Sensing system and method for diagnosing kidney disease

Abstract

Certain aspects of the present disclosure provide a monitoring system comprising a continuous analyte sensor configured to generate analyte measurements related to an analyte level of a patient, and a sensor electronics module coupled to the continuous analyte sensor and configured to receive and process the analyte measurements.

Description

Technical Field

[0001] (Cross - Reference to Related Applications) This application claims priority and benefit of U.S. Provisional Patent Application No. 63 / 365,702, filed on June 1, 2022; U.S. Provisional Patent Application No. 63 / 376,673, filed on September 22, 2022; U.S. Provisional Patent Application No. 63 / 387,078, filed on December 12, 2022; U.S. Provisional Patent Application No. 63 / 377,332, filed on September 27, 2022; U.S. Provisional Patent Application No. 63 / 403,568, filed on September 2, 2022; and U.S. Provisional Patent Application No. 63 / 403,582, filed on September 2, 2022, which are assigned to the assignee of this specification and are hereby expressly incorporated by reference in their entirety as if fully set forth herein and for all applicable purposes.

Background Art

[0002] The kidneys perform many important functions in the human body, including filtering waste products and excess fluids excreted in urine, and removing acids produced by the body's cells to maintain a healthy balance of water, salts, and minerals (such as sodium, calcium, phosphorus, and potassium) in the blood, but are not limited to these. Therefore, the kidneys play a major role in homeostasis through kidney mechanisms that transport water, salts, and minerals and regulate their secretion, reabsorption, and excretion. Additionally, the kidneys secrete renin (such as angiotensinogenase), which forms part of the renin-angiotensin-aldosterone system (RAAS) that mediates extracellular fluid and arterial vasoconstriction (e.g., blood pressure). More specifically, hypertension (e.g., hypertension) can be regulated through RAAS inhibitors such as angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARB). When the kidneys become diseased or damaged, impairment or loss of these functions can cause significant damage to the human body.

[0003] Kidney diseases occur when the kidneys become diseased or damaged. Kidney diseases are generally classified as either acute or chronic, based on the duration of the disease. Acute kidney injury (AKI) (also called "acute renal failure" or "acute renal insufficiency") is usually caused by sudden events that cause kidney dysfunction, such as dehydration, blood loss due to major surgery or injury, and / or the use of medications. On the other hand, chronic kidney disease (CKD) (or "chronic renal failure") is usually caused by long-term diseases such as hypertension or diabetes that gradually damage the kidneys and reduce their function over time.

[0004] As briefly described above, the kidneys play a major role in potassium homeostasis through kidney mechanisms that transport water, salts, and minerals and regulate their secretion, reabsorption, and excretion. In some cases, elevated potassium levels in patients with untreated CKD can lead to hyperkalemia. Hyperkalemia is a medical term describing blood potassium levels that are higher than normal (e.g., higher than the normal blood potassium level of 3.6 - 5.2 millimoles per liter (mmol / L)). Hyperkalemia can increase the risk of cardiac arrhythmia episodes and even sudden death. Symptoms associated with mild hyperkalemia include muscle weakness, numbness, tingling, nausea, or other abnormal sensations, while symptoms of very high potassium levels include palpitations, shortness of breath, chest pain, nausea, or vomiting. In more severe cases of hyperkalemia, patients may experience respiratory failure, cardiac sudden death, or other fatal events.

[0005] Similarly, low potassium levels in patients with untreated CKD can lead to the progression of hypokalemia. Hypokalemia is a medical term describing blood potassium levels that are lower than normal. Patients with CKD can develop hypokalemia due to, for example, gastrointestinal potassium losses from diarrhea or vomiting, or renal potassium losses from non - potassium - sparing diuretics (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). Similar to hyperkalemia, severe hypokalemia can result in symptoms of respiratory failure, cardiac sudden death, or other fatal events.

