inject or infuse, e.g., via an insulin pump) is often used to manage glucose in diabetic patients. However, insulin can also affect potassium levels, and in patients with CKD, the longer insulin half-life increases the patient's risk of hypokalemia. Thus, continuously monitored glucose data, in addition to continuously monitored potassium data, can be used to better inform the management of a patient's CKD, and more specifically, to provide more informed recommendations for the treatment and / or prevention of CKD, hyperkalemia, hypokalemia, hyperglycemia, and / or hypoglycemia.

[0037] Certain embodiments of the present disclosure provide techniques and systems for correcting a patient's potassium level by using measurements associated with other analyte sensor data, secondary sensor data, and / or other patient information, as described further below. As discussed above, the other analyte sensor data may include glucose, creatinine, lactate, blood urea nitrogen (BUN), ammonia, C-peptide, and / or cystatin C data. The other patient information includes patient information that may include information related to age, gender, family history of kidney disease, other health conditions, and the like. The secondary sensor data may include accelerometer data, heart rate data (ECG, HRV, HR, etc.), temperature, blood pressure, time of sensor initiation or remaining sensor life relative to initiation time, or any other sensor data other than analyte data.

[0038] According to certain embodiments of the present disclosure, the decision support system presented herein is designed to predict the risk or likelihood of one or more symptoms of renal disease occurring for a patient in real time or within a specified time period. In particular, the decision support system presented herein may be designed to predict the risk or likelihood of a patient experiencing (in real time or at a later time) (1) numbness, (2) tingling, (3) shortness of breath (or chest pain), (4) muscle weakness, and / or (4) arrhythmia. Especially in the early stages of renal disease, patients may have few symptoms associated with renal disease. Furthermore, patients may not realize that they have renal disease until the condition has progressed. Thus, being able to identify that a patient is experiencing symptoms of renal disease may be important for early detection of renal disease in patients, even though they may not realize that they are experiencing such symptoms.

[0039] According to embodiments of the present disclosure, the decision support system presented herein is designed to provide disease decision support to assist in diagnosing patients with or at risk for renal disease and in managing renal disease or risk thereof. Providing renal disease decision support may involve using collected data, including, for example, the analyte data, patient information, and secondary sensor data described above, to (1) automatically detect and classify abnormal renal function, (2) assess the presence and severity of renal disease, (3) risk stratify patients to identify patients at high risk for renal disease, (4) identify risks associated with a current renal disease diagnosis (e.g., risk of death, risk of hyperkalemia, risk of hypokalemia, etc.), (5) make patient-specific treatment decisions or recommendations for renal disease (and / or related diseases), and (6) provide information regarding the effect of an intervention (e.g., the effect of a patient's lifestyle changes, the effect of a surgical treatment, the effect of a dialysis session, the effect of a patient taking a new medication, etc.). In other words, the decision support system presented herein may provide information to help direct and improve the care of patients who have or are at risk for renal disease, hyperkalemia, hypokalemia, hyperglycemia, and / or hypoglycemia.

[0040] In certain embodiments, the decision support systems described herein may use various algorithms or artificial intelligence (AI) models, such as machine learning models, trained based on patient-specific and / or population data to provide real-time decision support to patients based on collected information about the patient. For example, certain aspects are directed to algorithms and / or machine learning models designed to assess the presence and severity of renal disease in patients. The algorithms and / or machine learning models may be used in combination with one or more continuous analyte sensors including at least a CPM to provide real-time renal disease assessment and staging. In particular, the algorithms and / or machine learning models may take into account parameters such as potassium levels, trends in potassium levels over time, such as the rate of change of the patient's potassium levels over time, and physiological parameters of the patient when experiencing different symptoms commonly associated with renal disease, when diagnosing and staging renal disease.

[0041] Based on these parameters, the algorithms and / or machine learning models may provide a risk assessment of different renal disease stages (and their corresponding severity) and the patient's progression toward one or more of these renal disease stages. The algorithms and / or machine learning models may take into account population data, personalized patient-specific data, or a combination of both when diagnosing and staging a patient's renal disease.

[0042] As described in more detail herein, the population data may be provided in the form of a data set including data records of historical patients having various stages of renal disease and / or experiencing potassium imbalance, and in some cases, different symptoms associated with renal disease. Each data record may be used as an input to a machine learning model to optimize such a model to generate accurate predictions associated with renal disease (e.g., a prediction of the presence and severity of renal disease in a patient, a prediction of one or more symptoms of renal disease occurring in real time or within a specified time period, etc.). According to certain embodiments, prior to deployment, the machine learning model is trained with training data including, for example, user-specific data and / or population data.

[0043] According to certain embodiments, the decision support systems described herein are designed to provide decision support in the form of risk assessment and treatment for renal disease and / or glucose homeostasis. For example, in certain embodiments, the decision support system is designed to continuously measure a patient's serum potassium levels and provide recommendations for treatment, specifically potassium imbalances associated with renal disease. In certain other embodiments, the decision support system is designed to continuously measure both a patient's serum potassium and glucose levels (e.g., via multiple sensors or a sensor capable of measuring both analytes) and provide recommendations for treatment of potassium and glucose imbalances.

[0044] In certain embodiments, the decision support system described herein provides risk stratification and treatment recommendations through algorithms and / or methods for the treatment and / or prevention of renal disease, hyperkalemia, hypokalemia, hyperglycemia, and / or hypoglycemia. In certain embodiments, the method is a tree-like rule-based model for providing decision support to the patient. Other types of algorithms and rule-based models are also within the scope of the present disclosure. In certain embodiments, the methods described herein may be used to make effective decisions regarding treatment and / or preventive medical recommendations. In particular, the methods described herein include different nodes (or steps) used in determining treatment recommendations, and in some cases, include risk stratification to inform the treatment recommendations. For example, in making medical decisions, the methods described herein may be used to aid in making diet-related decisions and decisions about whether and / or when to consume potassium and / or glucose and / or water for the purpose of maintaining potassium and / or glucose homeostasis. As another example, when making medical decisions, the methods described herein can be used to aid in decisions regarding insulin administration, dialysis, potassium binder intake, and to identify when a patient may need to consult with a medical professional and / or go to the emergency room for diagnosis and / or treatment.

[0045] In certain embodiments, decision support risk stratification and treatment recommendations may be based on the patient's potassium levels, trends, and / or thresholds. Different thresholds may be set based on the risk of potassium-related conditions such as hyperkalemia and / or hypokalemia, available treatments, efficacy of treatments, patient characteristics, patient activity, etc. Additionally, thresholds may be set by a user or medical professional, may be fixed and therefore may not be set by the user, may default to a particular value, or may be dynamically set and adjusted automatically or manually based on input 128. Thresholds may be absolute values ​​or may correspond to or be based on the user's trends.

[0046] Further, in certain embodiments, decision support risk stratification and treatment recommendations may be based on the patient's glucose levels, rates of change, trends, and / or thresholds. Different glucose thresholds may be set based on risk of glucose-related conditions such as hyperglycemia and / or hypoglycemia, and risk of hyperkalemia and / or hypokalemia, available treatments, efficacy of treatments, patient characteristics, patient activity, etc.

[0047] The combination of a continuous analyte monitoring system with machine learning models and / or algorithms provided by the decision support system described herein to (1) predict one or more symptoms associated with potassium imbalance (e.g., abnormal or out-of-range potassium levels) and (2) predict (e.g., assess) a user's renal health, including diagnosing, staging, and risk assessment of renal disease, allows for real-time diagnosis to enable early intervention. In particular, the decision support system may be used to provide early alerts of declining renal function and / or deliver information about other complications associated with the kidney. Early detection of such decompensation and / or other complications may enable intervention at the earliest possible stage to improve renal health and / or renal disease outcomes. For example, early intervention may reduce hospitalizations, complications, and mortality in some cases. Additionally, potassium levels and potassium level changes provided by the continuous analyte monitoring system may be used as inputs to machine learning models and / or algorithms to triage patients for more urgent care. In patients at risk for cardiac arrhythmia episodes and / or death, a sudden increase in potassium can be used to signal urgent medical intervention, even before the patient can report noticeable physiological symptoms.

[0048] In addition, through the combination of a continuous analyte monitoring system with machine learning models and / or algorithms (e.g., rules-based models) for (1) predicting one or more symptoms associated with potassium imbalance, and (2) predicting a user's renal health, including diagnosing, staging, and risk assessment of renal disease, the decision support system described herein may provide the necessary accuracy and reliability that patients expect. For example, bias, human error, and emotional influences may be minimized when assessing the presence and severity of renal health in a patient. Furthermore, the machine learning models and algorithms (e.g., rules-based models) combined with the analyte monitoring system may provide insight into patterns and / or trends of declining patient health, at least with respect to the kidneys and / or heart, that may have been previously overlooked. Thus, the decision support system described herein may aid in the identification of renal health for diagnostic, preventative, and therapeutic purposes.

[0049] Exemplary Decision Support System Including Exemplary Analyte Sensor FIG. 1 illustrates an exemplary decision support system 100 (also referred to herein as a “monitoring system”) for using a continuous analyte monitoring system 104 including at least a continuous potassium monitor (CPM) to (1) predict one or more symptoms associated with potassium imbalance and / or (2) predict a user's renal health, including diagnosing, staging, treating, and risk assessment of renal disease for a user 102 (individually referred to herein as a user and collectively referred to herein as a user). The user may be a patient, in certain embodiments, or in some cases, a caregiver for the patient. In certain embodiments, the decision support system 100 includes the continuous analyte monitoring system 104, a display device 107 running an application 106, a decision support engine 114, a user database 110, a medical history record database 112, a training server system 140, and the decision support engine 114, each of which is described in more detail below.

[0050] As used herein, the term "analyte" is a broad term used in its ordinary sense, including, but not limited to, referring to a substance or chemical constituent in a biological fluid (e.g., blood, interstitial fluid, cerebrospinal fluid, lymphatic fluid, or urine) that may be analyzed. Analytes may include naturally occurring substances, man-made substances, metabolites, and / or reaction products. Analytes measured by the present device and method may include, but are not limited to, potassium, glucose, acarboxyprothrombin; acylcarnitines; adenine phosphoribosyltransferase; adenosine deaminase; albumin; alpha-fetoprotein; amino acid profile (arginine (Krebs cycle), histidine / urocanic acid, homocysteine, phenylalanine / tyrosine, tryptophan); androstenedione; antipyrine; arabinitol enantiomers; arginase; benzoylecgonine (cocaine); biotinidase; biopterin; c-reactive protein; carnitine; carnosinase; CD4; ceruloplasmin; chenodeoxycholic acid; chloroquine; cholesterol; cholinesterase; conjugated 1-beta hydroxy-cholic acid; cortisol; creatine kinase; creatinine Enzyme MM;cyclosporine A;d-penicillamine;deethylchloroquine;dehydroepiandrosterone sulfate;DNA (acetylation polymorphism, alcohol dehydrogenase, alpha 1-antitrypsin, glucose-6-phosphate dehydrogenase, hemoglobin A, hemoglobin S, hemoglobin C, hemoglobin D, hemoglobin E, hemoglobin F, D-Punjab, beta-thalassemia, hepatitis B virus, HCMV, HIV-1, HTLV-1, MCAD, RNA, PKU, Plasmodium vivax, 21-deoxycortisol;desbutylhalofantrine;dihydropteridine reductase;diphtheria / tetanus antitoxin;endogenous insulin;erythrocyte arginase;erythrocyte protoporphyrin;esterase D;fatty acid / acylglycine;free beta-human chorionic gonadotropin;free erythrocyte porphyrin;free thyroxine (free thyroxine, FT4); free tri-iodothyronine, FT3; fumarylacetoacetase;Galactose / gal-1-phosphate;galactose-1-phosphate uridyltransferase;gentamicin;glucose-6-phosphate dehydrogenase;glutathione;glutathione peroxidase;glycocholate;glycosylated hemoglobin;halofantrine;hemoglobin variants;hexosaminidase A;human erythrocyte carbonic anhydrase I;17-α-hydroxyprogesterone;hypoxanthine phosphoribosyltransferase;immunoreactive trypsin;lactate;lead;lipoproteins ((a), B / A-1, β);lysozyme;mefloquine;netilmicin;phenobarbitone;phenytoin;phytanic acid / pristanic acid;progesterone;prolactin;prolidase;purine nucleoside phosphorylase;quinine;reverse triiodothyronine (reverse tri-iodothyronine, rT3); selenium; serum pancreatic lipase; sisomicin; somatomedin C; specific antibodies recognizing any one or more of the following, which may include adenoviruses, antinuclear antibodies, anti-zeta antibodies, arboviruses, pseudorabies virus, dengue virus, dracunculiasis, tapeworm, Entamoeba histolytica, enteroviruses, giardiasis, Helicobacter pylori, hepatitis B virus, herpes virus, HIV-1, IgE (atopic disease), influenza virus, Leishmania donovani, Leptospirosis, measles / mumps / rubella, Mycobacterium leprae, Mycoplasma pneumoniae, myoglobin, convolutus helminthiasis, parainfluenza virus, malaria parasite, poliovirus, Pseudomonas aeruginosa, respiratory syncytial virus, Rickettsia (tsutsugamushi disease), Schistosoma mansoni, Toxoplasma gondii, Treponema pallidum, Trypanosoma cruzi / langeri, Vesicular stomatitis virus, Wuchereria bancrofti, Yellow fever virus);Specific antigens (Hepatitis B virus, HIV-1);Succinylacetone;Sulfadoxine;Theophylline;Thyrotropin (TSH);Thyroxine (T4);Thyroxine-binding globulin;Trace elements;Transferrin;UDP-galactose-4-epimerase;Urea;Uroporphyrinogen I synthase;Vitamin A;Leukocytes;and zinc protoporphyrin. Salts, sugars, proteins, fats, vitamins, and hormones naturally present in blood or interstitial fluids can also constitute analytes in certain implementations. Ions are charged atoms or compounds that can include the following (e.g., sodium, potassium, calcium, chloride, nitrogen, or bicarbonate). Analytes can be naturally present in biological fluids, such as metabolites, hormones, antigens, antibodies, ions, etc. Alternatively, the analyte may be introduced into the body or may be exogenous, such as contrast agents for imaging, radioisotopes, chemical agents, fluorocarbon-based synthetic blood, challenge agent analytes (e.g., introduced for the purpose of measuring the increase and / or decrease in the rate of change of the concentration of a challenge agent analyte or other analytes in response to the introduced challenge agent analyte), or exogenous insulin; glucagon, ethanol; cannabis (marijuana, tetrahydrocannabinol, hashish); inhalants (nitrous oxide, amyl nitrite, butyl nitrite, chlorohydrocarbons, hydrocarbons); cocaine (crack cocaine); stimulants (amphetamine, methamphetamine, Ritalin, Silurt, Propylene glycol, selenium ... Relzine, Didrex, Prestate, Boranil, Sandrex, Pregin;Antidepressants (barbiturates, tranquilizers such as methaqualone, Valium, Librium, Miltown, Serax, Equuanil, Tranxine);Hallucinogens (phencyclidine, lysergic acid, mescaline, peyote, psilocybin);Nucleic acids (heroin, codeine, morphine, opium, meperidine, Percocet, Percodan, Tasionex, fentanyl, Darvon, Talwin, Lomotil);Synthetic narcotics (fentanyl, meperidine, amphetamine, methamphetamine, and phencyclidine analogues, e.g., ecstasy);Anabolic steroids;The analytes may be drugs or pharmaceutical compositions, including, but not limited to, nicotine and nicotine. Metabolites of drugs and pharmaceutical compositions are also contemplated analytes. Analytes such as neurochemicals and other chemicals produced in the body, such as ascorbic acid, uric acid, dopamine, noradrenaline, 3-methoxytyramine (3MT), 3,4-dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA), 5-hydroxytryptamine (5HT), and 5-hydroxyindoleacetic acid (FHIAA), as well as intermediates of the citric acid cycle, can also be analyzed;

[0051] Analytes measured and analyzed by the devices and methods described herein include potassium, and in some cases glucose, creatinine, blood urea nitrogen (BUN), ammonia, C-peptide, and cystatin C, although other analytes listed above may also be considered and measured, for example, by the analyte monitoring system 104, but are not limited to these.

[0052] In certain embodiments, the continuous analyte monitoring system 104 is configured to continuously measure one or more analytes and transmit the analyte measurements to an electronic medical records (EMR) system (not shown in FIG. 1 ). An EMR system is a software platform that enables electronic entry, storage, and maintenance of digital medical data. EMR systems are generally used throughout hospitals and / or other care facilities to document clinical information about patients over time. EMR systems organize and present data in a manner that aids clinicians, for example, in interpreting health conditions and providing ongoing care, scheduling, billing, and follow-up. Data contained in the EMR system may also be used to generate reports for the patient's clinical care and / or disease management. In certain embodiments, the EMR may communicate (e.g., over a network) with a decision support engine 114 to execute the techniques described herein. In particular, as described herein, the decision support engine 114 may obtain data associated with a user and use the obtained data as input to one or more trained models and output predictions. In some cases, the EMR may provide data to be used as input to one or more models to the decision support engine 114. Additionally, in some cases, the decision support engine 114, after making a prediction, may provide its output prediction to the EMR.

[0053] In certain embodiments, the continuous analyte monitoring system 104 is configured to continuously measure one or more analytes and transmit the analyte measurements to the display device 107 for use by the application 106. In some embodiments, the continuous analyte monitoring system 104 transmits the analyte measurements to the display device 107 through a wireless connection (e.g., a Bluetooth connection). In certain embodiments, the display device 107 is a smartphone. However, in certain other embodiments, the display device 107 may instead be any other type of computing device, such as a laptop computer, a smart watch, a tablet, or any other computing device capable of running the application 106. In some embodiments, the continuous analyte monitoring system 104 and / or the analyte sensor application 106 transmit the analyte measurements to one or more other individuals interested in the patient's health (e.g., a family member or a physician for real-time treatment and care of the patient). The continuous analyte monitoring system 104 may be described in more detail with respect to FIG. 2.

[0054] Application 106 is a mobile health application configured to receive and analyze analyte measurements from analyte monitoring system 104. In particular, application 106 stores information about the user in the user profile 118, including the user's analyte measurements, for processing and analysis, and for use by decision support engine 114 to provide decision support recommendations or guidance to the user.

[0055] The decision support engine 114 refers to a set of software instructions having one or more software modules, including a data analysis module (DAM) 116. In certain embodiments, the decision support engine 114 executes entirely on one or more computing devices in a private or public cloud. In such embodiments, the application 106 communicates with the decision support engine 114 via a network (e.g., the Internet). In some other embodiments, the decision support engine 114 executes partially on one or more local devices, such as the display device 107, and partially on one or more computing devices in a private or public cloud. In some other embodiments, the decision support engine 114 executes entirely on one or more local devices, such as the display device 107. As discussed in more detail herein, the decision support engine 114 may provide decision support recommendations to the user via the application 106. The decision support engine 114 provides decision support recommendations based on information contained in the user profile 118.

[0056] The user profile 118 may include information collected about the user from the application 106. For example, the application 106 provides a set of inputs 128 that include analyte measurements received from the continuous analyte monitoring system 104 that are stored in the user profile 118. In certain embodiments, the inputs 128 provided by the application 106 include other data in addition to the analyte measurements received from the continuous analyte monitoring system 104. For example, the application 106 may obtain the additional inputs 128 via manual user input, one or more other non-analyte sensors or devices, non-continuous analyte laboratory test results (e.g., a urine albumin test, GFR, or ACR), other applications running on the display device 107, etc. Non-analyte sensors and devices include, but are not limited to, one or more of an insulin pump, an electrocardiogram (ECG) sensor or heart rate monitor, a blood pressure sensor, a respiration sensor, a thermometer, a peritoneal dialysis machine, a hemodialysis machine, a sensor or device provided by the display device 107 (e.g., an accelerometer, a camera, a global positioning system (GPS), a heart rate monitor, etc.), or other user accessories (e.g., a smart watch), or any other sensor or device that provides relevant information about the user. The input 128 of the user profile 118 provided by the application 106 is described in further detail below with respect to FIG.

[0057] The DAM 116 of the decision support engine 114 is configured to process the set of inputs 128 to determine one or more metrics 130. The metrics 130, which will be discussed in more detail below with respect to FIG. 3, may, at least in some cases, generally be indicative of the user's health or condition, such as one or more of the user's physiological condition, trends associated with the user's health or condition, etc. In certain embodiments, the metrics 130 may then be used by the decision support engine 114 as inputs to provide guidance to the user. As shown, the metrics 130 are also stored in the user profile 118.

[0058] The user profile 118 also includes demographic information 120, disease progression information 122, and / or medical information 124. In certain embodiments, such information may be provided via user input or may be obtained from a particular data store (e.g., electronic medical record (EMR), etc.). In certain embodiments, the demographic information 120 may include one or more of the user's age, body mass index (BMI), ethnicity, gender, etc. In certain embodiments, the disease progression information 122 may include information regarding the user's disease, such as whether the user has previously been diagnosed with acute kidney injury (AKI) or chronic kidney disease (CKD), or has had a history of hyperkalemia, hypokalemia, etc. In certain embodiments, the information regarding the user's disease may also include the length of time since diagnosis, the level of disease control, the level of compliance with disease management therapy, predicted kidney function, other types of diagnoses (e.g., diabetes, heart disease, obesity), or health measurements (e.g., heart rate, exercise, stress, sleep, etc.), etc.

[0059] In certain embodiments, the medical information 124 may include information regarding the amount, frequency, and type of medication taken by the user. In certain embodiments, the amount, frequency, and type of medication taken by the user are time-stamped and correlated with the user's analyte levels, thereby indicating the effect the amount, frequency, and type of medication had on the user's analyte levels. In certain embodiments, the medical information may include information regarding the consumption of one or more diuretics. Diuretics may be prescribed to patients to treat excess fluid accumulation caused by, for example, congestive heart failure (CHF), liver failure, and / or nephritic syndrome. For example, CHF is a condition in which the heart cannot pump blood efficiently to meet the body's oxygen and nutrient needs. When the heart cannot pump blood efficiently, normal blood circulation is impaired, leading to excess fluid in the blood. The excess fluid leaks out of the blood vessels and accumulates in the lungs and other tissues. Therefore, patients may be prescribed diuretics to help the kidneys flush out excess fluid and maintain normal blood volume.

[0060] Different types of diuretics prescribed to patients may include loop diuretics, thiazide and thiazide-like diuretics, and potassium-sparing diuretics. Loop diuretics inhibit a protein found in a part of the nephron known as the loop of Henle. Loop diuretics may include, for example, furosemide (Lasix), bumetanide (Bumex), torsemide (Demadex), ethacrynic acid (Edecrin). Thiazide diuretics are commonly used to treat high blood pressure (hypertension), but are also used to manage heart failure. Thiazide diuretics inhibit a different protein than loop diuretics inhibit, which also aids in mineral reabsorption. Thiazide diuretics may include, for example, chlorothiazide (Diuril), hydrochlorothiazide (Hydrodiuril), metolazone (Zytonix). Finally, potassium-sparing diuretics (such as, for example, spironolactone) are weak diuretics used to increase the amount of fluid passing from the body in the urine while also preventing excess potassium from being lost from the body in the urine. As described in more detail below, the decision support system 100 may be configured to use the medical information 124 to determine optimal diuretics to be prescribed for different users. In particular, the decision support system 100 may be configured to identify one or more optimal diuretics for prescription based on the health of the patient at the time the one or more diuretics are prescribed, as well as the patient's condition being treated.

[0061] In certain embodiments, the pharmaceutical information may include information regarding the consumption of one or more drugs known to damage the kidney. The one or more drugs known to damage the kidney may include nonsteroidal anti-inflammatory drugs (NSAIDS), such as ibuprofen (e.g., Advil, Motrin) and naproxen (e.g., Aleve), vancomycin, iodized radiocontrast agents (e.g., refers to any contrast agent used in diagnostic tests), angiotensin-converting enzyme (ACE) agents, such as lisinopril, enalapril, and ramipril, aminoglycoside antibiotics, such as neomycdin, gentamicin, tobramycin, and amikacin, antiviral human immunodeficiency virus (HIV) medications, zoledronic acid (e.g., Zometa, Reclast), foscarnet, and the like.

[0062] In certain embodiments, the medical information may include information regarding the consumption of one or more drugs known to control complications of kidney disease. The one or more drugs known to control complications of kidney disease may include drugs for lowering blood pressure and maintaining kidney function, such as ACE inhibitors or angiotensin II receptor blockers, drugs for treating anemia, such as supplements of the hormone erythropoietin, drugs used to lower cholesterol levels, such as statins, drugs used to prevent weak bones, such as calcium and vitamin D supplements, phosphate binders, and the like.

[0063] In certain embodiments, the user profile 118 is dynamic because at least a portion of the information stored in the user profile 118 may be modified over time and / or new information may be added to the user profile 118 by the decision support engine 114 and / or the application 106. Thus, the information in the user profile 118 stored in the user database 110 provides an up-to-date repository of information related to the user.