[0006] The lack of easily noticeable symptoms in many patients is the reason why hyperkalemia and hypokalemia are often referred to as "silent killers," especially when patients are sensitive to irregular potassium levels. For example, in severe cases where hypokalemia or hyperkalemia leads to severe symptoms such as fatal events, the diagnosis of hypokalemia or hyperkalemia as the mediating mechanism may not become immediately apparent until the patient is evaluated by medical personnel.

[0007] CKD can alter a patient's glucose homeostasis, thereby making CKD an independent risk factor for hypoglycemia. In particular, the kidneys also play an important role in regulating blood glucose (e.g., blood sugar). Regarding the kidneys' involvement in glucose homeostasis, the main mechanisms include the release of glucose into the circulation via gluconeogenesis, the uptake of glucose from the circulation to meet the energy needs of the kidneys, and the reabsorption of glucose at the level of the proximal renal tubules. For example, gluconeogenesis is the formation of glucose from precursors (e.g., lactate, glycerol, and / or amino acids). During gluconeogenesis in the kidneys, glucose is formed by the kidneys and released into the circulation. Gluconeogenesis occurs mainly in the liver but also in the kidneys and maintains glucose homeostasis by preventing hypoglycemia.

[0008] However, as renal function declines, glucose formation also declines, and thus the kidneys' ability to respond to a drop in blood sugar is limited. In some cases, the reduced ability of the kidneys to respond to a drop in blood glucose levels can lead to hypoglycemia. Hypoglycemia is a medical term describing blood glucose (e.g., sugar) 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 in patients with early-stage CKD (CKD < 3). Furthermore, for hospitalizations related to end-stage renal disease (ESRD), 3.6% are due to hypoglycemia, and the mortality rate is 30%. In particular, severe hypoglycemia can lead to myocardial damage, neurocognitive impairment, and in certain cases, seizures or even death.

[0009] Therefore, renal dysfunction can result in lower and / or longer low glucose levels. A damaged kidney is slow to counteract a drop in glucose levels and in some cases may be less effective, resulting in inadequate glucose control (considering, for example, that the kidney filters insulin, reabsorbs filtered glucose from the proximal tubule, and produces glucose by gluconeogenesis). Hypoglycemic events in kidney disease are due to a decrease in insulin clearance by the kidney and impairment of gluconeogenesis. Regarding ESRD-related hospitalizations, dialysis is also a worsening factor.

[0010] Furthermore, a decrease in insulin metabolism and clearance can occur as a result of declining kidney health. Insulin is a hormone that enables the body to use glucose as energy or store glucose as fat. In other words, insulin stimulates the uptake of potassium and glucose by the patient's cells and reduces the levels of serum (e.g., extracellular) potassium and glucose. Insulin is removed by the kidney, and thus, as kidney function declines, insulin is removed more slowly. Therefore, typical dosing of insulin can have an extended and / or more pronounced effect on glucose in patients with renal dysfunction. As renal dysfunction progresses, insulin can have an even more extended and / or more pronounced effect on glucose. Therefore, patients with kidney disease can be at risk of hypoglycemia, and the risk increases with disease progression. For example, patients with CKD may find that insulin dosing, which predicts glucose clearance at a later point in time, results in hypoglycemia.

[0011] Kidney dysfunction can also result in higher and / or longer elevated glucose levels. For example, considering that the kidneys filter, reabsorb, and consume glucose from the blood, a malfunctioning kidney has a slower effect in reducing (i.e., removing) glucose levels and, in some cases, can be lower. Patients suffering from higher and / or longer elevated glucose levels can be diagnosed with hyperglycemia. Hyperglycemia is a medical term that describes blood sugar (e.g., glucose) levels that are higher than normal (e.g., significantly elevated blood sugar levels, usually above 180 - 200 mg / dL, or 10 - 11.1 mmol / L). Patients with hyperglycemia can be affected by 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 health complications, including but not limited to anemia, bone weakness, fluid retention, gout, heart disease, hypertension, hyperphosphatemia, metabolic acidosis, uremia, etc., may also occur with CKD. These complications, as well as the above-mentioned complications, are more frequent and severe as the kidney disease progresses and can lead to a decline in quality of life and an increase in morbidity and mortality.