[0064] User database 110, in some embodiments, refers to a storage server operating in a public or private cloud. User database 110 may be implemented as any type of data store, such as a relational database, a non-relational database, a key-value data store, a file system including a hierarchical file system, etc. In some example implementations, user database 110 is distributed. For example, user database 110 may comprise multiple persistent storage devices that are distributed. Additionally, user database 110 may be replicated such that the storage devices are geographically distributed.

[0065] The user database 110 includes user profiles 118 associated with multiple users who similarly interact with the applications 106 executing on the display devices 107 of other users. The user profiles stored in the user database 110 are accessible to the applications 106 as well as the decision support engine 114. The user profiles in the user database 110 may be accessible to the applications 106 and the decision support engine 114 via one or more networks (not shown). As described above, the decision support engine 114, and more specifically the DAM 116 of the decision support engine 114, can fetch inputs 128 from the user database 110 and then calculate multiple metrics 130, which may be stored as application data 126 in the user profiles 118.

[0066] In certain embodiments, the user profiles 118 stored in the user database 110 may also be stored in the medical history record database 112. The user profiles 118 stored in the medical history record database 112 may provide a repository of up-to-date information and medical history information for each user of the application 106. Thus, the medical history record database 112 essentially provides all data related to each user of the application 106, with the data stored according to an associated timestamp. The timestamp associated with the information stored in the medical history record database 112 may identify, for example, when information related to a user was obtained and / or updated.

[0067] Additionally, the medical history database 112 may maintain time series data collected for users over a period of time, including for users who use the continuous analyte monitoring system 104 and application 106. For example, analyte data for a user who used the continuous analyte monitoring system 104 and application 106 for five years to manage the user's renal condition may have the time series analyte data associated with the user maintained for the five years.

[0068] Additionally, in certain embodiments, historical record database 112 may include data for one or more patients who are not users of continuous analyte monitoring system 104 and / or application 106. For example, historical record database 112 may include information (e.g., user profiles) about one or more patients who have not been previously diagnosed with renal disease, as analyzed, for example, by a health care physician (or other known method), as well as information (e.g., user profiles) about one or more patients who have been previously diagnosed with renal disease (of various types and stages), as analyzed, for example, by a health care physician (or other known method). The data stored in historical record database 112 may be referred to herein as population data.

[0069] The data associated with each patient stored in the medical history database 112 may provide time series data collected over the patient's disease lifetime. For example, the data may include information about the patient before he or she was diagnosed with kidney disease, and information associated with the patient during the life of the disease, including information associated with each stage of kidney disease that has progressed and / or regressed in the patient, as well as information associated with other diseases, such as hyperkalemia, hypokalemia, diabetes, cardiac conditions and diseases, or similar diseases that coexist in association with kidney disease. Such information may be indicative of the patient's symptoms, the patient's physiological condition, the patient's potassium level, the patient's glucose level, the patient's creatinine level, the patient's BUN level, the patient's ammonia level, the patient's C-peptide level, the patient's cystatin C level, the status / condition of one or more organs of the patient, the patient's habits (e.g., activity level, food consumption, etc.), prescribed medications, etc.

[0070] Although depicted as separate databases for conceptual clarity, in some embodiments, the user database 110 and the medical history record database 112 may operate as a single database. That is, historical and current data associated with users of the continuous analyte monitoring system 104 and application 106, as well as historical data associated with patients who were not previous users of the continuous analyte monitoring system 104 and application 106, may be stored in a single database. The single database may be a storage server running on a public or private cloud.

[0071] As previously mentioned, the decision support system 100 is configured to predict a user's renal health, including renal disease diagnosis, staging, treatment, and risk assessment, for a user using a continuous analyte monitoring system 104 including at least a CPM, as well as predict the likelihood that the user will experience one or more symptoms commonly associated with potassium imbalance. In certain embodiments, the decision support engine 114 is configured to provide real-time and / or non-real-time decision support regarding renal health to the user and / or others, including, but not limited to, health care providers, the user's family, the user's caregivers, researchers, artificial intelligence (AI) engines, and / or other individuals, systems, and / or groups that assist with care or learning from data. In particular, the decision support engine 114 may be used to collect information associated with the user in a user profile 118 stored in the user database 110, determine the likelihood that the user will experience one or more renal disease symptoms, and in some cases, perform analysis on the information to provide a recommendation or warning to the user based on the determined likelihood. The decision support engine 114 may also be used to perform an analysis to collect information associated with the user in the user profile 118, determine a probability of the presence and / or severity of renal disease for the user, and provide one or more recommendations for treatment based, at least in part, on the determination. The user profile 118 may be accessible to the decision support engine 114 via one or more networks (not shown) to perform such an analysis.

[0072] In certain embodiments, the decision support system 100 is designed to predict the risk or likelihood of one or more symptoms of renal disease occurring for a patient in real time (including near real time) or within a specified time period. In certain embodiments, to enable such prediction, the decision support engine 114 is configured to collect information associated with the user in a user profile 118 stored in the user database 110 and perform an analysis to predict the risk or likelihood (in real time or at a later time) that the patient will experience (1) potassium fluctuations (indicative of hyperkalemia or hypokalemia), (2) numbness, (3) tingling, (4) shortness of breath (or chest pain), (5) muscle weakness, (6) arrhythmia, or other symptoms. In certain embodiments, the decision support engine 114 may attempt to verify whether the user is experiencing one or more predicted symptoms and store physiological parameters (e.g., potassium levels, glucose levels, C-peptide levels, heart rate, blood pressure, etc.) associated with the user, and / or one or more other diagnostic tests to verify that the user is in fact experiencing one or more predicted symptoms. Recording such physiological parameters may provide insight into which potassium thresholds (in addition to other measurements) correlate with which symptoms (e.g., common symptoms of renal disease) for each particular user. Further, in certain embodiments, each user's recorded potassium thresholds over time for different symptoms may be analyzed to provide an indication of improvement or worsening of the patient's renal disease.

[0073] In certain embodiments, the decision support engine 114 may utilize one or more trained machine learning models and / or algorithms (e.g., rule-based models) capable of predicting a user's symptoms based on information provided by the user profile 118. In the illustrated embodiment of FIG. 1, the decision support engine 114 may utilize trained machine learning models provided by a training server system 140. Although depicted as separate servers for conceptual clarity, in some embodiments, the training server system 140 and the decision support engine 114 may operate as a single server. That is, models may be trained and used by a single server, or trained by one or more servers and deployed for use on one or more other servers. In certain embodiments, models may be trained on one or many virtual machines (VMs) running at least in part on one or many physical servers in relational and / or non-relational database formats.

[0074] The training server system 140 is configured to train the machine learning models using training data, which may include data (e.g., from user profiles) associated with one or more patients (e.g., users or non-users of the continuous analyte monitoring system 104 and / or application 106) previously diagnosed with various stages of renal disease, as well as patients (e.g., healthy patients) not previously diagnosed with renal disease. The training data may be stored in the medical history record database 112 and may be accessible to the training server system 140 via one or more networks (not shown) to train the machine learning models. The training data may also include user-specific data, in some cases, for users over time.

[0075] Training data refers to a characterized and labeled data set. For example, the data set may include multiple data records, each containing information corresponding to a different user profile stored in the user database 110, and each data record is characterized and labeled. In machine learning and pattern recognition, a feature is an individual measurable characteristic or feature. Generally, the feature that best characterizes the patterns in the data is selected to create a predictive machine learning model. Data labeling is the process of adding one or more meaningful and informative labels to provide context to the data for learning by the machine learning model.

[0076] As an illustrative example, each relevant feature of a user reflected in a corresponding data record may be a feature used in training a machine learning model. Such features may include age, sex, potassium level change (e.g., delta) from a first timestamp to a second timestamp, change in renal disease stage or severity (e.g., delta) from a first timestamp to a second timestamp, change in potassium thresholds for users experiencing different symptoms (e.g., delta) from a first timestamp to a subsequent timestamp, derivatives of measured linear systems of potassium measurements at points at particular timestamps, and / or differences in derivatives to determine the rate of change of slope of increase or decrease in potassium values, etc. Additionally, the data records are labeled with an indication of experienced symptoms, renal disease diagnosis, assigned disease severity, and / or identified risk of renal disease, etc., associated with the patient in the user profile.

[0077] The model is then trained by the training server system 140 using the characterized and labeled training data. In particular, the features of each data record may be used as input to a machine learning model, and the generated output may be compared to the label associated with the corresponding data record. The model may calculate a loss based on the difference between the generated output and the provided label. This loss is then used to modify the internal parameters or weights of the model. By iteratively processing each data record corresponding to each historical patient, in certain embodiments, the model may be iteratively refined to generate accurate predictions associated with the user's renal health, including renal disease risk, presence, progression, improvement, and severity in the patient. Furthermore, in certain other embodiments, by iteratively processing each data record corresponding to each historical patient, in certain embodiments, the model may be iteratively refined to generate accurate predictions of the risk and / or presence of one or more symptoms associated with potassium imbalance.

[0078] As illustrated in FIG. 1, the training server system 140 deploys these trained models to the decision support engine 114 for use during runtime. For example, the decision support engine 114 may obtain a user profile 118 associated with the user, use information in the user profile 118 as input to the trained models, and output a prediction. The prediction may indicate the user's renal health, including the presence and / or severity of the user's renal disease, or may indicate the user's presence or risk of experiencing one or more symptoms of renal disease in real time or within a certain time period (e.g., shown as output 144 in FIG. 1). The output 144 generated by the decision support engine 114 may also provide one or more recommendations for treatment based on the prediction. The output 144 may be provided to the user (e.g., through the application 106), to the user's caregiver (e.g., parent, relative, guardian, teacher, nurse, etc.), to the user's physician, or in some cases, to any other individual interested in the user's health, such as by achieving the recommended treatment.

[0079] In certain embodiments, the user's own data is used to personalize one or more models that were initially trained based on population data. For example, a model (e.g., trained using population data) may be deployed for use by the decision support engine 114 to predict the presence or risk of a particular user experiencing symptoms of renal disease in real time. After making a prediction using the model, the decision support engine 114 may be configured to ask the user if the user is experiencing the predicted symptoms and / or to use one or more diagnostic tests to confirm that the user is experiencing the predicted symptoms. In some cases, the user's answers and / or results from the diagnostic tests performed may negate that the user is experiencing the predicted symptoms. Thus, the model may continue to be retrained and personalized using the user's answers, diagnostic test results, and / or the user's physiological parameters used as inputs to the model to personalize the model for the user.

[0080] In certain embodiments, the output 144 generated by the decision support engine 114 may be stored in the user profile 118. In certain embodiments, the output 144 may be a predicted likelihood that the user will currently experience one or more renal disease symptoms, or a predicted likelihood that the user will experience one or more renal disease symptoms within a specified time period. In certain embodiments, the output 144 may be a patient-specific treatment decision or recommendation to prevent one or more predicted renal disease predictions from occurring in the patient. In certain embodiments, the output 144 may be a prediction regarding the user's renal health, including the presence and / or severity of renal disease in the user. In certain embodiments, the output 144 may be a prediction regarding the user's risk of having renal disease, hyperkalemia, and / or hypokalemia. In certain embodiments, the output 144 may be a prediction regarding the patient's mortality risk. In certain embodiments, the output 144 may be a patient-specific treatment decision or recommendation for the patient's renal disease. The output 144 stored in the user profile 118 may be continuously updated by the decision support engine 114. Thus, the user's previous diagnostic and / or physiological parameters corresponding to different symptoms commonly associated with renal disease, initially stored as output 144 in the user profile 118 in the user database 110 and then passed to the medical history record database 112, may provide an indication of the progression of renal disease in the user over time, as well as an indication as to the effectiveness of different treatments (e.g., medications) recommended to the user to help halt the progression of the disease.

[0081] In certain embodiments, the user's own historical data may be used to provide decision support and insight regarding the user's renal health, such as renal function and / or disease. For example, the user's historical data may be used by an algorithm as a baseline to indicate improvement or deterioration of the user's renal function. As an illustrative example, the user's data from a certain period of time, e.g., one month ago, two weeks ago, etc., may be used as a baseline that can be compared to the user's current data to identify whether the user's renal function has improved or deteriorated. In certain embodiments, the user's own historical data may be used by the training server system 140 to train a personalized model that may be able to further predict or estimate the user's renal function or future improvement / deterioration based on the user's recent data patterns (e.g., exercise data, food consumption data, etc.).

[0082] In certain embodiments, machine learning models and / or algorithms (e.g., rule-based models) may be trained or configured to provide lifestyle recommendations, exercise recommendations, food intake recommendations, hydration recommendations, medication recommendations, and other types of decision support recommendations to help the user prevent conditions, treat conditions, and improve kidney health and function based on the user's historical data, including how different types of medications, foods, and treatments (e.g., dialysis, etc.) have affected the user's renal function in the past. In certain embodiments, the models may be trained to predict the underlying causes of a particular improvement or deterioration in a patient's renal function. For example, the application 106 may display a user interface with a graph showing the patient's renal function or its score along with a trend line, e.g., retrospectively, what caused the function of the kidneys at a particular time point (e.g., excess potassium intake, taking a non-potassium sparing diuretic, etc.) or what caused a particular symptom.

[0083] 2 is a diagram 200 that conceptually illustrates an example continuous analyte monitoring system 104 including an example continuous analyte sensor with sensor electronics, according to certain aspects of the present disclosure. For example, the system 104 can be configured to continuously monitor one or more analytes of a user, according to certain aspects of the present disclosure.

[0084] The continuous analyte monitoring system 104 in the illustrated embodiment includes a sensor electronics module 204 and one or more continuous analyte sensors 202 (individually referred to herein as a continuous analyte sensor 202 and collectively referred to herein as continuous analyte sensors 202) associated with the sensor electronics module 204. The sensor electronics module 204 may wirelessly communicate (e.g., directly or indirectly) with one or more of the display devices 210, 220, 230, and 240. In certain embodiments, the sensor electronics module 204 may also wirelessly communicate (e.g., directly or indirectly) with one or more medical devices, such as a medical device 208 (individually referred to herein as a medical device 208 and collectively referred to herein as medical devices 208), and / or one or more other non-analyte sensors 206 (individually referred to herein as a non-analyte sensor 206 and collectively referred to herein as non-analyte sensors 206).

[0085] In certain embodiments, the continuous analyte sensor 202 may include a sensor for detecting and / or measuring an analyte. The continuous analyte sensor 202 may be a multi-analyte sensor configured to continuously measure two or more analytes, or a single analyte sensor (e.g., a potassium sensor) configured to continuously measure a single analyte as a non-invasive, subcutaneous, transdermal, transdermal, and / or intravascular device. In certain embodiments, the continuous analyte sensor 202 may be configured to continuously measure an analyte level of a user using one or more measurement techniques, such as enzymatic, chemical, physical, electrochemical, spectrophotometric, polarimetric, calorimetric, iontophoretic, radiometric, immunochemical, etc. In certain aspects, the continuous analyte sensor 202 provides a data stream indicative of the concentration of one or more analytes in the user. The data stream may include raw data signals that are then converted to a calibrated and / or filtered data stream used to provide an estimated analyte value to the user.

[0086] In certain embodiments, the continuous analyte sensor 202 may be a multi-analyte sensor configured to continuously measure multiple analytes in a user's body. For example, in certain embodiments, the continuous multi-analyte sensor 202 may be a single sensor configured to measure potassium, glucose, lactate, ketones, creatinine, blood urea nitrogen (BUN), ammonia, C-peptide, and cystatin C in a user's body.

[0087] In certain embodiments, one or more multi-analyte sensors may be used in combination with one or more single-analyte sensors. As an illustrative example, a multi-analyte sensor may be configured to continuously measure potassium and glucose, and in some cases may be used in combination with an analyte sensor configured to measure only BUN levels. Information from each of the multi-analyte and single-analyte sensors may be combined to provide renal disease decision support using the methods described herein.

[0088] In certain embodiments, the sensor electronics module 204 includes electronic circuitry associated with measuring and processing the continuous analyte sensor data, including predictive algorithms associated with processing and calibrating the sensor data. The sensor electronics module 204 can be physically connected to the continuous analyte sensor 202 and can be integral (non-releasably attached) or releasably attached to the continuous analyte sensor 202. The sensor electronics module 204 can include hardware, firmware, and / or software that enables measurement of levels of a display device via the continuous analyte sensor 202. For example, the sensor electronics module 204 may include a potentiostat, a power supply for providing power to the sensor, other components useful for signal processing and data storage, and a telemetry module for transmitting data from the sensor electronics module to one or more display devices. The electronics can be affixed to a printed circuit board (PCB) or the like and can take a variety of forms. For example, the electronics can take the form of an integrated circuit (IC), such as an Application-Specific Integrated Circuit (ASIC), a microcontroller, and / or a processor.

[0089] Display devices 210, 220, 230, and / or 240 are configured to display displayable sensor data, including analyte data, that may be transmitted by sensor electronics module 204. Each of display devices 210, 220, 230, or 240 may include a display, such as a touchscreen display 212, 222, 232, / or 242, for displaying sensor data to a user and / or accepting input from a user. For example, a graphical user interface (GUI) may be presented to the user for such purposes. In some embodiments, the display device may include other types of user interfaces, such as a voice user interface, instead of or in addition to a touchscreen display, for communicating sensor data to a user of the display device and / or accepting user input. Display devices 210, 220, 230, and 240 may be examples of display device 107 illustrated in FIG. 1 used to display sensor data to a user of FIG. 1 and / or receive input from a user.

[0090] In some embodiments, one, some, or all of the display devices are configured to display or otherwise communicate the sensor data as it is communicated from the sensor electronics module (e.g., in a data package sent to the respective display device), without any additional anticipated processing required for calibration and real-time display of the sensor data.

[0091] The display devices may include custom display devices designed specifically to display a particular type of displayable sensor data associated with the analyte data received from the sensor electronics module. In certain embodiments, the display devices may be configured to provide alerts / alarms based on the displayable sensor data. Display device 210 is an example of such a custom device. In some embodiments, one of the display devices is a smart phone, such as display device 220 representing a mobile phone, configured to display a graphical representation of continuous sensor data (e.g., including current and historical data) using a commercially available operating system (OS). Other display devices may include display device 230 representing a tablet, display device 240 representing a smartwatch, medical device 208 (e.g., an insulin delivery device or a blood glucose meter), and / or a desktop or laptop computer (not shown).

[0092] Because different display devices have different user interfaces, the contents of the data package (e.g., amount, format, and / or type of data displayed, alarms, etc.) can be customized (e.g., programmed differently by the manufacturer and / or end user) for each particular display device. Thus, in certain implementations, multiple different display devices can wirelessly communicate directly with the sensor electronics module (e.g., on-skin sensor electronics module 204 physically connected to the continuous analyte sensor 202, etc.) during a sensor session to enable multiple different types and / or levels of display and / or functionality associated with the displayable sensor data. In certain embodiments, the types of alarms customized for each particular display device, the number of alarms customized for each particular display device, the timing of alarms customized for each particular display device, and / or the threshold levels configured (e.g., for triggering) for each of the alarms are based on the outputs 144 stored in the user profile 118 for each user (e.g., as mentioned, the outputs 144 can indicate the user's current health, the user's renal status, current treatments recommended for the user, and / or the user's physiological parameters when experiencing different symptoms).

[0093] As mentioned, the sensor electronics module 204 may be in communication with a medical device 208. The medical device 208 may be a passive device in some exemplary embodiments of the present disclosure. For example, the medical device 208 may be an insulin pump for administering insulin to a user. For various reasons, it may be desirable for such an insulin pump to receive and track potassium and / or glucose values ​​transmitted from the continuous analyte monitoring system 104, and the continuous analyte sensor 202 is configured to measure potassium and / or glucose.

[0094] Additionally, as mentioned, the sensor electronics module 204 may also communicate with other non-analyte sensors 206. In some cases, any one or more of the non-analyte sensors mentioned below may be integrated into the continuous analyte monitoring system 104 or may be separate and separate from the continuous analyte monitoring system 104. The non-analyte sensors 206 may include, but are not limited to, altimeter sensors, accelerometer sensors, temperature sensors, respiration rate sensors, sweat sensors, and the like. The non-analyte sensors 206 may also include monitors such as heart rate monitors, ECG monitors, blood pressure monitors, pulse oximeters, caloric intake, and drug delivery devices. The non-analyte sensors 206 may also include data systems for measuring non-patient specific phenomena such as time, ambient pressure, or ambient temperature, which may include barometric pressure sensors, external air temperature sensors, or clocks, timers, or other time measures of the time the sensor was initially inserted or the time of remaining sensor life compared to the insertion time, and may be used as calibration or other data inputs for the algorithmic models. One or more of these non-analyte sensors 206 may provide data to the decision support engine 114, which is described further below. In some embodiments, a user may manually provide a portion of the data for processing by the training server system 140 and / or the decision support engine 114 of FIG.

[0095] In certain embodiments, the non-analyte sensor 206 may be combined in any other configuration, such as, for example, combined with one or more continuous analyte sensors 202. As an illustrative example, a non-analyte sensor, such as a temperature sensor, may be combined with a continuous analyte sensor 202 configured to measure potassium to form a potassium / temperature sensor that is used to transmit sensor data to the sensor electronics module 204 using a common communications circuit. As another illustrative example, a non-analyte sensor, such as a temperature sensor, may be combined with a multi-analyte sensor 202 configured to measure potassium and glucose to form a potassium / glucose / temperature sensor that is used to transmit sensor data to the sensor electronics module 204 using a common communications circuit.

[0096] In certain embodiments, a wireless access point (WAP) may be used to couple one or more of the continuous analyte monitoring system 104, the multiple display devices, the medical devices 208, and / or the non-analyte sensors 206 to one another. For example, the WAP 138 may provide Wi-Fi and / or cellular connectivity among these devices. Near Field Communication (NFC) and / or Bluetooth may also be used between the devices depicted in the diagram 200 of FIG. 2.

[0097] Figure 3 illustrates example inputs and example metrics calculated based on the inputs for use by the decision support system of Figure 1, according to some embodiments disclosed herein. In particular, Figure 3 provides a more detailed illustration of the example inputs and example metrics introduced in Figure 1.

[0098] 3 illustrates exemplary inputs 128 on the left, the decision support engine 114 including the applications 106 and DAM 116 in the center, and alerts 130 on the right. In certain embodiments, each of the metrics 130 may correspond to one or more values, such as discrete numeric values, ranges, or qualitative values ​​(such as high / medium / low, or stable / unstable). The applications 106 obtain inputs 128 via one or more channels (e.g., manual user input, sensors / monitors, other applications running on the display device 107, EMR, etc.). As previously mentioned, in certain embodiments, the inputs 128 may be processed by the DAM 116 and / or the decision support engine 114 to output metrics 130. The inputs and metrics 130 may be used by the decision support engine 114 to provide decision support to a user. For example, the inputs 128 and metrics 130 may be used by the training server system 140 to train and deploy one or more machine learning models that are used by the decision support engine 114 to provide decision support regarding renal health, including the presence and / or severity of renal disease.

[0099] In some embodiments, beginning with input 128, food consumption information may include information about one or more of meals, snacks, and / or beverages, such as amount, content (milligrams (mg) of potassium, carbohydrates, fat, protein, etc.), order of consumption, and time of consumption. In certain embodiments, food consumption may be provided by the user through manual input, by providing a photo through an application configured to recognize food types and amounts, and / or by scanning a barcode or menu. In various examples, meal amounts may be manually entered as one or more of calories, amount (e.g., "3 cookies"), menu items (e.g., "Royale with Cheese"), and / or food exchanges (e.g., 1 fruit, 1 dairy product). In some examples, meal information may be received through a convenient user interface provided by application 106.

[0100] In certain embodiments, food consumption information (type of food (e.g., liquid or solid, snack or meal, etc.) and / or composition of food (e.g., carbohydrates, fat, protein, etc.)) may be automatically determined based on information provided by one or more sensors. Some exemplary sensors may include body sound sensors (e.g., abdominal sounds may be used to detect type of meal, e.g., liquid / solid meal, snack / meal, etc.), radio frequency sensors, cameras, hyperspectral cameras, and / or analyte (e.g., insulin, glucose, lactate, etc.) sensors for determining food type and / or composition.

[0101] In certain embodiments, the food consumption input by the user may be related to potassium consumed by the user. The potassium for consumption may include any natural or engineered food or beverage that contains potassium, such as, for example, apricot juice, bananas, or potatoes.