Brief Description of the Drawings

[0013] To enable a more detailed understanding of the features of the present disclosure listed above, a more specific description, briefly summarized above, can be made by referring to the embodiments, some of which are illustrated in the drawings. However, it should be noted that the attached drawings merely illustrate certain exemplary embodiments of the present disclosure and, therefore, should not be considered as limiting its scope since its description can be recognized for other equally effective embodiments.

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7A

Figure 7B

Figure 7C

Figure 7D

Figure 7E

Figure 8A

Figure 8B

Figure 8C

Figure 8D

Figure 8E

Figure 9A

Figure 9B

Figure 9C

Figure 10

Figure 11A

Figure 11B

Figure 11C

Figure 11D

Figure 12A

Figure 12B

Figure 13

Figure 14A

Figure 14B

Figure 14C

Figure 15

[0014] For ease of understanding, the same reference numbers are used, where possible, to designate the same elements common to each figure. It is contemplated that elements disclosed in one aspect may be beneficially utilized in other aspects without specific recitation.

DETAILED DESCRIPTION OF THE INVENTION

[0015] An accurate assessment of renal function (and in some cases, cardiac function) is important as a screening tool and to monitor disease progression and guide prognosis, at least with respect to chronic kidney disease (CKD). However, conventional disease diagnosis methods and systems for such diseases, including, but not limited to, glomerular filtration rate (GFR) tests, albumin-to-creatinine ratio (ACR) tests, electrocardiogram (ECG) monitoring, and blood tests to monitor a patient's potassium level, face many challenges with respect to accuracy and reliability. Furthermore, conventional disease diagnosis methods and systems generally cannot provide an efficient and complete analysis of factors that may contribute to CKD, and thus, CKD is difficult to diagnose at its early stages.

[0016] The current standard for diagnosing CKD is based on the glomerular filtration rate (GFR). GFR is an assessment of the flow rate of filtered fluid through the kidneys and is determined by measuring or estimating the clearance rate of an exogenous or endogenous filtration marker that is removed solely by glomerular filtration. The clearance rate of the marker is the volume of plasma from which the marker is removed per unit time and is used to approximate the GFR. Thus, GFR is used as a rough measure of renal function and can help determine the presence and severity of CKD.

[0017] GFR can be evaluated from urine or plasma clearance measurements of exogenous filtration markers (e.g., measured GFR (mGFR)), or from measured serum levels of endogenous filtration markers using an equation (e.g., estimated GFR (eGFR)). However, both of these methods have limitations with respect to both the administration of such tests and the data collected therefrom. For example, in the mGFR test, multiple blood (or urine) tests are performed to calculate the clearance rate of an exogenous marker such as iothalamate or inulin over a period of several hours (e.g., 5 - 6 hours). Due to the early rapid decay of the marker, samples are collected over several hours, thereby making it impossible to determine the initial point of marker administration. Consequently, both the number of samples and the collection period increase to compensate for the lack of initial data. As a result, the mGFR test is costly, time-consuming, invasive, and potentially painful for the patient. Furthermore, the mGFR test requires the patient to visit a clinic or other healthcare facility, thereby adding to its costly and inconvenient nature.

[0018] Results from the mGFR method are also limited in the evaluation of kidney filtration and secretion. Renal reserve is the ability of the individual nephrons of the kidney to increase filtration by up to 30% in response to stress or a high-protein load. In other respects healthy individuals with intact renal reserve undergoing an mGFR test have increased filtration due to renal reserve. This can lead to an overestimation of actual kidney function by the mGFR test. To normalize mGFR results and account for renal reserve, renal reserve can be measured separately in a renal reserve test in which an amino acid solution is administered to the patient and urine output is then monitored. However, in the case of CKD patients undergoing an mGFR test, renal reserve may be impaired and any increase in filtration during the test, if any, may be limited, thereby creating uncertainty as to the extent, if any, to which mGFR results overestimate actual kidney function.

[0019] Furthermore, the mGFR method only determines overall kidney function and does not distinguish between filtration activity and secretion activity. Therefore, specific aspects of kidney health, such as glucose reabsorption, cannot be determined by mGFR alone. Rather, additional tests may be required, such as to determine the health status of the renal tubules and the presence of tubulointerstitial fibrosis, both of which are important aspects of kidney health. Thus, for a more comprehensive evaluation of global kidney health, in addition to the mGFR test, monitoring of additional kidney health markers, such as markers of renal tubule health, may be necessary.

[0020] Rather than measuring GFR, estimating GFR provides a more convenient and rapid analysis for assessing kidney function. Estimated GFR (eGFR) is typically determined based on the estimation of the clearance rate of an endogenous filtration marker, such as creatinine, which can be determined from a urine and / or blood sample from the patient. For example, a blood sample can be taken from the patient over 24 hours and the creatinine level therein measured to estimate the patient's creatinine clearance rate based on several assumptions. Generally, different equations can be used depending on whether the creatinine measurement was obtained from a blood sample or a urine sample, and further depending on the patient's age, gender, weight, and / or ethnicity. However, the equations for estimating GFR are based on specific assumptions and thus may not be generalizable across all populations. Furthermore, similar to mGFR, the eGFR method cannot take into account the patient's serum potassium level and insulin level.

[0021] As another example, the ACR test is commonly used to screen for and diagnose kidney disease. In particular, the ACR test measures both albumin and creatinine in a single urine sample, also known as a spot urine sample. ACR is a preferred first method for detecting elevated protein, specifically albumin, in the urine. A persistent increase in urinary albumin levels measured using the ACR test, i.e., an increase in albumin excretion, provides a marker of kidney damage, but the test is not without its drawbacks. For example, inaccurate results can be obtained and thus misinterpretation can occur if the effect of creatinine excretion on the ratio of albumin to creatinine in the current use of the ACR test is not taken into account. Furthermore, spot 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, and thus random spot urine ACR results are less consistent than timed urine samples.

[0022] Both the eGFR and ACR test methods also have the drawback of their dependence on the measurement of creatinine levels. Creatinine is a waste product produced by muscle from the breakdown of creatine, a non-protein compound that facilitates the recycling of adenosine triphosphate (ATP). Creatinine is then filtered from the body by the kidneys and excreted with urine. Since creatinine is produced by muscle, creatinine levels are related to an individual's muscle mass and can thus, in certain situations, be more reflective of a patient's muscle mass than of their kidney function.

[0023] Creatinine levels also undergo a 24 - 48 hour biological delay. For example, in patients suffering from acute kidney injury (AKI), the patient's creatinine level may not reflect the injury, and thus the change in kidney function, until 24 - 48 hours after the injury. Therefore, the measured creatinine level may not accurately reflect the patient's kidney function in real - time. In some cases, up to more than half of kidney function may be lost before the creatinine level rises to a level at which CKD can be diagnosed. In fact, kidney dysfunction is accurately reflected only in creatinine levels after a significant decrease in GFR (e.g., <50 mg / dL), which corresponds to CKD stage 3 or higher. Therefore, ACR and eGFR are not useful for diagnosing chronic kidney disease until stage 3 or later.