[0102] In certain embodiments, athletic information is also provided as an input. The athletic information can be any information surrounding an activity, such as an activity that requires physical exercise by the user. For example, the athletic information can range from information related to low intensity (e.g., walking a few steps) and high intensity (e.g., running 5 miles) physical exercise. In certain embodiments, the athletic information can be provided by an accelerometer sensor on a wearable device, such as a watch, fitness tracker, and / or patch. In certain embodiments, the athletic information can also be provided through manual user input and / or through a proxy sensor that measures changes in heart rate (or other cardiac metrics) and a predictive algorithm. When predicting that a user is exercising based on their sensor data, the user can be asked to confirm whether exercise is occurring, what type of exercise, and / or the level of intensity used during the exercise over a certain period of time. This data can be used to train the system to learn about the user's exercise patterns to reduce the need for confirmation questions over time. Other analytes and sensor data can also be included in this training set, including the analytes and other measured factors described herein, including time factors such as hours and days.

[0103] In certain embodiments, user statistics such as one or more of age, height, weight, BMI, body composition (e.g., body fat percentage), height, build, or other information may also be provided as input. In certain embodiments, user statistics may be provided through a user interface, by interfacing with an electronic source such as an electronic medical record, and / or from a measurement device. In certain embodiments, the measurement device may include one or more of a wireless, e.g., Bluetooth enabled, weight scale, and / or camera that may communicate with the display device 107 to provide user data, for example.

[0104] In certain embodiments, treatment / medication information is also provided as an input. The medical information may include information regarding the type, dosage, and / or timing of when one or more medications are taken by the user. As mentioned herein, the medical information may include information regarding one or more diuretics, one or more drugs known to reduce potassium, one or more drugs known to damage the kidney, one or more drugs known to control complications of kidney disease that are prescribed to the user, and / or one or more medications to treat one or more symptoms of kidney disease, hyperkalemia, hypokalemia, diabetes, and / or other conditions and diseases that the user may have. The treatment information may include information regarding different lifestyle habits, surgical procedures, and / or other non-invasive procedures recommended by the user's physician. For example, the user's physician may recommend to the user to increase / decrease potassium intake, exercise for at least 30 minutes per day, or increase insulin dosage or other medications, etc., to maintain and / or improve kidney health and reduce the occurrence of hyperkalemia and / or hypokalemia.

[0105] As another example, a medical professional may recommend that the user engage in home dialysis treatment and / or dialysis treatment at a clinic. Dialysis is a treatment for kidney failure that cleanses the body of unwanted toxins, waste products, and excess fluids by filtering the user's blood. A user with end stage renal disease (ESRD) (e.g., CKD stage 5) may be prescribed dialysis treatment to supplement and / or replace the filtration generally performed by the kidneys, given that dialysis helps balance potassium, phosphorus, and sodium levels in the patient's body. The dialysis treatment may be hemodialysis or peritoneal dialysis. In hemodialysis, blood is pumped from the user's body to an artificial kidney machine and returned to the body by tubes connecting the user to the machine. In peritoneal dialysis, the lining of the user's stomach acts as a natural filter. Thus, information regarding dialysis treatment for the user may be included in the treatment / medication information. In certain embodiments, the treatment / medication information may be provided through manual user input.

[0106] In certain embodiments, analyte sensor data may also be provided as an input, for example, through the continuous analyte monitoring system 104. In certain embodiments, the analyte sensor data may include potassium data (e.g., a user's potassium level) measured by at least a CPM (or a multi-analyte sensor configured to measure at least potassium) that is part of the continuous analyte monitoring system 104. In certain embodiments, the analyte sensor data may include glucose data measured by at least a glucose sensor (or a multi-analyte sensor configured to measure at least glucose) that is part of or separate from the continuous analyte monitoring system 104. In certain embodiments, the analyte sensor data may include creatinine data measured by at least a creatinine sensor (or a multi-analyte sensor configured to measure at least creatinine) that is separate from or part of the continuous analyte monitoring system 104. In certain embodiments, the analyte sensor data may include BUN data measured by at least a BUN sensor (or a multi-analyte sensor configured to measure at least BUN) that is separate from or part of the continuous analyte monitoring system 104.

[0107] In certain embodiments, the analyte sensor data may include ammonia data measured by at least an ammonia sensor (or a multi-analyte sensor configured to measure at least ammonia) that is part of the continuous analyte monitoring system 104. In certain embodiments, the analyte sensor data may include C-peptide data measured by at least a C-peptide sensor (or a multi-analyte sensor configured to measure at least C-peptide) that is part of the continuous analyte monitoring system 104. In certain embodiments, the analyte sensor data may include cystatin C data measured by at least a cystatin C sensor (or a multi-analyte sensor configured to measure at least cystatin C) in the continuous analyte monitoring system 104. In certain embodiments, the analyte sensor data may include lactate data measured by at least a lactate sensor (or a multi-analyte sensor configured to measure at least lactate) that is part of the continuous analyte monitoring system 104.

[0108] In certain embodiments, input may also be received from one or more non-analyte sensors, such as the non-analyte sensor 206 described with respect to Figure 2. Input from such non-analyte sensors 206 may include information related to the user's heart rate, heart rate variability (e.g., the variance in time between heart beats), ECG data, respiratory rate, oxygen saturation, blood pressure, or temperature (e.g., to detect illness, physical activity, etc.). In certain embodiments, electromagnetic sensors may also detect low power radio frequency (RF) fields emitted from an object or a tool in contact with or in proximity to an object, which may provide information about user activity or location.

[0109] In certain embodiments, the input received from the non-analyte sensor may include input regarding the user's insulin delivery. In particular, the input regarding the user's insulin delivery may be received via a wireless connection on the smart pen, via a user input, and / or from an insulin pump. The insulin delivery information may include one or more of insulin amount, delivery time, and the like. Other configurations, such as insulin action time or duration of insulin action, may also be received as input.

[0110] In certain embodiments, a time may also be provided as an input, such as a time of day or a time from a real-time clock. For example, in certain embodiments, the input analyte data may be time-stamped to indicate the date and time when the analyte measurement was made for the user.

[0111] User input of any of the inputs 128 referred to above may be through a user interface, such as the user interface of the display device 107 of FIG.

[0112] As mentioned above, in certain embodiments, the DAM 116 and / or decision support engine (e.g., using one or more trained models) determines or calculates metrics 130 for a user based on the inputs 128. An example list of metrics 130 is shown in FIG.

[0113] In certain embodiments, the potassium level may be determined from sensor data (e.g., the CPM of the continuous analyte monitoring system 104, potassium measurements obtained from a sweat sensor configured to measure potassium in sweat, where the sweat sensor may be one of the non-analyte sensors 206). For example, the potassium level may reference time-stamped potassium lactate measurements or values ​​that are continuously generated and stored over time.

[0114] In certain embodiments, the potassium baseline may be determined from sensor data (e.g., potassium measurements obtained from a continuous lactate sensor of the continuous analyte monitoring system 104). For example, the potassium baseline may be determined based on historical levels of potassium at rest (e.g., two weeks prior), an average of resting potassium levels over time, or time-split data of the user's potassium levels at rest. The potassium baseline represents the user's normal potassium levels during periods when significant fluctuations in potassium production are typically not expected. A user's baseline potassium is generally expected to remain constant over time unless challenged through actions such as consumption of potassium or potassium-rich foods, or unless changed as a result of declining renal health or impaired renal function.

[0115] Additionally, each user may have a different potassium baseline. In certain embodiments, a user's potassium baseline may be determined by calculating an average of the user's potassium levels over a particular amount of time where significant fluctuations are not expected. For example, a user's baseline potassium may be determined over a period of time where the user is asleep, sitting in a chair, or other time period where the user is sedentary and has not consumed food or medications that would reduce or increase potassium levels (e.g., there are no external conditions present that would affect the potassium baseline). In certain embodiments, the DAM 116 may continuously calculate the potassium baseline, timestamp the calculated potassium baseline, and store the corresponding information in the user's profile 118.

[0116] In certain other embodiments, the DAM 116 may use potassium levels measured over a period of time, during which the user has engaged in exercise and / or consumed potassium, and / or there are external conditions that affect the potassium baseline, for at least a subset of the period. In such embodiments, the DAM 116 may, in some examples, first identify which measured potassium values ​​should not be used to calculate the potassium baseline by identifying which potassium values ​​have been affected by external events, such as food consumption, exercise, medication, or other perturbations that interfere with the capture of the potassium baseline measurements. The DAM 116 may then eliminate such measurements when calculating the user's potassium baseline. In some other examples, the DAM 116 may calculate the potassium baseline by first determining the percentage of the number of potassium values ​​measured during this period that represent the lowest lactate value measured. The DAM 116 may then take an average of such potassium values ​​to determine the potassium baseline level.

[0117] In certain embodiments, the absolute maximum potassium level may be determined from sensor data (e.g., potassium measurements obtained from continuous CPM of the continuous analyte monitoring system 104), health / sickness metrics (e.g., as described in more detail below), and / or disease stage metrics (e.g., as described in more detail below). The absolute maximum potassium level represents the maximum potassium level of a user that is determined to be unsafe over a period of time (e.g., hourly, weekly, daily, etc.). In certain embodiments, the absolute maximum potassium level may be consistent across all users (e.g., set at 5.5 mmol / L based on current medical guidelines). In certain other embodiments, each patient may have a different absolute maximum potassium level. For example, the absolute maximum potassium level may be lower for a user diagnosed with stage 1 CKD (e.g., normal or high GFR (GFR>90 mL / min)) than a user diagnosed with stage 5 end-stage CKD (e.g., GFR<15 mL / min) who also has hyperkalemia. In certain embodiments, the absolute maximum potassium level per patient may change over time. For example, a user may initially be assigned an absolute maximum potassium level based on clinical input, which may be adjusted over time based on other sensor data, disease stage, comorbidities, etc. for the patient.

[0118] For example, a user's absolute maximum potassium level may change over time as the user's renal health, renal disease, and / or one or more other diseases progress and / or improve. In certain embodiments, a first absolute maximum potassium level may be determined for a period of time in which there are no external conditions that affect potassium levels, and a second absolute maximum potassium level may be determined for a period of time in which there are external conditions that affect potassium levels (e.g., a period of time in which the user is consuming potassium, exercising, taking medications that affect potassium levels, etc.).

[0119] In certain embodiments, the absolute minimum potassium level may be determined from sensor data (e.g., potassium measurements obtained from a continuous potassium sensor of the continuous analyte monitoring system 104), health / illness metrics (e.g., as described in more detail below), and / or disease stage metrics (e.g., as described in more detail below). The absolute minimum potassium level represents the minimum potassium level of a user that is determined to be unsafe over a period of time (e.g., hourly, weekly, daily, etc.). In certain embodiments, the absolute minimum potassium level may be consistent across all users (e.g., set based on current medical guidelines). In certain other embodiments, each user may have a different absolute minimum potassium level. For example, the absolute minimum potassium level may be lower for a user diagnosed with stage 1 CKD (e.g., normal or high GFR (GFR>90 mL / min)) than a user diagnosed with stage 5 end-stage CKD (e.g., GFR<15 mL / min) who also has hypokalemia. In certain embodiments, the absolute minimum potassium level per patient may change over time. For example, a user may initially be assigned an absolute minimum potassium level based on clinical input, which may be adjusted over time based on other sensor data, disease stage, comorbidities, etc. for the patient.

[0120] For example, a user's absolute minimum potassium level may change over time as the user's renal health, renal disease, and / or one or more other diseases progress and / or improve. In certain embodiments, a first absolute minimum potassium level may be determined for a period of time in which there are no external conditions that affect potassium levels, and a second absolute minimum potassium level may be determined for a period of time in which there are external conditions that affect potassium levels (e.g., a period of time in which the user is consuming potassium, exercising, taking medications that affect potassium levels, etc.).

[0121] In certain embodiments, the potassium level rate of change may be determined from sensor data (e.g., potassium measurements obtained over time from a CPM of the continuous analyte monitoring system 104). For example, the potassium level rate of change refers to the rate at which one or more time-stamped potassium measurements or values ​​change relative to one or more other time-stamped potassium measurements or values. The potassium level rate of change may be determined over a period of one second or more, one minute or more, one hour or more, one day or more, etc.

[0122] In certain embodiments, the potassium level rate of change may be determined for a number of potassium measurements using a line of best fit that represents the average rate of change for the number of potassium measurements. In this example, as current potassium measurements are obtained from a sensor (e.g., a continuous potassium sensor of the continuous analyte monitoring system 104), the line of best fit may change to better represent the rate of change based on the current potassium measurements. Additionally, based on the rate of change progression over time, different statistics (e.g., average, standard deviation, etc.) may be calculated to determine when significant changes in potassium level rate of change occur.

[0123] In certain embodiments, the determined potassium level rate of change may be marked as "rapidly increasing" or "rapidly decreasing". As used herein, "rapidly" may describe a potassium level rate of change that is clinically significant and indicates a tendency that the patient's potassium level is likely to breach an absolute maximum potassium level or an absolute minimum potassium level (e.g., a potassium threshold) within the next defined time period. In other words, a predictive trend (e.g., produced by the decision support engine 114 using one or more trained models) may, in some cases, indicate that the patient is likely to reach an absolute maximum potassium level within a specified time period (e.g., one or two hours) based on the determined potassium level rate of change. Thus, such a potassium level rate of change may be marked as "rapidly increasing". Similarly, a predictive trend (e.g., produced by the decision support engine 114 using one or more trained models) may, in some cases, indicate that the patient is likely to reach an absolute minimum potassium level within a specified time period (e.g., one or two hours) based on the determined potassium level rate of change. Thus, such a potassium level rate of change may be marked as "rapidly decreasing".

[0124] In certain embodiments, the determined potassium level rate of change may be considered in light of non-analyte sensor data to determine whether there are one or more external factors affecting the user's potassium level. For example, if the potassium level rate of change is marked as "rapidly increasing" and it is known that the user is consuming a meal or completing an exercise session, the potassium level rate of change may be compared to other potassium level rate of change measurements when the user is consuming a meal or completing an exercise session. In other words, the potassium level rate of change when the user is consuming a meal or completing an exercise session may be analyzed separately from the potassium level rate of change when the user is at rest.

[0125] In certain embodiments, the potassium baseline rate of change may be determined from a potassium baseline determined over time for a user. For example, the potassium baseline rate of change refers to the rate at which one or more time-stamped potassium baselines for a user change relative to one or more other time-stamped potassium baselines for the same user. The potassium baseline rate of change may be determined over one second or more, one minute or more, one hour or more, one day or more, etc.

[0126] In certain embodiments, the potassium clearance rate may be determined from sensor data (e.g., potassium measurements obtained from a CPM of the continuous analyte monitoring system 104) following consumption of a known or estimated amount of potassium. The potassium clearance rate analyzed over time may be indicative of renal function. In particular, the slope of the potassium clearance curve during a first period of time (e.g., after consuming a known amount of potassium) compared to the slope of the potassium clearance curve during a second period of time (e.g., after consuming the same amount of potassium) may be indicative of the ability of the kidneys to function, and more specifically, to maintain potassium homeostasis (e.g., the potassium clearance rate may be slower when a user's kidneys are impaired than when the user's kidneys are healthy).

[0127] In certain embodiments, the potassium clearance rate may be determined by calculating the slope between an initial potassium value (e.g., during a period of increased potassium levels) and a potassium baseline associated with the user. In certain embodiments, the potassium clearance rate may be calculated over time until the user's increased potassium level reaches some value relative to the user's potassium baseline (e.g., a % of the user's potassium baseline). The potassium clearance rate calculated over time may be time-stamped and stored in the user's profile 118.

[0128] In certain embodiments, potassium trend may be determined based on potassium levels over a particular time period. In certain embodiments, potassium trend may be determined based on potassium baseline over a particular time period. In certain embodiments, potassium trend may be determined based on absolute minimum potassium levels over a particular time period. In certain embodiments, potassium trend may be determined based on absolute maximum potassium levels over a particular time period. In certain embodiments, potassium trend may be determined based on potassium level rate of change over a particular time period. In certain embodiments, potassium trend may be determined based on potassium baseline rate of change over a particular time period. In certain embodiments, potassium trend may be determined based on calculated potassium clearance rate, potassium level variability, potassium threshold-based trend, and / or potassium level average over a particular time period.

[0129] In certain embodiments, the glucose level may be determined from sensor data (eg, blood glucose measurements obtained from a continuous lactate sensor of the continuous analyte monitoring system 104).

[0130] In certain embodiments, the glucose level change rate may be determined from sensor data (e.g., glucose measurements obtained over time from a continuous glucose monitor (CGM) of the continuous analyte monitoring system 104). For example, the glucose level change rate refers to the rate at which one or more time-stamped glucose measurements or values ​​change relative to one or more other time-stamped glucose measurements or values. The glucose level change rate may be determined over a second or more, a minute or more, an hour or more, a day or more, etc.

[0131] In certain embodiments, blood glucose trends may be determined based on glucose levels over a particular period of time. In certain embodiments, glucose trends may be determined based on the rate of change of glucose levels over a particular period of time.

[0132] In certain embodiments, insulin sensitivity may be determined using medical history data, real-time data, or a combination thereof, and may be based on one or more inputs 128, such as, for example, one or more of food consumption information, continuous analyte sensor data, non-analyte sensor data (e.g., insulin delivery information from an insulin device), etc. Insulin sensitivity refers to how a user's cells respond to insulin. Improving insulin sensitivity for a user may help reduce insulin resistance in the user.

[0133] In certain embodiments, remaining insulin (insulin on board) may be determined using non-analyte sensor data inputs (e.g., insulin delivery information) and / or known or learned (e.g., from user data) insulin time-action profiles that may take into account both basal metabolic rate (e.g., insulin updates to keep the body moving) and insulin usage driven by activity or food consumption.

[0134] In certain embodiments, the creatinine level may be determined from sensor data (eg, creatinine measurements obtained from the continuous analyte monitoring system 104).

[0135] In certain embodiments, the absolute maximum creatinine level may be determined from sensor data (e.g., creatinine measurements obtained from a continuous creatinine sensor of the continuous analyte monitoring system 104), health / disease metrics (e.g., as described in more detail below), and / or disease stage metrics (e.g., as described in more detail below). The absolute maximum creatinine level represents the user's maximum creatinine level determined to be unsafe over a period of time (e.g., hourly, weekly, daily, etc.). Each user may have a different absolute maximum creatinine level. A user's absolute maximum creatinine level may change over time as the user's renal health, renal disease, and / or one or more other diseases progress and / or improve.

[0136] In certain embodiments, the rate of change of creatinine level may be determined from sensor data (e.g., creatinine measurements obtained over time from a continuous analyte monitoring system 104 creatinine sensor). For example, creatinine level rate of change refers to a rate describing how one or more time-stamped creatinine measurements or values ​​change relative to one or more other time-stamped creatinine measurements or values. The rate of change of creatinine level may be determined over one second or more, one minute or more, one hour or more, one day or more, etc. In certain embodiments, an average creatinine level may be calculated to determine the rate of change of the user's calculated average creatinine level.

[0137] In certain embodiments, the creatinine trend may be determined based on creatinine levels over a particular period of time.

[0138] In certain embodiments, the BUN level may be determined from sensor data (e.g., BUN measurements obtained from the continuous analyte monitoring system 104). In certain embodiments, the BUN trend may be determined based on the BUN level over a particular period of time.

[0139] In certain embodiments, the ammonia level may be determined from sensor data (e.g., ammonia measurements obtained from the continuous analyte monitoring system 104). In certain embodiments, the ammonia trend may be determined based on the ammonia level over a particular period of time.

[0140] In certain embodiments, C-peptide levels may be determined from sensor data (e.g., C-peptide measurements obtained from the continuous analyte monitoring system 104). In certain embodiments, C-peptide trends may be determined based on C-peptide levels over a particular period of time.

[0141] In certain embodiments, cystatin C levels may be determined from sensor data (e.g., cystatin C measurements obtained from the continuous analyte monitoring system 104). In certain embodiments, cystatin C trends may be determined based on cystatin C levels over a particular period of time.

[0142] In certain embodiments, the health and illness metrics may be determined, for example, from physiological sensors (e.g., temperature), activity sensors, or a combination thereof, based on one or more of user inputs (e.g., pregnancy information or known illness or disease information). In certain embodiments, based on the values ​​of the health and illness metrics, for example, the user's condition may be defined as one or more of healthy, sick, rest, or fatigued.

[0143] In certain embodiments, disease stage metrics, such as for renal disease, may be determined based on one or more of user inputs or outputs provided by the decision support engine 114 illustrated in FIG. 1. In certain embodiments, exemplary disease stages for renal disease may include AKI, stage 1 CKD with normal or high GFR (e.g., GFR>90 mL / min), stage 2 mild CKD (e.g., GFR=60-89 mL / min), stage 3A moderate CKD (e.g., GFR=45-59 mL / min), stage 3B moderate CKD (e.g., GFR=30-44 mL / min), stage 4 severe CKD (e.g., GFR=15-29 mL / min), and stage 5 end stage CKD (e.g., GFR<15 mL / min). In certain embodiments, exemplary disease stages may be expressed as GFR values / ranges, severity scores, etc.

[0144] In certain embodiments, the meal state metric may indicate the user's state with respect to food consumption. For example, the meal state may indicate whether the user is in one of a fasting state, a pre-meal state, a fed state, a post-meal reaction state, or a stable state. In certain embodiments, the meal state may also indicate remaining nutrients, e.g., meals, snacks, or beverages consumed, which may be determined from food consumption information, mealtime information, and / or digestibility information, which may be correlated, for example, to food type, amount, and / or order (e.g., which food / beverage was eaten first).

[0145] In certain embodiments, the eating habits metric is based on the content and timing of a user's meals. For example, if the eating habits metric is on a scale of 0 to 1, in an embodiment, the better / healthier the diet a user eats, the higher the user's eating habits metric will be toward 1. Also, in an embodiment, the more a user's food consumption adheres to a particular time schedule or recommended diet, the closer their eating habits metric will be toward 1.

[0146] In certain embodiments, medication adherence is measured by one or more metrics that indicate how committed a user is to their medication regimen. In certain embodiments, medication adherence metrics are calculated based on one or more of the timing of a user's taking medication (e.g., whether the user is on time or on schedule), the type of medication (e.g., whether the user is taking the correct type of medication), and the dosage of the medication (e.g., whether the user is taking the correct dosage). In certain embodiments, a user's medication adherence may be determined through user input and / or based on analyte data received from the analyte monitoring system 104 in a clinical trial in which medication consumption and the timing of such medication consumption are monitored.

[0147] In certain embodiments, the activity level metric may indicate a user's activity level. In certain embodiments, the activity level metric is determined based on input from an activity sensor or other physiological sensor, such as, for example, the non-analyte sensor 206. In certain embodiments, the activity level metric may be calculated by the DAM 116 based on one or more of inputs 128, such as one or more of motion information, non-analyte sensor data (e.g., accelerometer data), time, user input, etc. In certain embodiments, the activity level may be expressed as a user's step rate. The activity level metric may be time-stamped so that it may be contemporaneously correlated with the user's lactate level.

[0148] In certain embodiments, the exercise regimen metric may indicate one or more of what type of activity the user is engaged in, the corresponding intensity of such activity, the frequency with which the user is engaged in such activity, etc. In certain embodiments, the exercise regimen metric may be calculated based on one or more of analyte sensor data inputs (e.g., from a lactate monitor, a glucose monitor, etc.), non-analyte sensor data inputs (e.g., from an accelerometer, a heart rate monitor, a respiration rate sensor, etc.), calendar inputs, user inputs, etc.

[0149] In certain embodiments, body temperature metrics may be calculated by the DAM 116 based on the input 128, more specifically, non-analyte sensor data from a temperature sensor. In certain embodiments, heart rate metrics (including, for example, heart rate and heart rate variability) may be calculated by the DAM 116 based on the input 128, more specifically, non-analyte sensor data from a heart rate sensor. In certain embodiments, respiration metrics may be calculated by the DAM 113 based on the input 128, more specifically, non-analyte sensor data from a respiration rate sensor.

[0150] In certain embodiments, as described in more detail below, when it is determined that a user is experiencing one or more symptoms associated with potassium imbalance, physiological parameters associated with the user (e.g., potassium level, potassium level rate of change, glucose level, C-peptide level, heart rate, blood pressure, etc.) may be stored as metrics 130. In particular, physiological parameters associated with a user when the user is experiencing (1) potassium fluctuations, (2) numbness, (3) tingling, (4) shortness of breath (or chest pain), (5) muscle weakness, and / or (6) arrhythmia may be stored as metrics 130. In certain embodiments, such physiological parameters may be analyzed over time to provide an indication of improvement or worsening of the user's renal disease. In certain embodiments, user-specific values ​​of physiological parameters for different symptoms experienced by the user may be useful inputs for training one or a model designed to assess the presence and / or severity of renal disease in the user. In certain embodiments, the user-specific values ​​of the physiological parameters for different symptoms experienced by the user may be used to create one or more personalized models specific to the user to more accurately predict (1) the user's renal health, including the presence and / or severity of renal disease, and (2) one or more symptoms of the user associated with potassium imbalance.

[0151] Exemplary methods and systems for providing renal health decision support - Patents.com FIG. 4A is an example workflow 400A for determining (e.g., using a machine learning model and / or algorithm (e.g., a rule-based model)) the likelihood of one or more symptoms commonly associated with potassium imbalance (e.g., due to renal disease, in some cases) occurring in real time or within a specified time period and retraining or updating the model based on patient input and / or diagnostic tests, according to certain embodiments of the present disclosure. FIG. 4B is an example workflow 400B for determining the likelihood of one or more symptoms commonly associated with potassium imbalance occurring in real time or within a specified time period using a model and providing one or more recommendations for treatment for a user based on the determined likelihood, according to certain embodiments of the present disclosure. For example, workflow 400B may be executed to provide a user with a prediction of one or more symptoms and further provide one or more recommendations for treatment for the user based on the prediction, using a continuous analyte monitoring system 104 including at least a continuous potassium monitor (CPM) 202, as illustrated in FIGS. 1 and 2. In certain embodiments, operations 400A and 400B may be combined (e.g., such that in addition to using the model to provide a symptom prediction and retraining the model, treatment recommendations are also provided to the patient based on the prediction), while in some other embodiments operations 400A and 400B may be performed separately. Workflow 400A and workflow 400B may be executed by decision support system 100 to collect data, such as inputs 128 and metrics 130, including, for example, analyte data, patient information, and non-analyte sensor data described above, to provide a prediction of the risk or likelihood of one or more symptoms of potassium imbalance occurring for a patient in real time (e.g., a user is currently experiencing one or more of these symptoms) or within a specified time period (e.g., a user is expected to experience one or more of these symptoms within 5 minutes, 30 minutes, 1 hour, etc.). In certain embodiments, the prediction may be presented as a percentage of the likelihood of one or more symptoms of potassium imbalance occurring for a patient in real time or within a specified time period.

[0152] In certain embodiments, the decision support system 100 presented herein may be configured to identify or predict that a patient will experience one or more symptoms commonly associated with potassium imbalance, even when the patient is not aware that they are experiencing such symptoms. In particular, in the early stages of renal disease (e.g., when potassium imbalance is common), patients may have few symptoms associated with renal disease. However, patients may not realize that they have renal disease until the condition has progressed. Thus, by being able to identify the occurrence of such symptoms of potassium imbalance and confirm the occurrence of such symptoms based on user input or through one or more sensor data (e.g., generated by the non-analyte sensor 206), the decision support system 100 presented herein provides information that may be important for the early detection of renal disease in a patient.

[0153] In certain embodiments, the decision support engine 114 of the decision support system 100 may use various algorithms or artificial intelligence (AI) models, such as machine learning models trained based on patient-specific and / or population data, to provide renal disease symptom predictions. The algorithms and / or machine learning models may consider one or more inputs 128 and / or metrics 130 described with respect to FIG. 3 for a patient when predicting whether the patient is experiencing or is likely to experience one or more different symptoms commonly associated with potassium imbalance.

[0154] In certain embodiments, the one or more machine learning models may include a multiple input, single output (MISO) model trained to predict a patient's risk or likelihood of experiencing one of the symptoms of potassium imbalance (e.g., (1) potassium fluctuations (hypokalemia, hyperkalemia, rapid rate of change, variability, etc.), (2) numbness, (3) tingling, (4) shortness of breath, (5) muscle weakness, (6) arrhythmia, or (7) death). For example, one model may be trained to predict the likelihood of current or future occurrence of numbness, while another model may be trained to predict current or future occurrence of tingling.

[0155] In certain embodiments, the one or more machine learning models may include a multiple input, multiple output (MIMO) model trained to predict the risk or likelihood of a patient experiencing two or more of the symptoms of potassium imbalance. For example, a single model may be trained to predict the likelihood of current or future occurrence of numbness and tingling. In certain embodiments, the model may be trained to output a vector having multiple values, each value corresponding to the likelihood of one of the symptoms the model is trained to predict the likelihood of occurrence. As an illustrative example, the vector output by the MIMO model may include five values, each value corresponding to one of the six symptoms of potassium imbalance (e.g., listed above). A first value indicating 90% may indicate that there is a 90% chance that the patient is experiencing numbness, or that there is a 90% risk or likelihood that the patient will experience numbness within a given time limit. Meanwhile, a second value indicating 0% may indicate that there is a 0% change in the patient experiencing tingling, or that the patient has a 0% risk or likelihood of experiencing tingling within a given time limit.

[0156] One or more machine learning models described herein for making such predictions may first be trained using the population data. Methods for training one or more machine learning models may be described in more detail below with respect to FIG.

[0157] In certain embodiments, as an alternative to using machine learning models, the decision support engine 114 may use rule-based models to predict the risk or likelihood that a patient will experience one of the symptoms. Rule-based models involve using a set of rules to analyze data. These rules are often referred to as "If Statements" because they tend to follow the lines of "If X happens then do or conclude Y." In particular, the decision support engine 114 may apply rule statements (e.g., if, then statements) to predict one or more renal disease symptoms for a patient.

[0158] Such rules may be maintained in a reference library by the decision support engine 114. For example, the reference library may maintain ranges of potassium levels and ranges of potassium level change rates that may be mapped to different symptoms experienced by patients with renal disease. In certain embodiments, such rules may be determined based on empirical studies or analysis of historical patient records, such as records stored in the historical medical records database 112. In some cases, the reference library may be very granular. For example, other factors may be used in the reference library to create such "rules." Other factors may include gender, age, diet, medical history, family medical history, body mass index (BMI), etc. The increased granularity may provide a more accurate output.

[0159] As mentioned, Figure 4A illustrates a workflow 400A for updating a model used to determine the likelihood that one or more symptoms commonly associated with renal disease will occur in real time or within a specified time period. The model can be (1) a machine learning model that is initially trained on population data, or (2) a rule-based model. The workflow 400 is described below with reference to Figures 1 and 2 and their components.

[0160] At block 402, the workflow 400 begins by continuously monitoring one or more analytes of a patient, such as the user 102 illustrated in FIG. 1, over a number of time periods to obtain analyte data. The one or more analytes monitored may include at least potassium, and thus the analyte data may include potassium data. Block 402 may be performed in certain embodiments by the continuous analyte monitoring system 104 illustrated in FIGS. 1 and 2, and more specifically, the continuous analyte sensor 202 illustrated in FIG. 2. For example, the continuous analyte monitoring system 104 may include a CPM 202 configured to measure the patient's potassium level.

[0161] As mentioned, potassium is one of the most important minerals in the body. Potassium helps regulate fluid balance, muscle contraction, and nerve signals. A high-potassium diet can also help reduce blood pressure and water retention, protect against stroke, and prevent osteoporosis and kidney stones. Approximately 98% of potassium in the body is stored intracellularly, and the remaining 2% is stored extracellularly. The kidneys are primarily responsible for maintaining total body potassium content by matching potassium intake with potassium excretion. Regulation of renal potassium excretion occurs over several hours, and thus changes in extracellular potassium concentration are initially buffered by the movement of potassium in and out of skeletal muscle.

[0162] The regulation of potassium distribution between the intracellular and extracellular spaces is referred to as internal potassium balance. The most important factors regulating this movement under normal conditions are insulin and catecholamines (e.g., dopamine, epinephrine (adrenaline), and norepinephrine). In other words, the kidneys play a major role in potassium homeostasis by the renal mechanisms that transport potassium and regulate its secretion, reabsorption, and excretion. However, when the kidneys are damaged or lose the ability to function, they may no longer be able to remove excess potassium, and thus potassium levels may increase in the body. Thus, potassium monitoring may prove useful in assessing renal health, diagnosing or monitoring renal disease, and / or predicting and preventing symptoms associated with hyperkalemia that may result from renal disease.

[0163] In certain embodiments, techniques may be implemented to account for inaccurate potassium levels. For example, the same device or another device may measure free hemoglobin and / or use colorimetric measurements to monitor fluid abnormalities. Based on this information, the device may in some cases not report a potassium level and / or may alert the user that the potassium level is likely inaccurate. In some cases where the exact amount of free hemoglobin may be measured, a correction factor may then be applied to provide a more accurate potassium level for the patient.

[0164] The primary analyte for measurements described herein is potassium, although in certain embodiments, other analytes may be considered. In particular, combining potassium measurements with additional analyte data may help to further inform analysis regarding renal disease condition prediction. For example, monitoring additional types of analytes, such as glucose, creatinine, lactate, blood urea nitrogen (BUN), ammonia, C-peptide, and / or cystatin C, as measured by the continuous analyte monitoring system 104, may provide additional insight into the prediction of one or more patient conditions commonly associated with renal disease.

[0165] Additional insight gained from using a combination of analytes rather than just potassium may increase the accuracy of the prediction. For example, the probability of accurately predicting symptoms associated with renal disease may be a function of the number of analytes measured for a patient. For example, in some examples, the probability of accurately predicting that a patient is experiencing numbness using only potassium data (in addition to other non-analyte data) may be less than the probability of accurately predicting that a patient is experiencing numbness using potassium and glucose data (in addition to other non-analyte data), which may also be less than the probability of accurately predicting that a patient is experiencing numbness using potassium, glucose, and creatinine data (in addition to other non-analyte data) for analysis.

[0166] Additionally, multiple sensors or a single CPM measuring data over time to confirm or reject the presence of symptoms can be used to understand if the patient-reported data is accurate. For example, if a patient is entering a clinically dangerous threshold of hypokalemia or hyperkalemia, but has not yet confirmed that the patient is experiencing symptoms of potassium imbalance, the patient may be unaware of the pending danger. The use of additional analytes for analysis can help increase the patient's confidence that the lack of confirmed symptoms is a false negative response by the patient and that the patient is actually in danger. Thus, based on this confidence level, the system may be able to take one or more actions to help protect the patient. The one or more actions may include, but are not limited to, notifying family members, health care providers, emergency personnel, important contacts entered by the patient, and / or monitoring service companies trained to identify false negatives or confirm that the patient is stable. In some aspects, the same personnel-mediated triage methods may be used to flag false positives for patient-reported symptoms if a sensor is malfunctioning or if such a sensor is functioning as expected and reading normal levels for the patient, but the patient reports severe symptoms (e.g., intentionally or accidentally). Triage protocols or secondary confirmation sources, such as key contacts, family members, trained service company employees, or medical professionals previously designated by the patient, their care team, or their insurance provider, may be used to confirm or deny whether the patient is in fact experiencing the symptoms they claim to be experiencing.

[0167] In certain embodiments described herein, for example, the combination of analytes measured and collected by one or more (e.g., multi-analyte) sensors for renal disease condition prediction includes potassium and at least one of creatinine, BUN, ammonia, C-peptide, lactate, and cystatin C, however, other analyte combinations may be considered.

[0168] For example, in certain embodiments, the continuous analyte monitoring system 104 may continuously monitor the patient's potassium and glucose levels during multiple time periods at block 402. In certain embodiments, the measured glucose concentrations may be used in conjunction with the potassium levels to predict one or more conditions of the patient.

[0169] In particular, glucose is a simple sugar (e.g., monosaccharide). Glucose can be ingested as well as produced in the body from proteins, fats, and carbohydrates. Increased glucose stimulates insulin release. Insulin causes cells to take up glucose and potassium as fuel. Thus, insulin stimulates potassium and glucose uptake by cells, thereby reducing serum (e.g., extracellular) potassium and glucose levels. In some cases, when a patient's glucose level increases and the rate of change of glucose levels in the patient's body is high, excess insulin may be produced, thereby causing intracellular movement of potassium. On the other hand, when a patient's glucose level decreases and the rate of change of glucose levels in the patient's body is low, insulin secretion may be less. Low insulin may lead to limited access of glucose and potassium by cells, and therefore extracellular glucose and potassium levels may increase.

[0170] In some cases, the patient may suffer from insulin resistance. Insulin resistance occurs when the patient's muscle, fat and liver cells do not respond adequately to insulin. Thus, glucose metabolism and intracellular potassium movement may be impaired. As a result, the patient's pancreas makes more insulin to help glucose and insulin enter the patient's cells.

[0171] Insulin resistance may have different effects on glucose metabolism compared to potassium metabolism. Furthermore, the effects of insulin resistance on glucose metabolism and potassium metabolism may be different for different patients. In particular, a patient with a first insulin resistance may require X doses of insulin to reduce extracellular potassium levels by Y, while another patient with a second insulin resistance may require Z doses of insulin to reduce extracellular potassium levels by Y. For example, a diabetic patient with insulin resistance may be at higher risk for hyperkalemia and may require higher insulin concentrations when using insulin for hyperkalemia management.

[0172] In other words, a patient's glucose level may affect the amount of insulin produced by the body, which in turn is expected to reduce the amount of extracellular potassium available to be measured by CPM202 (e.g., which may measure potassium in interstitial fluid), but the effects of insulin resistance on potassium metabolism may cause less than expected amounts of potassium to move into cells even when the patient's glucose and / or insulin are elevated. Thus, making accurate predictions regarding a patient's potassium levels and better understanding the true health of the patient's kidneys may require understanding not only the effects of insulin resistance on glucose metabolism, but also the effects of insulin resistance on potassium metabolism. For example, in some cases, glucose levels, in addition to the patient's insulin resistance, may help understand whether the patient's abnormal potassium levels are actually due to a decline in the patient's kidney health or due to some other reason, such as increased insulin resistance, which may be caused by heart disease, liver disease, obesity, etc.

[0173] Furthermore, in certain embodiments, the measured glucose concentration may be used as an additional input when predicting one or more symptoms of a patient in addition to the patient's insulin resistance. For example, as glucose in the bloodstream increases over time, the kidneys may filter excess blood. Over time, this extra work puts more pressure on the nephrons, which often results in the nephrons losing their biofiltration capacity, thereby impairing kidney function. Thus, evaluation of glucose levels over time may provide insight into the overall health of a patient's kidneys, which may help determine whether a patient is experiencing or will experience one or more symptoms of kidney disease and / or determine the likelihood that the patient will develop kidney disease in the future.

[0174] In another example, at block 402, the continuous analyte monitoring system 104 may continuously monitor the patient's potassium and creatinine levels over multiple time periods. In certain embodiments, the measured level of creatinine may be used to assign a confidence level to the patient's measured potassium level, the confidence level indicating a level of certainty that the potassium level measured for the patient reflects the patient's actual potassium level. Additionally, in certain embodiments, the measured level of creatinine may be used to inform an analysis as to whether the patient is experiencing or is likely to experience one or more symptoms associated with potassium imbalance.

[0175] In particular, creatinine is a waste product produced by muscles from the breakdown of a compound called creatine. Creatinine is removed from the body by the kidneys, which filter almost all of it from the blood and release it into the urine. Thus, creatinine levels can provide insight into kidney health and function. Thus, a patient who experiences high levels of measured extracellular potassium and is expected to have impaired renal function (e.g., considering that excess potassium is not being filtered out of the body) can also be expected to experience high levels of measured creatinine (e.g., considering that impaired kidneys are likely unable to remove creatinine from the blood). Thus, if the measured creatinine level is also high, a higher confidence level can be assigned to the measured potassium level for the patient. Furthermore, the assumption that the patient's kidneys are impaired can be strengthened, thereby increasing the likelihood that the patient's kidneys are not functioning properly and increasing the likelihood that the patient is actually experiencing (or about to experience) one or more symptoms commonly associated with renal disease.

[0176] In yet another example, at block 402, the continuous analyte monitoring system 104 may continuously monitor the patient's potassium and BUN levels over multiple time periods. In certain embodiments, the measured level of BUN may also be used to assign a confidence level to the patient's measured potassium level. Additionally, in certain embodiments, the measured level of BUN may be used to inform an analysis as to whether the patient is experiencing or is likely to experience one or more symptoms associated with potassium imbalance.

[0177] In particular, the liver produces ammonia, which contains nitrogen, after breaking down proteins used by cells in the body. Nitrogen combines with other elements, such as carbon, hydrogen, and oxygen, to form urea, a chemical waste product. Urea travels from the liver to the kidneys through the bloodstream. Healthy kidneys filter the urea and remove other waste products from the blood, which then leave the body through urine. Thus, BUN levels (e.g., levels of nitrogen content in urea) can provide insight into kidney health and function. Thus, a patient who experiences high levels of measured extracellular potassium and is assumed to have impaired renal function may be expected to also experience high levels of measured BUN (e.g., given that impaired kidneys are likely unable to filter urea and remove other waste products from the blood). Thus, if the measured BUN level is also high, a higher confidence level may be assigned to the measured potassium level for the patient. Furthermore, the assumption that the patient's kidneys are damaged may be strengthened, thereby increasing the likelihood that the patient's kidneys are not functioning properly and that the patient is in fact experiencing (or about to experience) one or more symptoms commonly associated with kidney disease.

[0178] In yet another example, the continuous analyte monitoring system 104 may continuously monitor the patient's potassium and ammonia levels during multiple time periods in block 402. In certain embodiments, the measured ammonia levels may be used as an additional input when predicting one or more conditions of the patient.

[0179] In particular, ammonia is a waste product created by the body during the digestion of proteins. Normally, ammonia is processed by the liver, where it is converted into another waste product called urea. Thus, a patient's ammonia level can be evaluated to monitor the patient's liver decompensation. Endogenous insulin is processed by the liver, with up to 80% being removed, and is not released by the liver back into circulation. However, with reduced liver function, more endogenous insulin may be circulating, which, as previously mentioned, may shift extracellular potassium into cells due to an excess of insulin in patients who do not experience insulin resistance. In other words, the amount of extracellular potassium available to be measured by CPM202 may be reduced.

[0180] Thus, patients without renal disease and not experiencing insulin resistance, who would not be expected to experience one or more symptoms, may not exhibit "normal" extracellular measured potassium levels, because although their kidneys are functioning properly, the patient may have excess endogenous insulin. Thus, in these cases, the patient's insulin levels may mask the actual health of the patient's kidneys (e.g., that the kidneys are in good health), thus providing the patient with abnormal potassium levels caused by factors other than declining kidney health. Thus, evaluation of ammonia levels over time can provide insight into the health of the patient's liver, which can provide additional insight into insulin levels and their impact on measured potassium levels. With this information, more accurate predictions can be made when determining whether a patient is experiencing one or more symptoms of renal disease. For example, by continuously measuring ammonia, one or more models presented herein may be able to determine that a patient's low measured potassium levels are not due to the patient having damaged kidneys, but instead due to the fact that the patient's liver is deteriorating, which may result from several mechanisms, including excess insulin that pushes potassium into cells.

[0181] In yet another example, the continuous analyte monitoring system 104 may continuously monitor the patient's potassium and C-peptide levels during multiple time periods in block 402. In certain embodiments, the measured C-peptide levels may be used to determine the patient's endogenous insulin levels and provide additional context regarding the potassium measurements collected by the continuous analyte sensor 202.

[0182] In particular, C-peptide is a substance that is produced when the hormone insulin is produced and released into the body. Currently, there is no way to measure insulin in the body, so the level of C-peptide in the blood can be measured to indicate how much insulin is being made by the pancreas. In many cases, C-peptide is measured to indicate the difference between the insulin the body produces and the insulin that is injected into the body. Understanding a patient's insulin levels can provide insight into why a patient's measured potassium levels are lower than normal or higher than normal to better predict one or more symptoms of the patient.

[0183] In yet another example, in block 402, the continuous analyte monitoring system 104 may continuously monitor the patient's potassium and cystatin C levels during multiple time periods. In certain embodiments, the measured cystatin C levels may be used as an additional input when predicting one or more conditions of the patient. For example, cystatin C is a low molecular weight protein member of the cystatin superfamily of cysteine ​​protease inhibitors. Cystatin C is produced by nucleated cells and exhibits a steady production rate. Cystatin C is freely filtered by the glomerulus and metabolized after tubular reabsorption. Cystatin C has been proposed as a useful alternative marker to be tested in urine and blood, especially in situations where creatinine-based estimates of GFR fail to provide an accurate estimate. If renal function and GFR decline, blood levels of cystatin C are expected to rise. In some cases, serum levels of cystatin C provide a more accurate test of renal function (as represented by GFR) than serum creatinine levels. Thus, assessment of cystatin C levels over time can provide insight into the overall renal health of a patient, which can aid in determining whether a patient is experiencing one or more symptoms of renal disease.

[0184] In certain embodiments, for example, one or more algorithms and / or models described herein for predicting the risk or likelihood that a patient will experience one or more different symptoms may be configured to use input from one or more sensors measuring one or more of the aforementioned analytes. Parameters and / or thresholds of such algorithms and / or models may be altered based at least in part on the number of analytes being measured for an input to reflect knowledge gained from each of the other analytes being measured.

[0185] In addition to continuously monitoring one or more analytes of the user during a first time period to obtain analyte data in block 402, optionally in certain embodiments, workflow 400 may also include monitoring other sensor data (e.g., non-analyte data) during multiple time periods using one or more other non-analyte sensors or devices (e.g., non-analyte sensor 206 and / or medical device 208 of FIG. 2 ).

[0186] As previously mentioned, the non-analyte sensors and devices may include, but are not limited to, one or more of an insulin pump, a tactile sensor, an electrocardiogram (ECG) sensor or heart rate monitor, a blood pressure sensor, a sweat sensor, a respiration sensor, a thermometer, an impedance sensor, a peritoneal dialysis machine, a hemodialysis machine, a continuous positive airway pressure (CPAP) machine, a sensor or device provided by the display device 107 (e.g., an accelerometer, a camera, a global positioning system (GPS), a heart rate monitor, etc.) or other user accessory (e.g., a smart watch), or any other sensor or device that provides relevant information about the user. Metrics such as the metrics 130 illustrated in FIG. 3 may be calculated using measurement data from each of these additional sensors. As further illustrated in FIG. 3, the metrics 130 calculated from the non-analyte sensor or device data may include body temperature, heart rate (including heart rate variability), respiration rate, etc. In certain embodiments, described in more detail below, metrics 130 calculated from non-analyte sensor or device data may be used to further inform analysis regarding renal disease condition prediction.

[0187] In certain embodiments, one or more of these non-analyte sensors and / or devices may be worn by the user to aid in detecting periods of increased physical activity by the user. Such non-analyte sensors and / or devices may include accelerometers, ECG sensors, blood pressure sensors, heart rate monitors, impedance sensors, dialysis machines, etc. In certain embodiments, data measured and collected from periods of increased physical activity and periods of sedentary activity by the user may be used to analyze potassium levels during each of these identified periods and predict whether the patient is experiencing one or more symptoms of renal disease.

[0188] As an example, in certain embodiments, a sudden pattern of sedentary activity in a generally active patient detected by such a non-analyte sensor and / or device may be a trigger to ask the patient if the patient is feeling well or experiencing one or more symptoms. As another example, in certain embodiments, a sudden pattern of sedentary activity in a generally active patient detected by such a non-analyte sensor and / or device may be a trigger to engage the patient in one or more diagnostic tests to determine if the patient is not feeling well and / or experiencing one or more symptoms (e.g., a resting respiration rate test to determine if the patient is experiencing shortness of breath, a common symptom of renal disease). An embodiment relating to the use of non-analyte sensor and / or device data to detect abnormal patterns in a patient for symptom prediction is described in more detail below with respect to block 418 in FIG. 4.

[0189] In certain embodiments, one or more non-analyte sensors and / or devices that may be worn by the patient may include a temperature sensor. The temperature sensor may be worn to help correct the measured potassium levels to predict one or more symptoms of the patient. In particular, a correlation exists between body temperature and potassium release in the human body.

[0190] For example, at higher body temperature, the body sweats and potassium is excreted through sweat. Thus, higher body temperature may induce lower potassium levels. In some cases where a patient's measured potassium level appears to be lower than normal for that patient, it may be concluded that the patient's lower than normal potassium level is due to an inability to concentrate urine due to impaired tubular response to vasopressin (ADH), which in some cases leads to the excretion of large amounts of diluted urine containing potassium. However, such reduced potassium levels may actually be due to excess sweat, e.g., excess potassium leaving the patient's body. In other words, the patient's high body temperature may affect the amount of extracellular potassium measured by the CPM 202. Thus, in some embodiments, a secondary sensor, such as a temperature sensor, may be used to indicate that dynamic and sudden changes to potassium may be the result of sweating (e.g., associated with exercise, hot and / or humid weather, etc.) compared to negative health events.

[0191] There may also be a correlation between the temperature of the environment and the user's potassium levels. For example, during exercise, a user may experience high potassium levels when exercising outdoors in hot temperatures. Thus, when a user is outdoors in a hot environment, they may need to reduce intensity to avoid high potassium levels, even though potassium is excreted through sweat.

[0192] As another example, in some cases, potassium levels measured on a patient initially experiencing hypothermia (e.g., occurring when the body loses heat faster than it can produce heat, resulting in a dangerously low body temperature) may appear to be lower than normal potassium levels associated with the patient. Such reduced potassium levels may be due to the patient experiencing the onset of hypothermia. In particular, hypothermia may cause an initial decrease in extracellular potassium levels. Hypothermic hypokalemia is associated with an intracellular shift rather than an actual net loss. The intracellular shift is caused by a variety of factors, such as enhanced function of Na+K+ATPase, beta-adrenergic stimulation, pH, and membrane stabilization in deep hypothermia.