[0024] Outside of GFR and ACR, another conventional method for identifying potassium imbalances, and thus kidney problems, is electrocardiogram (ECG) monitoring. More specifically, ECG monitoring is touted as a method for recognizing the arrhythmogenic effects of severe hyperkalemia and / or hypokalemia, such as peaked T waves, QRS widening, shortened PR (e.g., the PR interval is the time from the start of the P wave (atrial depolarization) to the start of the QRS complex (ventricular depolarization), and a shortened PR interval may indicate a specific disease), bradycardia (e.g., a heart rate slower than expected), and other indicators of heart function. Therefore, since hyperkalemia and / or hypokalemia can result from chronic kidney dysfunction, changes in ECG measurements reflecting hyperkalemia and / or hypokalemia may indicate the need for further examination of kidney dysfunction.

[0025] However, there is a treatment paradox in that an ECG device or the ability to interpret such an ECG device is not readily accessible to most patients, for example, in the patients' own homes. Further, the evidence is conflicting regarding whether ECG findings are reliable, particularly in patients with chronic hyperkalemia and / or hypokalemia. Further, changes in potassium levels occur well before the corresponding changes in cardiac function detected by an ECG. Since changes in potassium levels underlie the corresponding changes in cardiac function that can be detected by an ECG, potassium levels change before cardiac function. Thus, it is desirable to know whether potassium levels are changing outside of a safe range. This is because impending cardiac arrhythmias can be indicated before these dangerous arrhythmias occur in the patient. 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 indicating that a given condition exists when it does not, and a false negative is a result indicating that a given condition does not exist when it actually does. Further, such monitoring lacks the ability for continuous monitoring required to provide an overall picture of a patient's health.

[0026] Additionally, in some cases, T-wave abnormalities detected by ECG monitoring may be due to factors unrelated to potassium levels. For example, T-wave abnormalities can result from subarachnoid hemorrhage, ischemic stroke, subdural hematoma, heart failure, myocardial edema, viral infection (e.g., Covid-19), traumatic brain injury, rare diseases, and, without limitation, as a result of certain tumor pathways such as pheochromocytoma. The value of using ECG alone to diagnose clinically significant hyperkalemia is further complicated by the coexistence of T-wave abnormalities in the context of renal disease. In these cases, it is even more difficult to determine whether the result of the T-wave abnormality is due to the result of renal disease or another clinical situation. Therefore, T-wave abnormalities recognized by ECG monitoring may not always be caused by a decrease in renal function (e.g., abnormal potassium levels may, in some cases, be due to a decrease in renal function). Therefore, ECG alone may not provide sufficient information to evaluate kidney health and may also not provide sufficient information regarding the extent of renal disease in a patient's body.

[0027] Here too, adverse events associated with chronic kidney dysfunction are often related to high or low serum potassium levels (i.e., hyperkalemia or hypokalemia, respectively), and if left untreated, can create a medical situation requiring urgent medical attention. Potassium is an important electrolyte that helps regulate fluid balance, muscle contraction, and nerve signaling in the human body. The kidneys primarily play a role in maintaining potassium homeostasis in the body through control of potassium secretion, reabsorption, and excretion mechanisms. Therefore, when renal function declines, the control mechanisms for maintaining potassium homeostasis can also decline. Therefore, since potassium levels can indicate changes in renal function, measuring a patient's potassium level is important for the screening, diagnosis, staging, and monitoring of CKD.

[0028] However, the current clinical standard for potassium measurement, and thus for kidney health assessment, is a blood test. In some cases, whole blood samples are obtained by pricking a finger with a lancet. In some cases, blood samples are obtained using venipuncture. In venipuncture, a needle is inserted into a vein to collect a blood sample for testing. Venous blood samples are often ordered for patients weekly. However, measuring whole blood carries the risk of hemolysis (e.g., rupture of red blood cells by external force), which is common particularly in the finger prick method of blood collection, and this can lead to false positive measurements due to the high intracellular concentration of potassium released upon cell rupture. Since the red blood cell potassium concentration is much higher (about 20 times higher) than that of plasma, among all routine blood tests, plasma / serum potassium measurement is one of the most sensitive measurements to the effects of hemolysis. Thus, even slight hemolysis can cause spuriously high plasma potassium concentrations, which can render the screening, diagnosis, and staging of CKD based on potassium measurements unreliable.