[0193] However, as hypothermia progresses in a patient, irreversible cellular damage may occur. In particular, the body may experience a lack of enzymes that function at low temperatures and a blockage of active transport. Thus, as hypothermia progresses, the patient's measured potassium levels may increase from lower than normal to higher than normal levels. In other words, the patient's low body temperature may affect the amount of extracellular potassium measured by the CPM 202 over time. Thus, monitoring the patient's body temperature may help inform the patient's measured potassium levels so that the measured potassium levels may be corrected before predicting one or more symptoms of the patient. Additionally, in certain aspects, an external temperature probe may be used to measure the temperature of the patient's immediate surroundings (e.g., a very close area) to predict whether the patient is actually experiencing hypothermia or a decrease in body temperature. Furthermore, in some aspects, an accelerometer or piezoelectric sensor may be used to identify whether the patient is shivering to determine whether the patient is experiencing a decrease in body temperature that may affect the patient's measured potassium levels.

[0194] In certain embodiments, the one or more non-analyte sensors and / or devices that may be worn by the patient may include a blood pressure sensor. Blood pressure measurements collected from the blood pressure sensor are used to provide additional insight into the patient's renal health.

[0195] In particular, CKD and hypertension are closely related. Typically, as blood pressure increases, renal function declines. Thus, evaluation of a patient's blood pressure level over time can provide additional insight into the patient's renal health. Thus, a patient who experiences high levels of measured extracellular potassium and is assumed to have impaired renal function (e.g., considering that excess potassium is not being filtered out of the body) can be expected to also experience high blood pressure levels. Thus, if the patient's blood pressure level is also high, the assumption that the patient's kidneys are impaired can be strengthened, thereby increasing the likelihood that the patient's kidneys are not functioning properly and that the patient is actually experiencing (or about to experience) one or more symptoms commonly associated with renal disease.

[0196] In certain embodiments, the one or more non-analyte sensors and / or devices that may be worn by a patient may include an ECG sensor and / or a heart rate monitor. As known in the art, an ECG device is a device that measures the electrical activity of the heart. In certain embodiments, heart rate measurements and heart rate variability information collected from an ECG sensor and / or a heart rate monitor may be used in combination with CPM to more accurately predict the risk or likelihood that a patient will be diagnosed with hyperkalemia and / or experience one or more cardiac events (e.g., arrhythmia and / or sudden cardiac death).

[0197] In particular, a patient's potassium levels measured using CPM can be used to detect hyperkalemia or hypokalemia, including levels at which a cardiac event is likely to occur. Hyperkalemia increases the risk of cardiac arrhythmia episodes and sudden death. However, in some cases, potassium measurements alone may not be an adequate indicator of the severity of hyperkalemia, and therefore certain embodiments described herein propose to use CPM and ECG / heart rate monitors in combination to more accurately predict or determine the risk or likelihood that a patient will experience one or more cardiac events.

[0198] For example, sometimes higher than normal levels of potassium measured on a patient can lead to changes in cardiac behavior that can be reflected on the ECG. Specifically, classical teachings of time-series ECG changes of hyperkalemia can include peaked T waves, prolonged PR intervals, widening of the QRS complex, loss of P waves, "sinusoids," and asystole (e.g., colloquially referred to as flat-line, representing the cessation of electrical and mechanical activity of the heart). Thus, ECG measurements on a patient, in combination with CPM measurements, can be crucial to provide a full-picture assessment of the physiological significance of hyperkalemia. Furthermore, ECG findings generally correlate with measured potassium levels; however, potentially life-threatening arrhythmias can occur without warning at almost any level of hyperkalemia. In other words, ECG / heart rate monitors can be used in combination with CPM to better predict the risk or likelihood that a patient will experience an arrhythmia, to (1) avoid the arrhythmia in some cases when it would be catastrophic, and (2) intervene in some cases before the arrhythmia becomes fatal.

[0199] Additionally, in certain embodiments, heart rate measurements and heart rate variability information collected from an ECG sensor and / or heart rate monitor may be used in combination with measured potassium levels to predict one or more conditions of a patient. More specifically, heart rate measurements and heart rate variability information may be analyzed to determine a time delay constant between measured local potassium levels (e.g., potassium levels in a local region where the patient's potassium is being measured) and their effect on the heart.

[0200] In particular, a CPM 202 measuring extracellular potassium may show a different potassium level in the local area where potassium is being measured than the potassium level in the whole body or in cardiac tissue. This is because potassium levels rise systemically first and then interstitially. In other words, interstitial potassium levels may lag systemic potassium levels. For example, potassium levels in cardiac tissue may precede interstitial levels depending on where the potassium sample is taken to determine such potassium levels in a patient.

[0201] Thus, in certain embodiments, a standardized or personalized delay constant may be calculated, where the delay constant corresponds to the amount of time it takes for the patient's interstitial potassium level to reach the patient's systemic potassium level. The calculated delay constant may be applied to the measured potassium levels to better understand systemic potassium levels (e.g., higher potassium levels) for more accurate prediction of hyperkalemia and / or one or more cardiac events. In particular, in certain embodiments, one or more algorithms may use the calculated delay constant in combination with the patient's interstitial potassium trend, interstitial potassium rate of change, and / or measured interstitial potassium levels to predict the risk or likelihood that the patient is currently experiencing or is about to experience an arrhythmia (or be diagnosed with hyperkalemia) within a certain time period.

[0202] Additionally, in certain embodiments, recorded episodes of vomiting and / or diarrhea for a patient may be collected and used in combination with the measured potassium levels to predict one or more conditions for the patient, particularly since in some cases prolonged vomiting, diarrhea, or both may result in excessive potassium loss from the digestive system.

[0203] At block 404, the workflow 400 continues by processing the analyte data from the multiple time periods to determine at least one rate of change of the patient's potassium. Block 404 may be performed by the decision support engine 114 in certain embodiments.

[0204] As referred to, potassium level rate of change refers to a rate indicative of change of one or more time-stamped potassium measurements or values ​​relative to one or more other time-stamped potassium measurements or values. In certain embodiments, the machine learning models described herein used to provide a symptom prediction may include one or more features related not only to the patient's potassium level but also to the patient's potassium level rate of change. For example, an exemplary machine learning model may include weights applied to features associated with one or more rates of change in potassium levels. Thus, in certain embodiments, prior to use of the machine learning model, at least one potassium level rate of change for the patient may need to be calculated for input into the model.

[0205] Furthermore, in certain embodiments, the rule-based models described herein used to provide abnormal potassium level symptom predictions may include one or more rules related not only to a patient's potassium level, but also to the patient's potassium level rate of change. For example, a reference library used to define one or more rules for a rule-based model may maintain a range of potassium levels and a range of potassium level rate of change that may be mapped to different symptoms experienced by patients with renal disease. Thus, prior to use of a rule-based model, at least one potassium level rate of change for a patient may need to be calculated for input into the model. Additional features, such as potassium level rate of change, added to the model may, in some cases, enable a more accurate prediction of one or more symptoms of renal disease for a patient.

[0206] At block 406, workflow 400 continues by determining the likelihood that the patient is experiencing (or will experience) a symptom using (1) at least one rate of change of the patient's potassium (e.g., determined in block 404) and (2) the trained model or one or more rules (e.g., a rule-based model).

[0207] As mentioned, different methods for determining the likelihood that a patient is experiencing (or will experience) a symptom may be used by the decision support engine 114. One method may include using a machine learning model, and another method may include using a rule-based model that defines one or more rules. In certain embodiments, such machine learning models are trained and such rule-based models include rules determined based on empirical studies or analysis of historical patient records from the medical history record database 112. Either the machine learning model or the rule-based model may be deployed for a particular patient using at least the CPM 202.

[0208] After determining in block 410 that the patient is likely experiencing (or will experience) the symptom, the workflow 400 continues by the decision support engine 114 confirming or denying whether the patient is experiencing the symptom that the decision support engine 114 predicted for the patient in block 406. For example, the presence of the symptom may be confirmed or denied through manual input by the patient, contextual data (e.g., camera), other input data (e.g., sensor data), or observations and / or input by someone other than the patient (e.g., a medical professional).

[0209] In certain embodiments, the decision support engine 114 may confirm or deny whether the patient is experiencing a symptom by transmitting instructions to the patient prompting the patient to confirm or deny whether the patient is actually experiencing the predicted symptom. In certain embodiments, the instructions may be transmitted to the patient as a message via the application 106.

[0210] In some embodiments, the actions at block 410 may occur only if the likelihood determined at block 406 exceeds a threshold. For example, if the decision support engine 114 determines that there is a 20% likelihood that the patient is experiencing numbness and the predefined threshold is set at 70%, then the decision support engine 114 may not determine at block 410 whether the patient is experiencing the symptom. In certain embodiments, the actions at block 410 may occur when the model is less confident as to whether the patient is experiencing the symptom. For example, in some aspects, a prediction may be considered uncertain (e.g., less confident) if the decision support engine 114 determines that there is between 40% and 60% likelihood that the patient is experiencing the symptom. In this case, the decision support engine 114 may determine at block 410 whether the patient is experiencing the symptom. In certain other embodiments, the actions at block 410 may occur periodically (e.g., not necessarily in response to a likelihood determination at block 406) such that the decision support engine 114 periodically determines whether the patient is experiencing the symptom.

[0211] While the embodiment of FIG. 4A illustrates the use of a MISO model used to predict whether a patient is experiencing (or likely to experience) only one symptom commonly associated with renal disease, in embodiments in which a MIMO model (e.g., a model used to predict whether a patient is experiencing (or likely to experience) two or more symptoms commonly associated with renal disease) is used, different thresholds may be predefined for different symptoms. Further, the determination in block 410 as to whether the patient is experiencing a symptom may be based on a predefined threshold for that symptom and a determined percentage of the likelihood that the patient is experiencing that symptom. Thus, in some cases, the decision support engine 114 may make one or more determinations in block 410, where a MIMO model is used, and the vector output for the MIMO model indicates a percentage for two or more symptoms that are greater than their respective predefined thresholds.

[0212] If, at block 410, the decision support engine 114 confirms that the patient is experiencing the symptom predicted by the decision support engine 114 (e.g., the decision support engine predicts that the patient is experiencing numbness and the decision support engine 114 confirms that the patient is experiencing numbness), then, at block 412, the workflow 400 continues by associating the patient's physiological parameters with the patient experiencing the symptom. Specifically, in certain embodiments, after confirming that the patient is experiencing a symptom commonly associated with renal disease, the decision support engine 114 may be configured to determine which of the patient's physiological parameters are associated with the patient experiencing the identified symptom. For example, the decision support engine 114 may determine the patient's potassium level range, potassium level rate of change range, heart rate, respiratory rate, values ​​of other metrics 130, etc., at the time the patient is confirmed to be experiencing numbness. In certain embodiments, the decision support engine 114 may determine the patient's physiological parameters (e.g., inputs to a model) for which a prediction was made as the physiological parameters corresponding to the particular patient's symptom. In certain embodiments, the decision support engine 114 may be configured to measure one or more physiological parameters of the patient after determining that the patient is experiencing a symptom and associate these physiological parameters with the patient experiencing the identified symptom.

[0213] In certain embodiments in which the patient is prompted to confirm or deny that they are experiencing a symptom, the decision support engine 114 may be configured to record the answers provided by the patient in block 410. In certain embodiments, the answers may be time-stamped to indicate the date and time the answers were recorded to the patient. By recording such answers of the patient, the decision support engine 114 may learn about different physiological parameter thresholds and their association with different subjective sensations (e.g., sensation of numbness, sensation of muscle weakness, etc.) being experienced by the patient.

[0214] In certain embodiments, the decision support engine 114 may be configured to record such physiological parameters in the metrics 130 stored for the patient in the user profile 118. In certain embodiments, the physiological parameters may be time-stamped to indicate the date and time the patient's physiological parameters were recorded.

[0215] In certain embodiments, the patient-specific values ​​of the physiological parameters for the different symptoms experienced by the user may be used to train a machine learning model (e.g., originally trained using only population data) and update one or more rules for a rule-based model (e.g., if the rules were originally created based on population data) to more accurately predict one or more symptoms of the patient related to potassium imbalance, to create one or more personalized models specific to the patient. Thus, at block 414, the workflow 400 continues by using the associations created at block 412 to update one or more rules used for the trained machine learning model or the rule-based model.

[0216] In other words, each model may use information provided by the decision support engine 114 and / or the patient as feedback to personalize the model for the patient. In particular, through this feedback loop, each model may learn about the patient's personalized symptom thresholds (e.g., potassium level thresholds when the patient will experience tingling pain, etc.). This approach may also allow each model to be personalized for each patient and to continue to learn over time as the patient's baseline and symptom thresholds (e.g., potassium levels, potassium rate of change, etc.) change (e.g., as the patient's renal health gets better or worse).

[0217] In certain embodiments, updating one or more rules for the rule-based model may include updating existing rules and / or creating new rules based on an analysis of parameters collected for the patient when the patient is identified as experiencing the identified symptom. In certain embodiments, updating the trained machine learning model may include retraining the machine learning model using physiological parameters collected for the patient when the patient is identified as experiencing the identified symptom. In some cases, retraining the machine learning model may include adjusting one or more weights for one or more features of the machine learning model.

[0218] For example, a machine learning model initially trained based on population data to predict whether a patient will experience tingling may include at least one or more features associated with different potassium ranges, potassium change rates, and / or other parameters discussed herein. Weights applied to the model's features may remain unchanged or be altered (e.g., increased or decreased) based on feedback from the patient indicating that the patient is or is not experiencing tingling as predicted by the model.

[0219] Similar methods for adjusting the weights of a machine learning model based on different feedback provided by the patient can be used to optimize and personalize the model based on the patient's specific data.

[0220] Referring back to block 410, if the decision support engine 114 determines that the patient is not experiencing a symptom predicted by the decision support engine 114 (e.g., the decision support engine predicts that the patient is experiencing numbness and the patient indicates to the decision support engine 114 that they are not experiencing numbness), then in block 416 the workflow 400 continues by the decision support engine 114 initiating one or more diagnostic tests. As used herein, a diagnostic test is a variety of evaluations performed to screen and detect a particular condition in a patient.

[0221] In certain embodiments, one or more diagnostic tests initiated by the decision support engine 114 may include using haptic feedback to communicate with the patient after the patient denies that they are experiencing numbness (e.g., a symptom predicted by the decision support engine 114 using workflow 400). The decision support engine 114 may use the patient's response or lack thereof to confirm / deny whether the patient is experiencing numbness.

[0222] As an illustrative example, the decision support engine 114 may use the deployed machine learning model to predict that there is an 80% likelihood that the patient is experiencing numbness. Thus, the decision support engine 114 may confirm or deny whether the patient is experiencing such numbness, and the patient may, in some cases, identify that he or she is not experiencing numbness. To confirm the determination, the decision support engine 114 may be configured to require the patient to engage with a tactile sensor (e.g., positioned on the user's body as a separate sensor or as part of the CPM). The tactile sensor may be configured to create a combination of force, vibration, and / or kinesthetic sensations to the patient. The patient may confirm or deny whether or not they were able to feel this sensation. In a situation where the patient was unable to feel this sensation, the decision support engine 114 may determine that although the patient reported that they were not experiencing numbness, the patient did in fact experience numbness, however, the patient was unaware.

[0223] In certain embodiments, the one or more diagnostic tests initiated by the decision support engine 114 may include using a respiratory sensor to determine the patient's respiratory rate after the patient has ruled out experiencing shortness of breath (e.g., a symptom predicted by the decision support engine 114 using workflow 400). The decision support engine 114 may use the patient's determined respiratory rate to confirm / rule whether the patient is experiencing shortness of breath.

[0224] In certain embodiments, one or more diagnostic tests initiated by the decision support engine 114 may include using haptic feedback to communicate with the patient after the patient denies that they are experiencing a tingling pain (e.g., a symptom predicted by the decision support engine 114 using workflow 400). The decision support engine 114 may use the patient's reaction or lack thereof to confirm / deny whether the patient is experiencing a tingling pain.

[0225] In certain embodiments, the one or more diagnostic tests initiated by the decision support engine 114 may include using an electromyography (EMG) sensor to determine the patient's muscle response after ruling out that the patient is experiencing muscle weakness (e.g., a symptom predicted by the decision support engine 114 using workflow 400). As known in the art, an EMG sensor is a sensor that measures small electrical signals generated by the patient's muscles as the patient moves their muscles. The decision support engine 114 may use the patient's determined muscle response to confirm / rule out whether the patient is experiencing muscle weakness.

[0226] In certain embodiments, one or more diagnostic tests initiated by the decision support engine 114 may include triggering the patient to squeeze a widget (e.g., a stress ball) after the patient denies experiencing muscle weakness (e.g., a symptom predicted by the decision support engine 114 using workflow 400). In certain embodiments, the decision support engine 114 may determine the patient's strength by asking the patient if the patient was able to fully squeeze the widget and confirming with the patient as to whether the patient was able to fully squeeze the widget. In certain embodiments, the widget may include a pressure sensor, and thus the decision support engine 114 may determine the patient's strength based on the pressure measured by the pressure sensor in the widget. The decision support engine 114 may use the patient's strength, or lack thereof, in squeezing the widget to confirm / deny whether the patient is experiencing muscle weakness.

[0227] In certain embodiments, the one or more diagnostic tests initiated by the decision support engine 114 may include using a heart rate monitor (in combination with one or more sensors used to detect that the patient is sedentary) to determine the patient's resting heart rate after ruling out that the patient is experiencing arrhythmia (e.g., a symptom predicted by the decision support engine 114 using workflow 400). The decision support engine 114 may use the patient's determined resting heart rate to confirm / rule out whether the patient is experiencing arrhythmia.

[0228] In certain embodiments, one or more other diagnostic tests similar to those mentioned herein may be used to confirm / rule whether a patient is experiencing symptoms predicted by the decision support engine 114 using the workflow 400.

[0229] In certain embodiments, the decision support engine 114 may be triggered to initiate one or more diagnostic tests in block 416 after determining the likelihood that the patient is experiencing (or will experience) the symptom. In other words, in some embodiments, the operation in block 410 may be skipped. Instead, the decision support engine 114 may be configured to initiate one or more diagnostic tests to determine whether the patient is experiencing the identified symptom without the decision support engine 114 determining whether the patient is experiencing the symptom. In some cases, the patient may not engage or respond to a prompt asking the user to confirm / deny experiencing the identified symptom, and thus, skipping the operation in block 410 may prove more efficient to predict the patient's one or more symptoms. In certain embodiments, if the decision support engine 114 determines in block 406 that the patient is likely currently experiencing the symptom, the decision support engine 114 may be triggered to automatically initiate one or more tests. On the other hand, if in block 406 the decision support engine 114 determines that the patient is likely to experience symptoms within the specified period of time (e.g., five minutes), the decision support engine 114 may wait the specified period of time before initiating one or more diagnostic tests.

[0230] At block 418, the workflow 400 continues by the decision support engine 114 determining whether the results of one or more diagnostic tests confirm that the patient is experiencing an identified symptom (e.g., a symptom predicted by the decision support engine 114 using the workflow 400). If at block 418 the decision support engine 114 determines that the results of one or more diagnostic tests confirm that the patient is experiencing an identified symptom, then at block 412 the workflow 400 continues by associating the user's physiological parameters with the user experiencing the symptom. Specifically, in certain embodiments, after confirming that the patient is experiencing a symptom commonly associated with renal disease (even after the patient denies experiencing the symptom in some cases), the decision support engine 114 may be configured to determine which physiological parameters of the patient have been associated with the patient experiencing the identified symptom. At block 414, the workflow 400 continues by using the association created at block 412 to update one or more rules used for the trained machine learning model or rule-based model.

[0231] On the other hand, if at block 418 the decision support engine 114 determines that the results of the one or more diagnostic tests do not confirm that the patient is experiencing the identified symptom, then at block 420 the workflow 400 continues by associating the user's physiological parameters with a user not experiencing the symptom. Specifically, in certain embodiments, after confirming that the patient is not experiencing symptoms commonly associated with renal disease, the decision support engine 114 may be configured to determine which physiological parameters of the patient are associated with the patient not experiencing the identified symptom. At block 414, the workflow 400 continues by using the associations made at block 412 to update one or more rules used for the trained machine learning model or rule-based model.

[0232] In certain embodiments, updating one or more rules for the rule-based model may include the decision support engine 114 or some other server analyzing parameters collected about the patient when it is determined that the patient is not experiencing the identified symptom to update the rule-based model and defining and / or modifying one or more rules based on the updated rule-based model. In certain embodiments, updating the trained machine learning model may include the training server system 140 retraining the machine learning model using diagnostic test results and / or physiological parameters collected about the patient when it is determined that the patient is not experiencing the identified symptom. In some cases, retraining the machine learning model may include adjusting one or more weights for one or more features of the machine learning model.

[0233] Although the embodiment of FIG. 4A illustrates the decision support engine 114 determining whether the patient is experiencing the symptom at block 410, in certain other embodiments, the decision support engine 114 may not provide a determination. Instead, in certain embodiments, diagnostic tests may be performed to confirm or negate the predicted symptom. In certain other embodiments, neither a determination from the decision support engine 114 nor results from diagnostic tests may be obtained, and instead, after block 406, the decision support engine 114 may associate the patient's physiological parameters with a patient experiencing the symptom (e.g., as illustrated in block 412) or may associate the patient's physiological parameters with a patient not experiencing the symptom (e.g., as illustrated in block 420).

[0234] Additionally, while the embodiment of FIG. 4A illustrates the initiation of one or more diagnostic tests only after the decision support engine determines that the patient is not experiencing a symptom, in certain other embodiments, one or more diagnostic tests may also be initiated when the decision support engine 114 confirms that the patient is experiencing a symptom, to further increase confidence that the patient is actually experiencing the predicted symptom that the decision support engine 114 has also confirmed.

[0235] 4A illustrates the initiation of one or more diagnostic tests after the decision support engine 114 determines that the patient is not experiencing the symptom, in certain other embodiments, the one or more diagnostic tests may not be used. Instead, in certain embodiments, the decision support engine 114 may determine solely whether the patient is experiencing the predicted symptom, and based on the decision support engine 114's determination, the decision support engine 114 may associate the patient's physiological parameters with the patient experiencing the symptom (e.g., as illustrated in block 412) or may associate the patient's physiological parameters with the patient not experiencing the symptom (e.g., as illustrated in block 420).

[0236] As previously mentioned, a model (e.g., a machine learning model or a rules-based model) utilized and personalized for a patient based on the operations illustrated in workflow 400A can be deployed to provide a prediction of the patient experiencing one or more symptoms and further provide one or more recommendations for treatment based on the prediction. Figure 4B is an example workflow 400B for using a model to determine the likelihood of one or more symptoms commonly associated with potassium imbalance occurring in real time or within a specified time period and providing one or more recommendations for treatment based on the determined likelihood, according to certain embodiments of the present disclosure.

[0237] Actions 402-406, 410, 416, and 418 in FIG. 4B may be similar to actions 402-406, 410, 416, and 418 illustrated in FIG. 4A, however, FIG. 4B further illustrates decision support (e.g., one or more recommendations) being provided to the patient based on the predicted one or more symptoms.

[0238] Although FIG. 4B illustrates that (1) decision support engine 114 provides decision support to the patient after determining whether the patient is experiencing the predicted symptoms in block 406, or (2) after decision support engine 114 determines that the patient is not experiencing the predicted symptoms and diagnostic testing confirms that the patient is in fact experiencing the symptoms, in certain other embodiments, decision support is provided to the patient after (3) determining in block 406 that the patient is experiencing or is likely to experience the symptoms, or (4) receiving feedback from the patient in block 410.

[0239] In certain embodiments, the decision support provided to the patient may tell the patient how to control potassium levels to avoid symptoms before they occur if the decision support engine 114 predicts that the patient is likely to experience symptoms in the near future. In certain embodiments, the decision support provided to the patient may provide one or more recommendations to avoid future symptoms (e.g., consume less potassium, adjust insulin dosage, etc.). In particular, the decision support engine 114 may provide recommendations to increase or decrease the patient's potassium levels, such as to correct a potassium imbalance in the patient that is currently causing the predicted one or more symptoms. Such recommendations may include dietary recommendations, hydration recommendations, lifestyle recommendations, treatment recommendations, service intervention recommendations, insulin administration recommendations, medication recommendations, alarm / alert recommendations, or other recommendations for managing the patient's potassium levels. The decision support engine 114 may output such recommendations for treatment to a user (e.g., through the application 106). Each of these recommendations may be described in more detail with respect to FIG. 5.