[0029] Furthermore, ensuring that a patient continues to participate in such potassium monitoring activities as daily or weekly blood tests can itself pose problems. In the case of patients with CKD, venous blood samples are often ordered weekly for potassium monitoring. However, such regular blood sampling can be costly, time-consuming, and painful for the patient. As a result, a patient may decide to forgo such potassium monitoring activities. And patients who forgo engaging in such potassium monitoring activities that help stabilize their chronic condition may not be able to manage their condition outside of such tests. If the condition is left unmanaged for too long, the patient's condition can deteriorate significantly, additional health problems can arise, and in some cases, this can lead to an increased risk or likelihood of death.

[0030] Overall, existing diagnostic methods as described above have a first technical problem in that they cannot continuously monitor the concentration of a changing analyte containing potassium to provide a continuous readout. As used herein, the term "continuous" can mean fully continuous, semi - continuous, periodic, etc., and thus, "monitor continuously" can mean monitor continuously, monitor semi - continuously, monitor periodically, etc. Such continuous monitoring of an analyte is advantageous for screening, diagnosing, and staging a patient's disease when the continuous measurements provide continuously up - to - date measurements, as well as information regarding the trend and rate of change of the analyte concentration over a continuous period. Such information can be used to make more informed decisions in the assessment of kidney health and the treatment of kidney diseases, more specifically chronic kidney disease (CKD).

[0031] Second, existing diagnostic methods have another technical problem in that they cannot simultaneously and continuously monitor the concentrations of multiple changing analytes such as potassium and, for example, glucose. In particular, continuous monitoring of multiple analytes such as potassium, glucose, creatinine, lactate, urea (via blood urea nitrogen (BUN)), cystatin C, and / or C - peptide can provide additional insights when assessing the presence and / or severity of a patient's kidney disease, hyperkalemia, hypokalemia, as well as hyperglycemia and / or hypoglycemia. Furthermore, the additional insights obtained from using combinations of analytes rather than a single analyte can increase the accuracy of predictions and can help in making more informed patient - specific decisions and / or recommendations for the screening, diagnosis, staging, and monitoring of CKD.

[0032] As a result of these technical problems, the diagnosis of kidney disease or its risk using conventional techniques can be not only inaccurate but also impossible, and in some cases, it may turn out to be life-threatening for patients having such diseases. Specifically, considering the dynamic and latent nature of early-stage kidney diseases, as well as the heterogeneity of patients, it may be necessary to predict the progression of chronic kidney disease (CKD) in an individualized manner with reasonable accuracy. Therefore, improved methods for screening, diagnosing, and stage-classifying CKD in patients, as well as methods for understanding the interaction between the progression of CKD and measured analyte levels in patients, are desirable.

[0033] Accordingly, certain embodiments described herein provide a technical solution to the above technical problems by using a continuous analyte monitoring system to provide decision-making support around kidney disease, particularly chronic kidney disease (CKD). Further, certain other embodiments described herein provide a technical solution to the above technical problems by using a continuous analyte monitoring system including at least a continuous potassium monitor (CPM) to provide decision-making support around kidney disease. The decision-making support can be provided in the form of risk assessment (e.g., screening), diagnosis, stage classification, and / or monitoring of kidney disease. As used herein, risk assessment can refer to the evaluation or estimation of the current or future incidence of one or more symptoms associated with kidney disease, such as kidney dysfunction, kidney disease, hyperkalemia, and / or hypokalemia.

[0034] According to embodiments of the present disclosure, the decision-making support system presented herein is designed to provide disease decision-making support for the risk assessment, diagnosis, and / or staging of patients with or at risk of kidney disease, and to assist patients in preventing, attenuating, and / or managing kidney disease or its risk. Providing kidney disease decision-making support may include, for example, using a large amount of collected data including the above analyte data, patient information, and secondary sensor data to (1) automatically detect and classify abnormal kidney function, (2) evaluate the risk of kidney disease; and (3) evaluate the presence and stage of kidney disease. In other words, the decision-making support system presented herein may provide information useful for instructing and improving the care of patients with or at risk of kidney diseases such as CKD.