[0240] In certain embodiments, the decision support engine 114 may use the patient's predicted one or more symptoms to understand the patient's risk of hospitalization / re-admission. In certain embodiments, the patient's predicted one or more symptoms may be used to better understand the patient's post-discharge stability when determining whether to discharge the patient. In certain embodiments, the decision support engine 114 may use the patient's predicted one or more symptoms to stratify the level of care the patient should receive upon admission. This may be described in more detail with respect to FIG. 5.

[0241] In certain embodiments, the decision support engine 114 may use one or more other machine learning models trained based on patient-specific and / or population data to provide recommendations for the treatment or prevention of conditions associated with potassium imbalance. The algorithms and / or machine learning models may consider one or more inputs 128 and / or metrics 130 (e.g., including potassium levels and / or potassium level rate of change) described with respect to FIG. 3 to determine an optimal recommendation for the patient for the prevention and / or management of such conditions.

[0242] In certain embodiments, as an alternative to using a machine learning model, the decision support engine 114 may use one or more methods to provide recommendations for prevention of symptoms associated with potassium imbalance. The methods may be rules-based and may include one or more rules that are used to provide recommendations regarding management of the patient's potassium levels.

[0243] The one or more models developed and optimized in workflow 400 for a patient may provide insight into the patient's baseline potassium levels and how these baselines change over time for the patient. The one or more models may also provide insight into which potassium level ranges and / or potassium level rate of change ranges (in addition to other physiological parameters) correlate with different symptoms experienced by the patient and how these ranges change over time for the patient. In particular, the patient's personalized potassium thresholds (as well as other physiological parameters) corresponding to various symptoms may be analyzed over time to provide indications regarding the improvement or worsening of the patient's kidneys, as well as other organs and / or conditions / diseases in the patient's body.

[0244] In certain embodiments, user-specific values ​​of physiological parameters for different symptoms experienced by the user may be useful inputs to one or a model designed to assess the user's renal health, including the presence and / or severity of renal disease.

[0245] 5 is a flow diagram illustrating an example method 500 for providing decision support using a continuous analyte sensor including at least a CPM, according to certain exemplary embodiments of the present disclosure. For example, the method 500 may be performed to provide decision support to a user using a continuous analyte monitoring system 104 including at least a CPM 202, as illustrated in FIGS. 1 and 2.

[0246] The method 500 may be executed by the decision support system 100 to collect / generate data, such as inputs 128 and metrics 130 including analyte data, patient information, and non-analyte sensor data described above, to (1) automatically detect and classify abnormal renal conditions, (2) assess the renal health of a user, including the presence and severity of renal disease, hyperkalemia, and / or hypokalemia, (3) risk stratify patients to identify patients at high risk for renal disease, hyperkalemia, and / or hypokalemia, (4) identify risks (e.g., risk of death, risk of significant cardiac events, etc.) associated with a current renal disease diagnosis, (5) make patient-specific treatment decisions or recommendations for renal disease, and (6) provide information regarding the effectiveness of an intervention (e.g., the effectiveness of a patient's lifestyle changes, the effectiveness of a surgical treatment, the effectiveness of a patient taking a new medication, etc.). In other words, the decision support system presented herein may provide information useful in directing and improving care for patients with or at risk for renal disease. The method 500 is described below with reference to FIGS. 1 and 2 and their components.

[0247] At block 502, the method 500 begins by continuously monitoring one or more analytes of a patient, such as the user 102 illustrated in FIG. 1, over a plurality of time periods to obtain analyte data. The one or more analytes monitored may include at least potassium, and thus the analyte data may include potassium data. Block 502 may be performed in certain embodiments by the continuous analyte monitoring system 104 illustrated in FIGS. 1 and 2, and more specifically, the continuous analyte sensor 202 illustrated in FIG. 2. For example, the continuous analyte monitoring system 104 may include a CPM 202 configured to measure the patient's potassium level.

[0248] Similar to workflow 400 of FIG. 4, the primary analyte for measurements described herein is potassium, although in certain embodiments, other analytes may be considered. In particular, combining potassium measurements with additional analyte data may help to further inform analysis related to renal disease diagnosis and staging. For example, monitoring additional types of analytes, such as glucose, creatinine, lactate, BUN, ammonia, C-peptide, and / or cystatin C, measured by continuous analyte monitoring system 104, may be used to correct the potassium levels measured for the patient, provide additional insight into renal disease diagnosis, and / or supplement the information used to determine optimal treatment for preventing disease progression (and in some cases, for disease regression).

[0249] Additional insight gained from using a combination of analytes rather than just potassium may increase the accuracy of renal disease diagnosis prediction. For example, the probability of correctly diagnosing and / or staging renal disease may be a function of the number of analytes measured for a user. More specifically, in some examples, the probability of correctly staging renal disease using only potassium data (in addition to other non-analyte data) may be less than the probability of correctly staging renal disease using potassium and other analyte data for one or more other analytes for analysis.

[0250] In certain embodiments, how other analytes are used to provide a renal disease diagnosis may vary from patient to patient. For example, different models may be created for different groups of patients with common defining characteristics (e.g., a first group having patients who regularly experience high potassium levels, a second group having patients who regularly experience low potassium levels, etc.). The role and weight that each of these other analytes plays in making a renal disease diagnosis may vary for each determined patient group.

[0251] In certain embodiments described herein, for example, for staging renal disease, the combination of analytes measured and collected by one or more sensors (e.g., a multi-analyte sensor) includes potassium and at least one of glucose, creatinine, BUN, ammonia, C-peptide, and / or cystatin C; however, other analyte combinations may be considered to inform the measured potassium levels (e.g., to better understand why a patient's measured potassium level is lower or higher than expected potassium levels for a particular patient), correct the measured potassium levels, and diagnose and stage renal disease.

[0252] In certain embodiments, AI models, such as machine learning models and / or algorithms (e.g., rules-based models), may be used to provide real-time decision support for diagnosing and staging renal disease. In certain embodiments, such models may be configured to use input from one or more sensors measuring multiple analyte data to diagnose, stage, treat, and assess risk of renal disease. Thus, to account for such comorbidity interactions, parameters and / or thresholds of such algorithms and / or models may be altered based at least in part on the number of analytes being measured for input to reflect knowledge gained from each of the prevalences associated with the other / additional analytes being measured.

[0253] In addition to continuously monitoring one or more analytes of the patient during multiple time periods to obtain analyte data in block 502, optionally in certain embodiments, method 500 may also include monitoring other sensor data (e.g., non-analyte data) during multiple time periods using one or more other non-analyte sensors or devices (e.g., non-analyte sensor 206 and / or medical device 208 of FIG. 2 ).

[0254] As previously mentioned, non-analyte inputs may include, but are not limited to, outputs from one or more of an insulin pump, a tactile sensor, an ECG sensor or heart rate monitor, a blood pressure sensor, a sweat sensor, a respiration sensor, a thermometer, a peritoneal dialysis machine, a hemodialysis machine, a sensor or device provided by the display device 107 (e.g., an accelerometer, a camera, a global positioning system (GPS), a heart rate monitor, etc.), or other user accessories (e.g., a smart watch), or any other sensor or device that provides relevant information about the user.

[0255] At block 504, the method 500 continues by processing the analyte data from the multiple time periods to determine at least one rate of change of potassium for the patient. Block 504 may be performed by the decision support engine 114 in certain embodiments. Block 504 in the method 500 of FIG. 5 may be similar to block 404 in the workflow 400 of FIG.

[0256] At block 506, the method 500 continues by generating a disease prediction using the analyte data associated with the one or more analytes and at least one rate of change of the potassium level. Block 506 may, in certain embodiments, be performed by the decision support engine 114 illustrated in FIG.

[0257] Different methods for generating disease predictions may be used by the decision support engine 114. In particular, in certain embodiments, the decision support engine 114 may use rule-based models to provide real-time decision support for diagnosing and staging renal disease. As previously mentioned, rule-based models involve using a set of rules to analyze data. In particular, the decision support engine 114 may apply rule statements (e.g., if, then statements) to assess the presence and severity of renal disease in a patient, perform renal disease risk stratification for the patient, and / or identify risks associated with the patient's current renal disease diagnosis.

[0258] For example, one rule may relate to the patient's current absolute maximum potassium level or may be based on the absolute maximum potassium level change over time. Another rule may relate to the patient's current absolute minimum potassium level or may be based on the absolute minimum potassium level change over time. Another rule may be based on the change in the patient's potassium baseline over time. Another rule may relate to the patient's potassium level rate of change, whether the potassium level rate of change is marked as "rapidly increasing" or "rapidly decreasing" (e.g., as described with respect to FIG. 3) or based on the change in the potassium level rate of change over time. Another rule may relate to an insulin metric, a creatinine metric, a BUN metric, an ammonia metric, and / or a C-peptide metric, as described with respect to FIG. 3, or based on the change in such metrics over time.

[0259] Another rule may relate to whether a patient experiences an acute or sudden increase in creatinine concentration (and what that increase is) due to AKI (e.g., blood loss, vomiting, diarrhea, heart failure, etc.). Another rule may relate to the patient's potassium clearance rate following consumption of a known or estimated amount of potassium. For example, in patients with impaired renal function, the clearance rate may be slower than in healthy control subjects.

[0260] Another rule may relate to the absolute maximum potassium level following some event, such as the consumption of a known or estimated amount of potassium. For example, the absolute maximum potassium level following the consumption of a known or estimated amount of potassium may be greater in a patient with renal impairment than in a healthy control subject. Another rule may relate to the correlation between the patient's absolute maximum potassium level and the patient's potassium clearance rate. For example, an increase in the absolute maximum potassium level with reduced potassium clearance may be observed as renal disease progresses in a patient. One or more other rules may be based on data from one or more non-analyte sensors in combination with the patient's measured potassium level. Any of the rules identified above may be used in combination with each other or with one or more other rules when using a rule-based model.

[0261] Such rules may be maintained in a reference library by the decision support engine 114. For example, the reference library may maintain ranges of potassium levels and / or potassium level change rates that may be mapped to different severities of renal disease. In certain embodiments, such rules may be determined based on empirical studies and analyzing historical patient records from the medical history records database 112.

[0262] In certain embodiments, as an alternative to using rule-based models, AI models, such as machine learning models, may be used to provide real-time decision support for renal disease diagnosis and staging. In certain embodiments, the decision support engine 114 may deploy one or more of these machine learning models to perform renal disease diagnosis, staging, and risk stratification of patients. Risk stratification may refer to the process of assigning health risk states to patients and using the patient's assigned risk states to direct and improve care.

[0263] In particular, the decision support engine 114 may retrieve information from a user profile 118 associated with the patient stored in the user database 110, characterize the information about the patient stored in the user profile 118 into one or more features, and use these features as inputs to such models. Alternatively, the information provided by the user profile 118 may be characterized by another entity, and the features may then be provided to the decision support engine 114 to be used as inputs to the ML model. In certain embodiments, the features associated with the patient may be used as inputs to one or more of the models to assess the patient's renal health, including the presence and severity of renal disease.

[0264] In certain embodiments, features associated with a patient may be used as inputs to one or more of the models for risk stratifying a patient to identify whether the patient is at high or low risk of developing renal disease. In certain embodiments, features associated with a patient may be used as inputs to one or more of the models to identify the risk associated with the patient's current renal disease diagnosis (e.g., risk of death, risk of being diagnosed with one or more other diseases, etc.). In certain embodiments, features associated with a patient may be used as inputs to one or more of the models to perform any combination of the functions described above. Details associated with how one or more machine learning models can be trained to provide real-time decision support for renal disease diagnosis and staging are further discussed in connection with FIG. 6.

[0265] As mentioned, in certain embodiments, other analyte data, in addition to potassium, may be used by the decision support engine 114 to generate a disease prediction for the patient, in block 506. Analyte data including potassium data, lactate data, creatinine data, BUN data, ammonia data, C-peptide data, and / or cystatin C data (e.g., from measurements by the continuous analyte monitoring system 104) may be used as inputs to such machine learning models and / or rule-based models to predict the user's renal health, including the presence and severity of the user's renal disease.

[0266] The decision support engine 114 may use machine learning models and / or rule-based models to generate disease predictions based on continuous analysis of the patient's data (e.g., analyte data, and in some cases, non-analyte data) collected over various time periods. Analysis of data collected for a patient over various time periods may provide insight into whether the patient's health and / or disease is improving or worsening. For example, a patient previously diagnosed with renal disease using the models discussed herein may continue to be constantly monitored (e.g., collected continuously for the patient) to determine whether the disease is worsening or improving, etc. As an example, comparison of potassium levels, potassium baseline, absolute maximum potassium levels, absolute minimum potassium levels, potassium level rate of change (e.g., in response to consumption of a known amount of potassium), potassium baseline rate of change, and / or physiological parameters recorded for different symptoms of a patient over multiple months may indicate the patient's disease progression.

[0267] In certain embodiments, the decision support engine 114 outputs the generated disease prediction to the user at block 506. In some embodiments, the method 500 ends at block 506 by outputting the disease prediction. In other words, in certain embodiments, the operation at block 508 may be skipped and the decision support engine 114 may not provide further decision support regarding recommendations for treatment.

[0268] In other cases, the method 500 continues at block 508 by the decision support engine 114 generating one or more recommendations for treatment based at least in part on the disease prediction generated at block 506. In particular, the decision support engine 114 may provide recommendations for treatment or prevention of renal disease, such as lifestyle recommendations, medication recommendations, service intervention recommendations, or other recommendations for managing renal health. The decision support engine 114 may output such recommendations for treatment to a user (e.g., through the application 106).

[0269] In certain embodiments, the decision support engine 114 may use one or more other machine learning models trained based on patient-specific data and / or population data to provide recommendations for the treatment or prevention of renal disease. The algorithms and / or machine learning models may consider one or more inputs 128 and / or metrics 130 (e.g., including potassium levels and / or potassium level rate of change) described with respect to FIG. 3 to determine an optimal recommendation for the prevention and / or management of the patient's renal disease. In certain embodiments, the models may look at different patterns of analyte measurements collected for a patient to guide the patient in the management of the patient's disease.

[0270] In certain embodiments, as an alternative to using a machine learning model, the decision support engine 114 may use one or more methods to provide a recommendation for the treatment or prevention of renal disease. The methods may be rule-based, such that the methods include one or more rules that are used to provide a recommendation regarding the management of the patient's renal disease.

[0271] Recommendations for the treatment or prevention of kidney disease may, in some cases, be based on the determined optimal balance of the patient's potassium (e.g., intracellular and extracellular) and insulin levels. In particular, in certain embodiments, one or more algorithms may be used to determine the optimal balance of the patient's potassium and insulin levels, which is then used to form one or more recommendations for the patient regarding diet, lifestyle changes, treatments, insulin dosages, and / or medications.

[0272] As mentioned herein, insulin stimulates potassium uptake by cells, thereby reducing serum (e.g., extracellular) potassium levels when a patient does not experience insulin resistance or insulin sensitivity (e.g., when potassium levels are not very responsive to insulin, often caused by chronic hyperinsulinemia). In some cases, as a patient's renal function decreases, the patient may need to resort to excess insulin to move some of the excess extracellular potassium into cells. While this provides an immediate solution in avoiding hyperkalemia and its symptoms, this solution may not be a long-term treatment. In particular, at some point, intracellular potassium may need to be pushed out of the cells so that the kidneys can remove the excess potassium (e.g., through urine). Whether the patient's kidneys can remove the amount of potassium pushed out of the cells (e.g., into the blood) may be an important point. In particular, the rate of potassium movement from within the patient's cells to the extracellular fluid may need to match the rate of potassium clearance by the kidneys to ensure that excessive levels of potassium are not present in the extracellular fluid. In some other cases, excess insulin may move too much potassium into the cells. Therefore, determining a patient's optimal balance of intracellular potassium, extracellular potassium, and insulin based on the patient's renal status can help better inform recommendations made to the patient for managing renal health and / or disease.

[0273] In certain embodiments, dietary recommendations may include recommending to the patient to consume a certain amount of potassium daily, weekly, etc. For example, the patient may be recommended to eat a certain amount of potassium per day based on the level of potassium that the patient's kidneys are currently removing. Monitoring potassium consumption can help ensure that excessive levels of potassium are not pushed into cells (e.g., due to excess insulin), while also ensuring that consumed potassium can be removed by the patient's kidneys.

[0274] In certain embodiments, the dietary recommendations may include recommendations for the patient to increase their potassium consumption. For example, if excess insulin and / or one or more diuretics are causing a significant fall in measured extracellular potassium (e.g., which may lead to hypokalemia), the patient may be recommended to increase potassium consumption.

[0275] In certain embodiments, the hydration recommendation may include a recommendation that the patient increase their fluid intake. For example, the patient may be recommended to increase fluid intake to allow the kidneys to properly process potassium and thus reduce potassium levels in the blood. Lack of proper hydration may cause hyperkalemia as the body cannot effectively process potassium.

[0276] In certain embodiments, lifestyle recommendations (including, for example, exercise recommendations) may include recommendations for the patient to increase physical activity daily, weekly, etc., given that increased physical activity may be one way to remove excess potassium from the body, for example, through sweating. For example, for a patient with hyperkalemia, one or more models may be used to determine when such a patient should engage in physical activity based on potassium levels, renal function, and / or current insulin production / infusion for the patient. The patient may be recommended to engage in a modified physical activity schedule or to engage in additional rest based on the determined schedule around when the patient should engage in physical activity. In certain embodiments, exercise recommendations may be made to optimize sweating while limiting the patient's physical activity to reduce potassium, while preventing potassium from increasing due to exercise activity by the patient. In certain embodiments, other analyte or non-analyte data, such as heart rate or respiration rate data, may be used in combination with potassium data to provide such exercise recommendations.

[0277] In certain embodiments, the treatment recommendation may include a recommendation of dialysis for the patient. In certain embodiments, determining the optimal balance of the patient's potassium and insulin levels may help inform whether dialysis is a recommended treatment for the patient. In particular, dialysis is a treatment for kidney failure that cleanses the body of unwanted toxins, waste products, and excess fluids by filtering the patient's blood. Dialysis helps balance potassium, phosphorus, and sodium levels in the patient's body. Thus, understanding the optimal balance of potassium and insulin in the body may help inform such treatment where dialysis is recommended.

[0278] In certain embodiments, the service intervention recommendation may include a recommendation that the patient seek medical assistance. For example, in certain embodiments, the service intervention recommendation may indicate to the patient, or another individual interested in the patient's health, that the patient needs to go to an emergency room immediately and / or contact the patient's physician. In certain other embodiments, the service intervention recommendation may automatically alert the patient's physician regarding the patient's condition for physician intervention. In certain other embodiments, the service intervention recommendation may alert medical personnel to send assistance to the patient, for example, triggering an ambulance service or emergency medical service to provide the patient with emergency pre-hospital treatment and stabilization and / or transport the patient to definitive care. In certain embodiments, the decision support engine 114 may make the service intervention recommendation based on the patient's ability to seek medical assistance and / or the patient's accessibility to medical relief.

[0279] In certain embodiments, the insulin dosage recommendation may include a recommendation for a combination dosage such as insulin / glucose (e.g., to prevent hypoglycemia but restore eukalemia). In certain embodiments, one or more algorithms may be used to determine the combination dosage recommended for the patient. In particular, insulin may be used to treat hyperkalemia such that an increased dose of insulin may be used to reduce extracellular potassium. In a CKD patient who is diabetic and receiving insulin and has increasing potassium levels, an algorithm may be created and used for administration of insulin to avoid or treat hyperkalemia while also preventing hypoglycemia. For example, the algorithm may calculate the amount of insulin required to reduce potassium levels by a value X (e.g., X is a value greater than zero), as well as the amount of glucose and timing of glucose consumption to prevent hypoglycemia. Additionally, the insulin dosage recommendation may be personalized for each patient to account for differences in insulin resistance across patients. For example, insulin resistance alters the ability of insulin to push potassium into cells, and thus a patient with higher insulin resistance may require a different insulin dosage compared to another patient with a different level of insulin resistance, hi certain aspects, the personalized insulin dosage recommendation may be modified over time as insulin resistance increases or decreases in a patient.

[0280] In certain embodiments, the medication recommendation may include recommending to the patient to take a new medication if the patient has not previously taken a similar medication. In certain embodiments, the medication recommendation may include a recommendation for the patient to stop taking a previously prescribed medication, and in some cases may recommend an alternative medication for consumption by the patient. In certain embodiments, the medication recommendation may include recommending to the patient to take a lower or higher dosage of a previously prescribed medication. In certain embodiments, the medication recommendation may include a recommendation regarding titration of a medication previously prescribed to the patient to determine an ideal dosage for the patient (e.g., while monitoring the user's renal and cardiac health).

[0281] In certain embodiments, the decision support engine 114 may determine that renal disease in a patient is progressing and correlate such progression with medications previously prescribed for the patient. The decision support engine 114 may make this determination based on inputted medical consumption information for the patient (in combination with other factors). In certain embodiments, determining the optimal balance of a patient's potassium and insulin levels, as well as understanding the interactions between potassium levels, insulin, and one or more types of medications for the patient, may help inform which medications (including dosage and frequency) are most appropriate for the patient.

[0282] In certain embodiments, the medical recommendation may include a recommendation for the patient to take a potassium binder. For example, in certain embodiments, the medical recommendation may include a recommendation for the patient to take a potassium binder rectally as an enema. In certain embodiments, the medical recommendation may include a recommendation for the patient to take an oral potassium binder, such as Valtassa.

[0283] In certain embodiments, the medical recommendation may include a recommendation for the patient to take a potassium binder in combination with insulin. For example, in certain embodiments, insulin can be used to reduce serum potassium levels because it stimulates potassium uptake by cells. However, while insulin promotes the movement of extracellular potassium into cells, it does not reduce total body potassium. Therefore, to prevent a subsequent increase in total body potassium levels after taking insulin, a potassium binder may be taken in combination with it to remove potassium.

[0284] In certain embodiments, the medication recommendation may include a recommendation for the patient for a particular type of diuretic and / or dosage. As previously mentioned, medications such as diuretics may be prescribed to a patient for the purpose of treating excess fluid accumulation caused by congestive heart failure (CHF), liver failure, and / or nephritic syndrome. The types of diuretics prescribed may include loop diuretics, thiazides and thiazide-like diuretics, and / or potassium-sparing diuretics. Each of the identified diuretic types may be prescribed to a patient for different purposes. Thus, in certain embodiments, the decision support engine 114 may determine the optimal diuretic for prescription based on the patient's health and the patient's condition being treated. For example, a patient with CHF (but not renal disease) may typically be prescribed a thiazide-like diuretic. In particular, thiazide-like diuretics may help clear out excess fluid caused by CHF, however, such diuretics may dehydrate the patient in some cases. Dehydration in a patient with impaired kidneys may further damage the patient's kidneys. In particular, dehydration can clog the kidneys with muscle proteins (myoglobin). Thus, if a patient is experiencing only CHF and not renal disease, prescribing the patient a thiazide-like diuretic may not cause significant harm to the patient's kidneys. However, if the patient is experiencing renal failure, such a diuretic may not be optimal to prescribe. Therefore, another diuretic type may be considered.

[0285] In certain embodiments, the medical recommendation may include a recommendation for the patient to take a diuretic in combination with taking insulin. Also, insulin promotes the movement of extracellular potassium into cells but does not reduce total body potassium. Thus, to prevent an increase in total body potassium levels after taking insulin, a diuretic may be taken in combination therewith.

[0286] Additionally, in certain embodiments, the decision support engine 114 may determine the optimal diuretic for prescription by also considering the side effects that the prescribed diuretic may have on other organs of the patient. By considering the effects of different medications on other organs, the decision support engine 114 may assist the patient in managing the condition of other organs in the patient's body. In certain embodiments, the CPM 202 may assist in making this determination, at least with respect to the patient's kidneys. For example, the CPM 202 may be used to monitor the effect of a medication prescribed to the patient for treating CHF to better determine its side effects on the patient's renal function. If potassium levels are observed to decrease over a period of time after the patient has been prescribed a diuretic for CHF, it may be assumed that the patient is experiencing dehydration, and further concluded that such dehydration may adversely affect the patient's kidneys. Thus, a new diuretic may be considered for prescription.

[0287] In certain embodiments, the alarm / alert recommendations may include recommendations for adding new types of alarms / alerts, removing existing alarms / alerts, increasing or decreasing the frequency of existing alarms / alerts, and / or changing existing threshold levels for existing alarms / alerts configured for devices used by the patient. As mentioned, in certain embodiments, the types of alarms / alerts customized for each particular display device, the number of alarms / alerts customized for each particular display device, the timing of the alarms / alerts customized for each particular display device, and / or the threshold levels configured for each of the alarms (e.g., for triggering) are based on the patient's current health, the patient's renal status, the current treatments recommended for the patient, the patient's physiological parameters when experiencing different symptoms stored in the user profile 118 for each patient, and / or in some cases, the disease predictions generated in block 506.

[0288] In certain embodiments, the disease prediction generated in block 506 may be used to stratify the patient's risk of hospitalization / readmission. In certain embodiments, the disease prediction generated in block 506 may be used to better understand the patient's post-discharge stability when determining whether to discharge the patient. In some embodiments, the disease prediction generated in block 506 may be used to monitor the patient's post-discharge stability and determine if and when the patient should be readmitted. In certain embodiments, the disease prediction generated in block 506 may be used to stratify the level of care the patient should receive upon admission. For example, a patient admitted to a hospital may be one of many patients at the hospital. Thus, in certain aspects, the disease prediction generated for the patient may be compared to similarly generated disease predictions of other patients in the hospital to better inform medical personnel in the hospital of where the patient ranks among other patients in terms of the level of care required, as well as the urgency of attention required by the patient among other patients. This may be particularly important when patients are at high risk of experiencing an acute life-threatening event within a short time after being admitted to hospital (as opposed to, for example, waiting four or more hours before receiving assistance or care from medical personnel).

[0289] In certain embodiments, the machine learning models deployed by the decision support engine 114 include one or more models trained by the training server system 140, as illustrated in Figure 1. Figure 6 describes in further detail a technique for training the machine learning models deployed by the decision support engine 114 to generate predictions associated with renal health, according to certain embodiments of the present disclosure.

[0290] In certain embodiments, the method 600 is used to train a model to generate predictions associated with renal health. The predictions associated with renal disease may include (1) a prediction of one or more symptoms associated with renal disease, (2) a prediction regarding a patient's renal health, including the presence and / or severity of renal disease (e.g., a user as illustrated in FIG. 1 ), and / or (3) a prediction regarding an optimal treatment for the patient.

[0291] The method 600 begins at block 602 with retrieving data from a historical medical record database, such as the historical medical record database 112 illustrated in Figure 1, by a training server system, such as the training server system 140 illustrated in Figure 1. As referred to herein, the historical medical record database 112 may provide a repository of up-to-date and historical medical information for users of the continuous analyte monitoring system and connected mobile health application, such as users of the continuous analyte monitoring system 104 and application 106 illustrated in Figure 1, as well as data for one or more patients who are not or were not previously users of the continuous analyte monitoring system 104 and / or application 106. In certain embodiments, the historical medical record database 112 may include one or more data sets of historical patients who do not have renal disease or who have various stages of renal disease.

[0292] Retrieval of data from the historical record database 112 by the training server system 140, in block 602, may include retrieval of all or any subset of the information maintained by the historical record database 112. For example, if the historical record database 112 stores information about 100,000 patients (e.g., non-users and users of the continuous analyte monitoring system 104 and application 106), the data retrieved by the training server system 140 to train one or more machine learning models may include information about all 100,000 patients, or only a subset of the data for those patients, e.g., data associated with only 50,000 patients, or only data from the past 10 years.

[0293] As an illustrative example, integration with home-grown or cloud-based medical record databases through Fast Healthcare Interoperability Resources (FHIR), web application programming interfaces (APIs), Health Level 7 (HL7), and / or other computer interface languages ​​may enable aggregation of historical medical records for baseline assessment in addition to aggregation of de-identified patient data from cloud-based repositories.

[0294] As an illustrative example, in block 602, the training server system 140 may retrieve information about 100,000 patients with various stages of renal disease stored in the medical history record database 112 to train a model to predict the presence and / or severity of renal disease in a user. Each of the 100,000 patients may have a corresponding data record (e.g., based on their corresponding user profile) stored in the medical history record database 112. Each user profile 118 may include information such as the information discussed with respect to FIG.

[0295] The training server system 140 then uses the information in each of the records to train an artificial intelligence or ML model (for simplicity, referred to herein as an "ML model"). Examples of the types of information contained in a patient's user profile are provided above. The information in each of these records may be characterized (e.g., manually or by the training server system 140) to yield features that may be used as input features to train the ML model. For example, the patient records may include or be used to generate features related to the patient's age, the patient's gender, the patient's occupation, the patient's analyte levels over time, the patient's analyte level change rates and / or trends over time, physiological parameters associated with the patient's different renal disease stages over time, etc. The features used to train the machine learning model may vary in different embodiments.

[0296] In certain embodiments, each historical patient record retrieved from the historical record database 112 is further associated with a label indicating whether the patient was healthy or experienced any fluctuations in renal disease, the patient's previously determined renal disease diagnosis and / or stage of renal disease, renal disease risk assessment, treatment, and / or similar metrics. What the record is labeled depends on what the model is trained to predict.

[0297] At block 604, the method 600 continues by the training server system 140 training one or more machine learning models based on the selected historical patient records. In some embodiments, the training server does so by providing the features (e.g., extracted at block 606) as input to the model. The model may be a new model initialized with random weights and parameters, or may be partially or fully pre-trained (e.g., based on a previous training round). Based on the input features, the model in training generates some output. In certain embodiments, the output may indicate whether the patient was healthy or experienced some variation in renal disease, a renal disease diagnosis and / or renal disease stage for the patient, a risk assessment, previously recommended treatment, or similar metrics. In certain embodiments, the output may indicate whether the patient has experienced one or more symptoms commonly associated with renal disease, or the level of risk that the patient will experience one or more symptoms commonly associated with renal disease.

[0298] In certain embodiments, the training server system 140 compares this generated output to the actual labels associated with the historical patient record and calculates a loss based on the difference between the actual outcome and the generated outcome. This loss is then used to refine (e.g., via backpropagation) one or more internal weights and parameters of the model so that the model learns to more accurately predict the user's renal health, including the presence and / or severity of renal disease (or its recommended treatment).

[0299] One of a variety of machine learning algorithms may be used to train the models described above, such as a supervised learning algorithm, a neural network algorithm, a deep neural network algorithm, etc.

[0300] At block 606, the training server system 140 deploys the trained model to make predictions associated with renal health during runtime. In some embodiments, this includes transmitting some instructions of the trained model (e.g., a weight vector) that can be used to instantiate the model on another device. For example, the training server system 140 may transmit the weights of the trained model to the decision support engine 114. The model can then be used to assess the presence and / or severity of a user's renal disease in real time using the application 106, provide treatment recommendations, and / or make other types of predictions discussed above. In certain embodiments, the training server system 140 may continue to train the model in an "online" manner by using input features and labels associated with new patient records.

[0301] Additionally, the same method for training illustrated in FIG. 6 using historical patient records may also be used to train a model using patient-specific records to create a more personalized model for making predictions associated with renal health. For example, a model trained using historical patient records deployed for a particular user may be further retrained after deployment. For example, a model may be retrained after it has been deployed for a particular patient to create a more personalized model for the patient. The more personalized model may be able to more accurately predict a patient's symptoms based on the patient's own data (as opposed to only historical patient record data), including the patient's own potassium thresholds.

[0302] Exemplary methods and systems for providing kidney health and / or diabetes decision support As noted, in certain embodiments, the decision support systems described herein are designed to provide decision support in the form of risk assessment and treatment for renal disease and / or glucose homeostasis. For example, the methods may be used to provide risk stratification and treatment recommendations for the treatment and / or prevention of renal disease, hyperkalemia, hypokalemia, hyperglycemia, and / or hypoglycemia.

[0303] In a particular embodiment, a method is a decision tree, which is a tree-like collection of (1) attribute nodes linked to two or more subtrees and (2) decision nodes (also commonly referred to as leaf nodes). The attribute nodes compute some outcome based on the attribute values ​​of the case (e.g., in this case, the user) being evaluated, and each possible outcome is associated with one of the subtrees. A decision node is a node that does not split into further nodes and provides a decision based on attributes previously evaluated for the case. Cases may be classified by starting at the root node of the method and working down / through the method until a decision node is encountered.

[0304] For example, the methods described herein may be used to classify a user, and more specifically, to classify the user as (1) a patient at risk for hyperkalemia, (2) a patient at risk for hypokalemia, or (3) a patient at neither risk for hyperkalemia nor hypokalemia, and further provide preventative and / or therapeutic recommendations based on the initial classification. Thus, the attribute nodes of the methods may include evaluations of the user's continuously measured potassium and / or glucose levels (e.g., attribute values) relative to threshold potassium and / or glucose thresholds, respectively. Additionally, in some cases, the attribute nodes of the methods may include evaluations of user medications and / or user activities related to insulin administration, dialysis treatment, potassium consumption, glucose consumption, and the like. The decision-making nodes of the methods may include recommended actions to correct and / or maintain potassium homeostasis, and additionally, in some cases, may include recommended actions to correct and / or maintain glucose homeostasis. Exemplary recommended actions may include recommendations for potassium intake, glucose intake, fluid intake, insulin administration, dialysis, potassium binder intake, and / or the like.

[0305] In certain embodiments, the decision-aided risk stratification and treatment recommendation may be based on a comparison of the user's potassium level to different potassium thresholds. For example, the decision-aided risk stratification and treatment recommendation may be based on a comparison of the user's potassium level to one or more different potassium thresholds. The different potassium thresholds utilized in the method include: (1) a strongly elevated potassium threshold, (2) a moderately elevated potassium threshold, (3) an elevated potassium threshold, (4) a depressed potassium threshold, and (5) a strongly depressed potassium threshold.

[0306] In certain embodiments, the decision support risk stratification and treatment recommendations may be based on a comparison of the user's glucose level to one or more different glucose thresholds. For example, the different glucose thresholds utilized in the method include: (1) an elevated glucose threshold, and (2) a lowered glucose threshold.

[0307] In certain embodiments, one or more of these potassium and / or glucose thresholds are predefined for all users (e.g., based on historical data). In certain embodiments, one or more of these potassium and / or glucose thresholds are predefined for different populations of similarly situated users (e.g., based on historical data for similarly situated users). In certain embodiments, one or more of these thresholds are personalized for different users through manual threshold setting by the user or a healthcare professional (HCP).

[0308] In certain embodiments, one or more of these thresholds may be personalized for different users based on the user's potassium baseline and / or glucose baseline, respectively. For example, as mentioned herein, each user may have a different potassium baseline. In certain embodiments, a user's potassium baseline may be determined by calculating an average of the user's potassium levels over a certain amount of time during which significant fluctuations (e.g., rapid increases or decreases in potassium levels) are not expected. For example, a user's baseline potassium may be determined over periods during which the user is asleep, sitting in a chair, or other periods during which the user is sedentary and has not consumed foods and / or medications (e.g., insulin administration, medications known to affect potassium levels, etc.) that would reduce or increase potassium levels (e.g., no external conditions are present that would affect the potassium baseline). However, a particular personalized potassium threshold may constantly change for a user. Thus, in certain embodiments, the DAM 116 may continuously calculate the potassium baseline, timestamp the calculated potassium baseline, and store the corresponding information in the user profile 118. Additionally, in certain embodiments, potassium thresholds (1) for a population of users or (2) personalized for a particular user may be altered based at least in part on the number of analytes being measured for an input to reflect knowledge gained from each of the other analytes being measured. Similar methods may be used to calculate, determine, and / or store a user's glucose baseline.

[0309] Since the potassium baseline and / or glucose baseline may vary from user to user, the potassium and / or glucose thresholds set for the user's potassium and / or glucose baseline, respectively, may also vary from user to user. In certain embodiments, individual thresholds may be set based on a set difference above or below the baseline. For example, a first user with significant renal impairment may have a higher potassium baseline, and a potassium threshold specific to the first user may be set based on a difference of 0.5 mmol / L compared to the first user's baseline. Thus, if the first user's potassium baseline is 4.5 mmol / L, the elevated potassium threshold specific to the first user may be set to 5.0 mmol / L, and in some examples, the lowered potassium threshold specific to the first user may also be set to 4.0 mmol / L. Meanwhile, a second user with healthy renal function may have a lower potassium baseline than the first user, for example, a potassium baseline of 3.8 mmol / L. Thus, the second user-specific elevated potassium threshold may be set to 4.3 mmol / L and / or the second user-specific lowered potassium threshold may be set to 3.3 mmol / L (e.g., if a difference of 0.5 mmol / L is used).

[0310] In certain examples, user-specific elevated and lowered analyte thresholds (e.g., potassium and / or glucose) may be determined separately, with only the elevated or lowered analyte thresholds being set based on a difference from the user's analyte baseline. In certain examples, the user-specific elevated analyte threshold may be adjusted by an absolute value (e.g., absolute set difference) that is different (e.g., greater or smaller) than the absolute value of the adjustment of the user-specific lowered analyte threshold based on the difference from the user's analyte baseline. In further embodiments, analyte thresholds, such as potassium and / or glucose thresholds, may instead be set based on population-based guidelines and may not be modified or set based on the user's corresponding analyte baseline.

[0311] In certain embodiments of analyte thresholds, the determination of elevated and decreased analyte levels may be made based on a potassium range for a healthy user. For example, a potassium range of 3.0 mmol / L to 5.5 mmol / L may represent a potassium range for healthy users based on population data. In this embodiment, a user with a potassium baseline of 4 mmol / L would only be notified of an elevated potassium threshold if the user's baseline potassium is above 5.5 mmol / L. Conversely, if the user's baseline potassium falls below 3.0 mmol / L, the user would be notified of a decreased potassium level.

[0312] As described below with respect to Figures 7-13, the methods described herein may be divided into five groups: (1) risk stratification, (2) hyperkalemia decision aid, (3) hypokalemia decision aid, (4) glucose / exogenous insulin decision aid, and (5) dietary decision aid. For example, the method may first begin (e.g., at a root node) by stratifying users into different risk categories based on their risk of hyperkalemia or hypokalemia. Thus, the method may be divided into two sections, including (1) a first section for users determined to be at risk for hyperkalemia, and (2) a second section for users determined to be at risk for hypokalemia. Users determined to be at risk for hyperkalemia may proceed to the first section of the method, which provides treatment recommendations for users at risk for hyperkalemia. In other words, the recommendations may include dietary recommendations, medication recommendations, and / or recommendations to take one or more actions that take into account the user's risk of hyperkalemia. Alternatively, a user determined to be at risk for hypokalemia may proceed to a second section of the method, which provides treatment recommendations for users at risk for hypokalemia. In other words, the recommendations may include dietary recommendations, medication recommendations, and / or recommendations to take one or more actions that take into account the user's risk of hypokalemia.

[0313] 7-13 collectively illustrate an exemplary method used to provide such risk stratification and treatment (or prevention) recommendations to a user. More specifically, FIG. 7 is an exemplary workflow for stratifying users into different risk categories based on risk of hyperkalemia or hypokalemia, while FIGs. 8A, 8B, 9, and 10 constitute a first section of the method (e.g., for users determined to be at risk for hyperkalemia), and FIGs. 11, 12, and 13 constitute a second section of the method (e.g., for users determined to be at risk for hypokalemia). In certain embodiments, the decision support system 100 illustrated in Figure 1 may be configured to collect data, such as inputs 128 and metrics 130, including, for example, analyte data (e.g., from a continuous potassium monitor (CPM) and / or a continuous glucose monitor (CGM)) and the patient information described above, use this information as input for the methods illustrated in Figures 7-13 to (1) classify users into different risk categories based on their risk of hyperkalemia or hypokalemia, and (2) provide one or more recommendations for treatment (or prevention) for the user based on the classification. More specifically, the decision support engine 114 of the decision support system 100 may be configured to perform such (1) classification and (2) provide one or more recommendations. In some cases, the decision support engine 114, after making one or more recommendations, may provide the output recommendations to the user (e.g., through application 106 illustrated in FIG. 1), to the user's caregiver (e.g., a parent, relative, guardian, teacher, nurse, etc.), to the user's physician, or in some cases, to any other individual interested in the user's health with a view to improving the user's health, such as by accomplishing the recommended treatment and / or preventative action. FIGS. 7-13 are described in more detail below.

[0314] 7 is an exemplary workflow 700 for stratifying a user's risk of disease commonly associated with renal disease, including hyperkalemia and hypokalemia, according to certain embodiments of the present disclosure. Workflow 700 is described below with reference to FIGS. 1 and 2 and their components. For example, workflow 700 may be executed by components of decision support system 100 illustrated in FIG. 1 to continuously monitor one or more analytes of a user, such as user 102 illustrated in FIG. 1, over multiple time periods and stratify the user into different risk categories based on the continuously monitored analytes. In certain embodiments, decision support system 100 is configured to continuously monitor a user's potassium level and use workflow 700 to stratify the user's risk of hyperkalemia and / or hypokalemia based on the patient's potassium level.

[0315] The workflow 700 begins at block 702 by the decision support system 100, and more specifically, the decision support engine 114 of the decision support system 100, which determines whether a user is experiencing signs and / or symptoms of potassium imbalance. As previously described, signs are objective observable phenomena that can be identified by another person or device, whereas symptoms are the subjective experience of the patient.

[0316] Signs and symptoms of potassium imbalance may be determined through a number of means, including the use of one or more analyte sensors (e.g., CPM) and one or more non-analyte sensors (e.g., heart rate monitor, blood pressure monitor, respiratory monitor, ECG, accelerometer, temperature sensor, etc.). For example, in certain embodiments, the decision support engine 114 determines whether the user is experiencing (or will experience) one or more symptoms of potassium imbalance based on analyte data from one or more analyte sensors and / or non-analyte data from one or more non-analyte sensors according to workflow 400A of FIG. 4A and / or workflow 400B of FIG. 4B.

[0317] In certain embodiments, the decision support engine 114 determines whether the user is experiencing (or will experience) one or more symptoms and / or is showing one or more signs of potassium imbalance based on the user-entered symptoms and / or signs. For example, the user may manually input into the decision support system 100 that the user is experiencing one or more signs or symptoms.

[0318] In certain embodiments, the decision support engine 114 determines whether the user is experiencing (or will experience) one or more symptoms of potassium imbalance and / or is exhibiting one or more signs based on user input identifying such symptoms and / or signs. For example, the decision support system 100 may identify one or more signs and / or symptoms (e.g., predicted in some cases according to workflows 400A and 400B of FIG. 4A and FIG. 4B, respectively). In some cases, the decision support system 100 prompts the user based on analyte data (e.g., elevated or increasing potassium levels, decreased or decreasing potassium levels, etc.) and / or non-analyte data (e.g., increased respiration, ECG changes, blood pressure changes). In other cases, the decision support system 100 may identify one or more signs and / or symptoms based on contextual data (e.g., camera), other input data (e.g., sensor data), or observations and / or inputs from a person other than the patient (e.g., a health care provider).

[0319] In some cases, the decision support engine 114 may notify the user of a potassium imbalance based on abnormal analyte data (e.g., potassium data) collected from one or more analyte sensors. In some cases, the decision support engine 114 may determine signs and / or symptoms of potassium imbalance. Signs and / or symptoms of potassium imbalance may include muscle weakness or paralysis, nausea, tingling, or cardiac conduction abnormalities, and / or cardiac arrhythmias including sinus bradycardia, sinus arrest, slow idioventricular rhythm, ventricular tachycardia, ventricular fibrillation, and / or asystole. Additionally, electrocardiogram (ECG) abnormalities may be indicative of hyperkalemia. For example, a tall peaked T wave with a shortened QT interval followed by a progressive lengthening of the PR interval (e.g., the PR interval is the time from the onset of the P wave (atrial depolarization) to the onset of the QRS complex (ventricular depolarization)) and QRS duration may be indicative of hyperkalemia. In addition, the P waves may disappear and the QRS may widen into a sinusoid. One or more of these signs and / or symptoms of hyperkalemia and / or hypokalemia (and / or other signs and symptoms not listed) may indicate a serious condition requiring immediate treatment.

[0320] Thus, if, in block 702, the decision support engine 114 determines that the user is experiencing signs and / or symptoms of potassium imbalance, in block 704, the decision support engine 114 provides a recommendation for the user to go to an emergency room. In other words, emergency treatment beyond available home treatments is recommended by the decision support engine 114.

[0321] Any of the above signs or symptoms may indicate a severe risk of hyperkalemia and / or hypokalemia, or other serious medical conditions that may not be resolved with home treatments readily available to the user. Thus, due to the significant risks associated with the signs and / or symptoms of potassium imbalance, a recommendation to a user having one or more signs and / or symptoms is to go to an emergency room. Workflow 700 may end after providing this recommendation if emergency treatment beyond available home treatments is recommended.

[0322] Although potassium levels may indicate signs and / or symptoms of potassium imbalance, in certain cases, a user may experience the signs and / or symptoms of potassium imbalance described above even when the user's potassium levels do not indicate a severe risk. Thus, as shown in FIG 7, the decision support engine 114 may make a recommendation for the user to seek immediate treatment regardless of the user's potassium level (and in some cases, comparison to one or more potassium thresholds personalized for the user).

[0323] Alternatively, if in block 702 the decision support engine 114 determines that the user is not experiencing signs and / or symptoms of potassium imbalance, then in block 706 the decision support engine 114 determines different potassium thresholds to use to evaluate the user's continuously monitored potassium levels (e.g., for risk stratification). The different potassium thresholds determined by the decision support engine 114 may include: (1) a strongly elevated potassium threshold, (2) a moderately elevated potassium threshold, (3) an elevated potassium threshold, (4) a depressed potassium threshold, and / or (5) a strongly depressed potassium threshold.

[0324] In certain embodiments, the different potassium thresholds represent potassium level values ​​for the patient (e.g., 3.0 mmol / L, 5.5 mmol / L, etc.). In certain embodiments, the different potassium thresholds represent potassium level rate of change. As described herein, potassium level rate of change refers to the rate at which one or more time-stamped potassium measurements or values ​​change relative to one or more other time-stamped potassium measurements or values. Potassium level rate of change may be determined over one second or more, one minute or more, one hour or more, one day or more, etc.

[0325] According to FIG. 7, the decision support engine 114 determines one or more of these thresholds based, at least in part, on (1) whether the user has significant renal dysfunction or tissue destruction and / or (2) whether the user is currently exercising (e.g., currently engaged in physical exercise) or has just completed exercise (e.g., recently in time).

[0326] In particular, various conditions can result in elevated potassium levels, including both chronically elevated and acutely elevated serum potassium levels. Some conditions that result in acutely elevated serum potassium can lead to a rapid increase in potassium levels and severe hyperkalemia. For example, ongoing tissue destruction, ongoing potassium absorption, and / or significant non-anion gap metabolic or respiratory acidosis can result in a large and / or rapid release of potassium into the serum. Furthermore, some conditions that result in chronically elevated serum potassium levels can also mask changes in the user's potassium levels. For example, CKD, especially the later stages of CKD, can result in chronically elevated serum potassium.

[0327] In certain embodiments, thresholds relied upon by the decision support engine 114 are adjusted (e.g., increased) to account for circumstances in which these conditions elevate a user's potassium level. For example, thresholds for decision support for users with CKD may be increased to account for the impact of CKD on potassium levels. Thus, in certain embodiments, the (1) strongly elevated potassium threshold, (2) moderately elevated potassium threshold, (3) elevated potassium threshold, (4) depressed potassium threshold, and / or (5) strongly depressed potassium threshold for users with CKD may be higher than the (1) strongly elevated potassium threshold, (2) moderately elevated potassium threshold, (3) elevated potassium threshold, (4) depressed potassium threshold, and / or (5) strongly depressed potassium threshold for users without CKD.

[0328] In certain embodiments, the adjustment to the potassium threshold comprises adjusting the potassium threshold to a particular threshold. In certain embodiments, the predetermined threshold is associated with a different condition. For example, an elevated potassium threshold associated with CKD may be 5.8 mmol / L. Thus, a user with CKD may have an adjusted potassium threshold of 5.8 mmol / L.

[0329] In certain embodiments, adjustments to potassium thresholds include adjusting such thresholds relative to the user's potassium baseline. In certain embodiments, certain adjustments (or changes) are associated with different conditions. For example, an adjustment of +0.5 mmol / L may be associated with CKD. Thus, a user with CKD may have an adjusted potassium threshold equal to their potassium baseline +0.5 mmol / L.

[0330] In certain embodiments, adjustments to potassium thresholds include adjusting such thresholds for a user's potassium change rate. For example, such thresholds may include different potassium change rates. These different potassium change rates may be adjusted by a common / single absolute value (e.g., rate of change value) or by different absolute values. In certain embodiments, the predetermined adjustments, rate of change values, are associated with different conditions.

[0331] Further, in certain embodiments, the thresholds relied upon by the decision support engine 114 are adjusted (e.g., increased) based on information about the exercise performed by the user. In particular, the user's serum potassium levels may be affected by exercise. Specifically, potassium is released from inside the cells (i.e., intracellularly) into the serum. Potassium levels may increase up to 8 mmol / L during exercise and normalize within minutes after exercise has stopped. Because potassium levels may rise significantly and / or rise rapidly during exercise, while dropping significantly after exercise has stopped, the thresholds relied upon by the decision support engine 114 may need to be adjusted for the period of exercise and / or the period after exercise has stopped. For example, the thresholds during exercise may be increased to account for the effect of exercise on potassium levels. In another example, the decreasing rate of change thresholds may be decreased after exercise to account for the effect of exercise on potassium levels after exercise has stopped.