[0035] In certain embodiments, the decision-making support is provided in the form of a risk assessment of a patient who has developed a kidney disease, such as CKD. In other words, the decision-making support is provided in the form of a kidney disease screening. For example, periodic or continuous analyte measurements provided by one or more analyte sensors can indicate an increased risk of developing a kidney disease, and additional diagnostic tests for kidney disease may be recommended. In another example, periodic or continuous analyte measurements provided by one or more analyte sensors can indicate a low risk of developing a kidney disease, and additional diagnostic tests for kidney disease are not recommended. Such analyte sensors, which may include a continuous glucose monitor (CGM) and / or a continuous potassium monitor (CPM), can be specifically used by a patient to screen for kidney disease according to the methods described herein. Alternatively, if a patient is already utilizing an analyte sensor such as a CGM or CPM, the decision-making support system can periodically and / or continuously monitor or screen for an increased risk of kidney disease and can alert the user (e.g., the patient) if the risk of developing a kidney disease has increased. For example, a CGM of a diabetic patient can be utilized to monitor an increased risk of kidney disease in the diabetic patient. In some examples, when such a risk reaches or exceeds a predetermined threshold, a CPM may be recommended for the patient to screen for kidney disease.

[0036] In certain embodiments, the decision support is provided in the form of a diagnosis of a kidney disease, such as CKD. In other words, the decision support is provided in the form of an assessment of the presence and / or stage classification of a kidney disease. For example, the periodic or continuous analyte measurements provided by one or more analyte sensors can indicate the presence and / or stage of a patient's kidney disease, which can be confirmed by additional diagnostic tests. In another example, the periodic or continuous analyte measurements provided by one or more analyte sensors can indicate a patient's healthy kidney function. For example, the periodic or continuous analyte measurements can include the time that a patient's analyte concentration rotates around a given set point over 24 hours, or a portion of a day (e.g., daytime or nighttime). Further examples can include the time exceeding a given threshold during a 24-hour day or a portion of a day (e.g., daytime or nighttime). Such metrics can be based on analyte data such as potassium level, potassium level threshold, potassium level change rate, change in potassium change rate and rate of change, average potassium level, standard deviation of potassium level, potassium clearance rate, personalized potassium data, and / or other changes in potassium data.

[0037] In certain examples, the kidney disease decision support can further include assessing the risk of adverse health events associated with the kidney disease and / or providing patient-specific treatment recommendations for the kidney disease. For example, in certain embodiments, the decision support is provided in the form of an assessment of the risk of mortality due to the kidney disease. In certain embodiments, the decision support is provided in the form of an assessment of the risk of health adverse events such as hyperkalemia, hypokalemia, cardiac events. In certain embodiments, the decision support is provided in the form of an assessment of the risk of comorbidities such as hypoglycemia, hyperglycemia, liver disease.

[0038] In certain embodiments, the decision support system can provide decision support to a patient based on various collected data, including analyte data, patient information, secondary sensor data (e.g., non-analyte data), and the like. For example, the analyte data can include continuously monitored potassium data, in addition to other continuously monitored analyte data such as glucose, creatinine, urea (blood urea nitrogen (BUN)), inulin, dextran, saccharin, iothalamic acid, iohexol, 125l-iothalamate, cystatin C, C-peptide, 51Cr-EDTA, lactate, aspartic acid, polyfructosan, and betaine, collected by a continuous analyte monitoring system.

[0039] 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 kidneys play a major role in potassium homeostasis through kidney mechanisms that transport potassium and regulate its secretion, reabsorption, and excretion. Thus, the continuously monitored analyte data can include potassium data measured by a continuous potassium monitor (CPM) that indicates or determines the patient's potassium level and / or the rate of change of the patient's potassium level over time for assessing the health and function of the patient's kidneys.