[0332] In certain embodiments, the potassium threshold of an exercising user (e.g., referred to herein as the "exercise potassium threshold") is standardized based on historical data or data from similarly situated users. The historical data may include potassium levels, rate of change, and / or trends from previous exercise sessions. The exercise potassium threshold may be personalized for each patient through user and / or HCP input thresholds. The exercise potassium threshold may further be personalized based on each user's potassium baseline. For example, the exercise potassium threshold may be set relative to the patient's baseline potassium level and may be set as a ratio to the patient's baseline potassium level. The exercise potassium threshold may be adjusted based on the user's potassium levels, rate of change, and / or trends pre-exercise, during exercise, and post-exercise.

[0333] In certain embodiments, the decision support engine 114 uses exercise regime metrics of metrics 130 stored in the user profile 118 to determine whether and / or when the user is exercising, as illustrated in FIG. 1. As described herein, the exercise regime metrics may indicate one or more of when the user engages in what types of activities, the corresponding intensity of such activities, the frequency with which the user engages in such activities, etc. The exercise regime metrics may be calculated based on one or more of analyte sensor data inputs (e.g., from lactate monitors, glucose monitors, etc.), non-analyte sensor data inputs (e.g., non-analyte sensor data inputs from accelerometers, heart rate monitors, respiration rate sensors, etc.), calendar inputs, user inputs, etc.

[0334] In certain embodiments, the user input may include user-entered exercise information and / or confirmation of exercise. For example, in some cases, the user may provide input to the decision support system that the user is exercising. In some other cases, the decision support system 100 may prompt the user and ask for confirmation of exercise. The decision support system 100 may prompt the user based on analyte data (e.g., elevated or increasing potassium levels), other analytes (e.g., lactate data, glucose data, etc.), and / or non-analyte data (e.g., increased respiration, increased HR, accelerometer).

[0335] Thus, the potassium threshold for a user used in the method may be based, at least in part, on (1) whether the user has significant renal impairment or tissue destruction, and / or (2) whether the user is currently exercising (e.g., currently engaged in physical exercise) or has just completed exercise (e.g., recently in time). For example, decision support for a user with significant impairment is based on a threshold adjusted based on conditions that result in elevated potassium levels in the user (e.g., referred to herein as an "adjusted elevated significant impairment threshold"). In another example, decision support for an exercising user is based on a threshold adjusted based on the user currently or recently engaged in a level of exercise (e.g., referred to herein as an "elevated exercise threshold"). Thus, the (1) strongly elevated potassium threshold, (2) moderately elevated potassium threshold, and / or (3) elevated potassium threshold for a patient may be adjusted based on changes in the user's profile, some of which may include changes in potassium baseline, renal function / health, diabetes / glucose homeostasis, medications, treatments, other comorbidities (e.g., hypertension, liver disease, cardiac disease, metabolic disease, etc.), diet, exercise, adverse events (e.g., hyperkalemia / hypokalemia, hyperglycemia / hypoglycemia, cardiac events, etc.), signs of hyperkalemia / hypokalemia, and / or symptoms of hyperkalemia / hypokalemia.

[0336] 7, in block 714, the decision support engine 114 determines whether the user has significant renal impairment and / or tissue destruction. If the decision support engine 114 determines in block 714 that the user has significant renal impairment and / or tissue destruction, in block 716, the decision support engine 114 further determines whether the user is exercising (or has recently exercised).

[0337] Exercise-induced changes (e.g., increases) in potassium levels can be dramatic in rate and magnitude. Typically, for users with healthy renal function, exercise-induced changes in potassium levels are transient in nature. However, for users with impaired renal function, such exercise-induced potassium levels may persist, which may lead to further complications such as hyperkalemia and arrhythmias. Thus, identification of exercise may be important in determining a user's risk of hyperkalemia. In certain embodiments, the determination of whether a user is exercising (or has recently exercised) is based on the rate of change in the user's potassium levels, the duration of elevated potassium levels, and / or the variability (or stability) of potassium levels. In certain embodiments, in block 716, the decision support engine 114 may further determine whether the user has good control of exercise-induced potassium levels and thus is at low risk for prolonged exercise-induced potassium levels, or whether the user has reduced control of exercise-induced potassium levels and thus is at high risk for prolonged exercise-induced potassium levels. Such a determination may be based on the user's own historical data, including the duration of exercise-induced elevated potassium levels and / or exercise-induced variability in potassium levels. The determination at block 716 may then be utilized to determine the user's risk of hyperkalemia, for example, in FIG. 8A.

[0338] In cases where the decision support engine 114 determines in block 716 that the user is exercising (or has recently exercised), the decision support engine 114 adjusts the user's potassium threshold based on both significant impairment and exercise in block 720. In other cases where the decision support engine 114 determines in block 716 that the user is not exercising (or has not recently exercised), the decision support engine 114 adjusts the user's potassium threshold based on significant impairment (and not based on exercise) in block 722 (e.g., using the adjusted significant impairment elevation threshold for the remainder of the method).

[0339] In the event that the decision support engine 114 determines in block 714 that the user does not have significant renal dysfunction and / or tissue destruction, the decision support engine 114 further determines whether the user is (or has recently been) exercising in block 718. In certain embodiments, the determination of whether the user is (or has recently been) exercising is based on the rate of change in the user's potassium level, the duration of elevated potassium levels, and / or the variability (or stability) of the potassium level.

[0340] In cases where the decision support engine 114 determines in block 718 that the user is exercising (or has recently exercised), the decision support engine 114 adjusts the user's potassium threshold based on exercise (and not based on significant impairment) (e.g., uses the elevated exercise threshold for the remainder of the method) in block 724. In other cases where the decision support engine 114 determines in block 718 that the user is not exercising (or has not recently exercised), the decision support engine 114 does not adjust the user's potassium threshold based on either significant impairment and / or exercise in block 726.

[0341] As an illustrative example, following block 706, the decision support engine 114 may rely on (1) a strongly elevated potassium threshold equal to 6.5 mmol / L, (2) a moderately elevated potassium threshold equal to 5.5 mmol / L, (3) an elevated potassium threshold equal to 5.0 mmol / L, (4) a reduced potassium threshold equal to 3.5 mmol / L, and (6) a strongly reduced potassium threshold equal to 2.5 mmol / L to classify the user, and more specifically, as a user at risk for hyperkalemia, (2) a patient at risk for hypokalemia, or (3) a patient at neither risk for hyperkalemia nor hypokalemia, and provide decision support, and to further provide prevention and / or treatment recommendations based on the initial classification. In other words, the decision support engine 114 may use these thresholds throughout the remainder of the method to classify the user and provide one or more recommendations.

[0342] 7, at block 708, the decision support engine 114 determines whether the user's potassium level is above the elevated potassium threshold (e.g., >5.0 mmol / L) (e.g., as determined at block 706) or whether the user's potassium level will be above the elevated potassium threshold within a short period of time (e.g., ≦1 hour). The short period of time may be predefined. In certain embodiments, the elevated potassium level may be determined by the user's potassium level reaching and / or exceeding the elevated potassium threshold. In certain embodiments, the elevated potassium level may be determined by the user's potassium rate of change and / or trends indicating that the user is likely to reach and / or exceed the elevated potassium threshold within a defined short period of time.

[0343] In some cases where the decision support engine 114 determines in block 708 that the user's potassium level is above an elevated potassium threshold (e.g., the user's potassium level >5.0 mmol / L) or that the user's potassium level will be above the elevated potassium threshold within a short period of time, the decision support engine 114 determines that the user is at risk for hyperkalemia. Thus, as shown in Figure 7, the method proceeds to Figures 8A and 8B (e.g., starting with Figure 8A). Figures 8A and 8B illustrate an exemplary 800 method used to provide renal disease-related decision support to a user determined to be at risk for hyperkalemia.

[0344] Although an elevated potassium level may not require immediate treatment, the user's potassium level may need to be quickly reduced to a healthy level. Additionally, if the user's potassium level is not currently elevated but may be elevated soon (e.g., within a defined short period of time), the decision support system 100 may provide a recommendation to the user to initiate potassium-lowering medication to prevent the elevated potassium level or to reduce the rate at which the potassium level is rising. Reducing the potassium level or its rapid rate of change may help reduce instances of hyperkalemia, reduce instances of significant / urgent intervention, aid in renal disease management, and / or increase the effectiveness and availability of interventions. The exemplary method 800 illustrated in Figures 8A and 8B may assist the decision support system 100, and more specifically the decision support engine 114, in determining whether the user's potassium level requires immediate treatment, available home treatment, and / or other actions to recommend to the user to reduce the user's potassium level.

[0345] In some other cases where the decision support engine 114 determines in block 708 that the user's potassium level is not above the elevated potassium threshold (e.g., the user's potassium level <5.0 mmol / L) and the user's potassium level does not exceed the elevated potassium threshold within a short period of time, the decision support engine 114 determines that the user is not at risk for hyperkalemia. Thus, in block 710, the decision support engine 114 determines whether the user's potassium level is below the lowered potassium threshold (e.g., determined in block 706) (e.g., <3.5 mmol / L) and / or the user's potassium level will fall below the lowered potassium threshold within a short period of time (e.g., ≦1 hour). In certain embodiments, the lowered potassium level may be determined by the user's potassium level reaching and / or exceeding the lowered potassium threshold. In certain embodiments, the lowered potassium level may be determined by the user's potassium rate of change and / or a trend indicating that the user is likely to reach and / or exceed the lowered potassium threshold.

[0346] In some cases where the decision support engine 114 determines in block 710 that the user's potassium level is below an elevated potassium threshold (e.g., the user's potassium level <3.5 mmol / L) and / or the user's potassium level falls below a lowered potassium threshold within a short period of time, the decision support engine 114 determines that the user is at risk for hypokalemia. Thus, as shown in Figure 7, the method proceeds to Figure 11. Figure 11 illustrates an example method 1100 used to provide renal disease related decision support to a user determined to be at risk for hypokalemia.

[0347] Although a decreased potassium level may not require immediate treatment (e.g., as with increasing potassium levels), the user's potassium level may need to be increased quickly to a healthy level. Additionally, if the user's potassium level is not currently decreased but may be decreased soon (e.g., within a defined short period of time), the decision support system 100 may provide a recommendation to the user to perform one or more potassium-increasing actions (e.g., consuming potassium). Such recommendations may help reduce instances of hypokalemia, reduce instances of significant / urgent intervention, aid in renal disease management, and / or increase the effectiveness and availability of interventions. An exemplary method 1100 illustrated in FIG. 11 may assist the decision support system 100, and more specifically the decision support engine 114, in determining whether the user's potassium level requires immediate treatment, available home treatment, and / or other actions to recommend to the user for the purpose of increasing the user's potassium level.

[0348] In some other cases where the decision support engine 114 determines in block 710 that the user's potassium level does not fall below the reduced potassium threshold (e.g., the user's potassium level >3.5 mmol / L) and the user's potassium level does not fall below the reduced potassium threshold within a short period of time, the decision support engine 114 determines that the user's potassium level is in the healthy range and is projected to remain in the healthy range. Thus, if the user's potassium level is within the healthy range, no action may be required. However, the decision support system 100 may continue to monitor the user's potassium level for any changes.

[0349] As discussed above, the decision support engine 114 may proceed to (e.g., starting with) Figure 8A to determine one or more recommendations to provide to a user classified (e.g., in Figure 7) as at risk for hyperkalemia. Figures 8A and 8B illustrate an exemplary method 800 used to provide renal disease-related decision support to a user determined to be at risk for hyperkalemia.

[0350] Method 800 is described below with reference to Figures 1 and 2 and their components. For example, components of decision support system 100 illustrated in Figure 1 may use method 800 to determine treatment (or prevention) recommendations to recommend to a user for purposes of reducing the user's elevated potassium levels and / or preventing the user's elevated potassium levels from increasing further.

[0351] Method 800 begins at block 802 by the decision support system 100, and more specifically, the decision support engine 114 of the decision support system 100, which determines whether the user's potassium level is above a strongly elevated potassium threshold (e.g., >6.5 mmol / L) (e.g., as determined in block 706 of FIG. 7) and / or whether the user's potassium level is above the strongly elevated potassium threshold within a period of time, e.g., a short period of time (e.g., ≦1 hour), an intermediate period of time (e.g., ≦2 hours), or an extended period of time (e.g., 2 hours or more). The period of time may be predefined.

[0352] In certain embodiments in which the decision support engine 114 determines in block 802 that the user's potassium level is above a strongly elevated potassium threshold (e.g., the user's potassium level >6.5 mmol / L) and / or that the user's potassium level will be above a strongly elevated potassium threshold for a period of time, the decision support engine 114 determines in block 804 that the user is at severe risk for hyperkalemia. Thus, as shown in FIG. 8A, in block 806, the decision support engine 114 recommends that the user stop any consumption of potassium (e.g., current or future consumption). In certain embodiments in which the user is at severe risk for hyperkalemia, in block 806, the decision support engine 114 may further recommend that the user take insulin and / or take a potassium binder and / or take a diuretic to urgently remove extracellular potassium levels. Thereafter, in block 808, the decision support engine 114 recommends that the user visit an emergency room for treatment and / or analysis by a medical professional. In certain embodiments, the decision support engine 114 may simultaneously recommend to the user to stop consuming potassium and visit an emergency room for treatment and / or analysis by a medical professional.

[0353] In particular, significantly (or strongly) elevated potassium levels (e.g., potassium levels >6.5 mmol / L) may be associated with a severe risk of hyperkalemia. Furthermore, significantly elevated potassium levels may increase the risk of signs and / or symptoms of hyperkalemia and should be reduced immediately. Thus, users with significantly elevated potassium levels may be recommended to (1) seek immediate medical attention in an emergency room and / or (2) immediately discontinue potassium consumption.

[0354] Even if the user's potassium level is not yet at a significantly elevated potassium level, if the user's potassium level rate of change indicates that the user's potassium level will (or is likely to) reach a strongly elevated threshold (e.g., >6.5 mmol / L) during that time period, the user's potassium level may still be associated with a severe risk of hyperkalemia. A user with a significantly elevated potassium level should (1) seek immediate medical attention in an emergency room and / or (2) immediately discontinue potassium consumption.

[0355] In certain other embodiments in which the decision support engine 114 determines in block 802 that the user's potassium level is not above a strongly elevated potassium threshold (e.g., the user's potassium level is ≦6.5 mmol / L) and / or the user's potassium level does not exceed a strongly elevated potassium threshold within a time period, the decision support engine 114 determines in block 810 whether the user's potassium level is above a moderately elevated potassium threshold (e.g., >5.5 mmol / L) (e.g., as determined in block 706 of FIG. 7 ) and / or whether the user's potassium level exceeds the moderately elevated potassium threshold within a defined time period, e.g., within an intermediate time period (e.g., ≦2 hours).

[0356] In certain embodiments in which the decision support engine 114 determines in block 810 that the user's potassium level is above a moderately elevated potassium threshold (e.g., the user's potassium level >5.5 mmol / L) and / or the user's potassium level will be above a moderately elevated potassium threshold within a defined time period, the decision support engine 114 determines in block 812 that the user is at moderate risk for hyperkalemia.

[0357] In particular, the user's elevated potassium level and / or potassium rate of change may indicate that the user's potassium level will reach a moderately elevated threshold (e.g., >5.5 mmol / L) in a defined time period, for example, in a defined intermediate time period (e.g., next 2 hours), and may be associated with a moderate risk of hyperkalemia. A moderate risk of hyperkalemia may not be considered a hyperkalemic emergency requiring fast acting treatment, however, users with a moderate risk of hyperkalemia should attempt to reduce their potassium levels quickly (e.g., within 6-12 hours). In certain embodiments, the user may use home treatments and / or therapies to rapidly reduce their potassium levels. In certain embodiments, the decision support system 100 recommends available therapies to the user with the goal of reducing the user's potassium level to a healthy range. In certain embodiments, the decision support system 100 recommends available home therapies to the user with the goal of reducing the user's potassium level to a healthy range. In certain embodiments, the decision support system 100 recommends that the user immediately stop consuming potassium. In certain embodiments, the decision support system 100 may determine that there is no appropriate treatment available, and in such a case, the patient may be recommended to seek immediate medical care.

[0358] In certain embodiments in which the decision support engine 114 determines in block 812 that the user is at moderate risk for hyperkalemia, the decision support engine 114 may proceed to FIG. 10 to determine one or more dietary recommendations for treatment of renal disease and / or glucose homeostasis for the user classified (e.g., in FIG. 7) as at risk for hyperkalemia.

[0359] In certain embodiments, as shown in FIG. 8A, in block 814, the decision support engine 114 recommends that the user stop any consumption of potassium (e.g., current or future consumption) and increase fluid intake. Further, in block 834 (e.g., illustrated in FIG. 8B), the decision support engine 114 determines whether the user has been prescribed a medication that may cause hyperkalemia. In particular, some medications may increase potassium levels and / or otherwise induce hyperkalemia (e.g., drugs that inhibit the renin-angiotensin-aldosterone system (RAAS), nonsteroidal anti-inflammatory drugs, hypovolemia, etc.). In such embodiments, the user's potassium level may be monitored over an extended period of time, e.g., more than two hours, or more than seven days, to account for the user being prescribed or taking a new medication, such as a non-potassium sparing diuretic. In some examples, a newly prescribed medication may require monitoring of the patient's analyte levels over an extended period of time for a meaningful determination of whether the prescribed medication may actually cause hyperkalemia.

[0360] In certain embodiments in which the decision support engine 114 determines in block 834 that the user has been prescribed a medication that may cause hyperkalemia, in block 836, the decision support engine 114 recommends that the user consult with a medical professional for the purpose of adjusting the medication currently prescribed for the user.

[0361] In certain other embodiments (e.g., not illustrated) in which the decision support engine 114 determines in block 834 that the user has been prescribed a medication that may cause hyperkalemia, the decision support engine 114 recommends an adjustment to the administration of such medication. The adjustment recommendation of the decision support engine 114 may be based, at least in part, on the medication itself, the medication parameters (e.g., type, dosage, timing, frequency, etc.), the user's potassium level, and / or the user's potassium change rate.

[0362] Medication adjustments recommended by the medical professional and / or decision support system 100 may include (1) skipping a dose, (2) taking a lower dose, (3) taking a dose with a meal, (4) taking a dose at a later time, and / or (5) taking a dose with another potassium reducing mechanism (e.g., another medication). Additionally, medication adjustments recommended by the medical professional and / or decision support system 100 may include (1) recommended temporary adjustments, and / or (2) recommended permanent adjustments (e.g., adjustments to administration of medication beyond the next dose).

[0363] In certain other embodiments in which the decision support engine 114 determines in block 834 that the user is not prescribed a medication that may cause hyperkalemia, in block 838 the decision support engine 114 determines whether the user is receiving exogenous insulin (e.g., injecting and / or infusing insulin via an insulin pump).

[0364] In certain embodiments in which the decision support engine 114 determines in block 838 that the user is receiving exogenous insulin, the decision support engine 114 may proceed to FIG. 9 to determine one or more recommendations to provide for treatment of renal disease and / or glucose homeostasis for a user classified (e.g., in FIG. 7) as at risk for hyperkalemia. In particular, insulin stimulates potassium and / or glucose uptake by cells while reducing serum (e.g., extracellular) potassium and / or glucose levels. Thus, a user taking exogenous insulin may be recommended to administer additional insulin depending on the user's glucose level. FIG. 9 illustrates an exemplary method 900 of assessing a user's glucose level before recommending the user to administer exogenous insulin and / or consume glucose to stimulate endogenous release of insulin.

[0365] In certain embodiments in which the decision support engine 114 determines in block 838 that the user is not taking exogenous insulin, in block 840 the decision support engine 114 recommends that the user consume glucose to stimulate endogenous release of insulin. For example, even if the user is not taking exogenous insulin, the user may increase their insulin levels by stimulating insulin release. Insulin is released when glucose levels rise. Thus, the decision support engine 114 may recommend that the user consume additional glucose to increase serum glucose levels. The increase in serum glucose levels may trigger insulin release in the user. The released insulin may help reduce the user's glucose and / or potassium levels, thereby helping to reduce the user's risk of hyperkalemia.

[0366] In certain embodiments, if the decision support engine 114 determines in block 838 that the user is not taking exogenous insulin, then in block 840 the decision support engine 114 recommends that the user take insulin to reduce the user's potassium and / or glucose levels. For example, for a user experiencing a moderate risk of potassium (e.g., elevated potassium), administering insulin may help reduce the user's potassium level to...

Claims

1. 1. A surveillance system comprising: a continuous analyte sensor configured to generate an analyte measurement associated with an analyte level in a patient; a sensor electronics module coupled to the continuous analyte sensor and configured to receive and process the analyte measurements; A monitoring system comprising:

2. The continuous analyte sensor comprises: A substrate; A working electrode arranged on the substrate; A reference electrode arranged on the substrate; Equipped with 10. The monitoring system of claim 1, wherein the analyte measurements produced by the continuous analyte sensor correspond to an electromotive force based at least in part on a potential difference generated between the working electrode and the reference electrode.

3. the continuous analyte sensor is a continuous potassium sensor; The monitoring system of claim 1 , wherein the analyte measurement comprises a potassium measurement.

4. a memory containing executable instructions; one or more processors; Further comprising: The one or more processors are in data communication with the sensor electronics module and are configured by the executable instructions to: receiving, from the sensor electronics module, analyte data associated with the analyte measurements, the analyte measurements being associated with one or more analytes and generated by the continuous analyte sensor over a plurality of time periods, the analyte data including potassium data; processing the analyte data from the plurality of time periods to determine at least one rate of change of potassium for the patient based on the potassium data; 4. The monitoring system of claim 3, configured to generate a disease prediction using the analyte data for the one or more analytes, including the potassium data and at least one rate of change of the potassium for the patient.

5. The monitoring system of claim 4 , wherein the one or more analytes of the patient are monitored continuously, semi-continuously, or periodically during the multiple time periods.

6. The disease prediction comprises: an indication of the presence of one or more diseases in said patient; an indication of the severity of said one or more diseases in said patient; an indication of the level of risk that the patient will be diagnosed with the one or more diseases; and an indication of the level of improvement or worsening of said one or more diseases in said patient; The monitoring system of claim 4 , comprising at least one of:

7. The indicator of the level of improvement or worsening of the one or more diseases in the patient may comprise: any procedures previously performed on the patient; any medication previously taken by said patient; one or more actions taken by the patient; or one or more other clinical procedures; The surveillance system of claim 6 based in part on at least one of:

8. The disease prediction comprises: an indication of a level of risk that the patient will be diagnosed with at least one of hyperkalemia or hypokalemia within a first threshold time; an indication of the level of risk that the patient will experience a cardiac event within a second threshold time; or an indication of a level of risk that said patient will experience a serious medical outcome due to a potassium imbalance in said patient within a third threshold time; The monitoring system of claim 4 , comprising at least one of:

9. The monitoring system of claim 4 , wherein the processor is further configured to generate one or more recommendations for treatment based at least in part on the disease prediction.

10. The one or more treatment recommendations include: a predicted effect of said one or more recommendations on one or more organs of said patient; a predicted effect of said one or more recommendations on one or more conditions of said patient; the patient's insulin level, or 10. The monitoring system of claim 9, further generated based on at least one of: one or more other medications previously prescribed for the patient.

11. The one or more treatment recommendations include: Drug prescription recommendations, Medical supplements recommended, Insulin dosage recommendations, Recommendations for invasive or non-invasive procedures; a medical device recommendation for use by said patient; Recommendations for improving lifestyle habits, Exercise recommendation, or Recommendations for improving diet, The monitoring system of claim 9 , comprising at least one of:

12. The monitoring system of claim 11 , wherein the insulin dosage recommendation includes a recommended combined dosage of insulin and glucose.

13. The monitoring system of claim 4 , wherein the one or more analytes further comprise at least one of glucose, creatinine, blood urea nitrogen (BUN), ammonia, C-peptide, and cystatin C.

14. The processor, 5. The monitoring system of claim 4, further configured to monitor other sensor data for the patient during the multiple time periods using one or more other non-analyte sensors, and the disease prediction is further generated using the other sensor data.

15. 15. The monitoring system of claim 14, wherein the one or more other non-analyte sensors include at least one of an insulin pump, an accelerometer, a temperature sensor, an electrocardiogram (ECG) sensor, a heart rate monitor, a blood pressure sensor, an impedance, a peritoneal dialysis machine, a hemodialysis machine, a continuous positive airway pressure machine, an internal body sound sensor, or a respiration sensor.

16. 5. The monitoring system of claim 4, wherein the disease predictions are generated using a model trained using training data, the training data including records of historical patients having various stages of renal disease.

17. the processor is further configured to obtain at least one of demographic information, food consumption information, activity level information, medical information, health and disease information, or disease stage information regarding the patient; 5. The monitoring system of claim 4, wherein the disease prediction is further generated using at least one of the food consumption information, the activity level information, the medical information, the health and illness information, or the disease stage information regarding the patient.

Images