[0040] According to certain embodiments, the decision support system described herein is designed to provide decision support in the form of risk assessment and treatment for CKD and / or potassium homeostasis. For example, in certain embodiments, the decision support system is designed to continuously measure the patient's serum potassium level and provide recommendations for treatment, specifically for potassium imbalances associated with kidney disease.

[0041] In certain embodiments, the decision-making support risk assessment and treatment recommendation may be based on the patient's potassium level, rate of change, trend, and / or threshold. For example, different thresholds can be set based on the risk of hyperkalemia and / or hypokalemia, available treatments, treatment effectiveness, patient characteristics, patient activities, and the like.

[0042] Certain embodiments of the present disclosure also 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 further described below. As described above, the collected data may also include patient information that may include information related to age, gender, family history of kidney disease, other health conditions, and the like. Secondary sensor data may include accelerometer data, heart rate data (ECG, HRV, HR, etc.), temperature, blood pressure, time of sensor start or remaining sensor life relative to the start time, or any other sensor data other than analyte data.

[0043] In certain embodiments, the decision-making support system described herein may use various algorithms or artificial intelligence (AI) models, such as machine learning models, trained based on patient-specific data, historical data, and / or population data to provide real-time decision-making support to a patient based on the information collected about the patient. For example, certain aspects may target algorithms and / or machine learning models designed to evaluate the presence and severity of kidney disease in a patient. The algorithms and / or machine learning models may be used in combination with one or more continuous analyte sensors, such as CPM, to provide real-time kidney disease evaluation and stage classification. In particular, the algorithms and / or machine learning models can take into account parameters such as potassium level, rate of change of the patient's potassium level over time, and other physiological parameters of the patient that are generally associated with kidney disease when screening, diagnosing, and staging CKD.

[0044] Based on these parameters, an algorithm and / or a machine learning model can provide a risk assessment of different CKD stages (and their corresponding severities), as well as the progression of a patient towards one or more of these CKD stages. The algorithm and / or machine learning model can take into account population data, individualized patient-specific data, or a combination of both when screening, diagnosing, and staging CKD for a patient.

[0045] 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. As described in more detail herein, the population data can be provided in the form of a dataset containing data records of historical patients having various stages of kidney disease. Each data record can be used as an input to the machine learning model during training to optimize such a model to generate as output accurate predictions related to CKD, such as predictions of the risk, presence, and / or severity of kidney disease in a patient.

[0046] The combination of a continuous analyte monitoring system with a machine learning model and / or algorithm for screening, diagnosing, staging, and evaluating the risk of CKD provided by the decision-making support system described herein enables real-time diagnosis and early intervention. In particular, the decision-making support system can be used to provide early alerts of declining renal function and / or deliver information about other complications related to the kidneys. Early detection of such decompensation and / or other complications can enable intervention at the earliest possible stage to ultimately improve renal disease outcomes. For example, early intervention can, in some cases, reduce hospitalizations, complications, and deaths. Additionally, the potassium levels and changes in potassium levels provided by the continuous analyte monitoring system can be used as inputs to the machine learning model and / or algorithm to triage patients for more urgent care. In patients at risk of arrhythmia episodes and / or death, a sudden increase in potassium can be used to notify for urgent medical intervention even before the patient can report significant physiological symptoms.

[0047] In addition, through the combination of a continuous analyte monitoring system and machine learning and / or algorithms for screening, diagnosing, staging, and evaluating the risk of CKD, the decision-making support system described herein can provide the accuracy and reliability required by the patient. For example, when assessing the presence and severity of kidney disease in a patient, bias, human error, and emotional influences can be minimized. Furthermore, the machine learning model and algorithm combined with the analyte monitoring system can provide insights into patterns and / or trends that may have previously gone undetected and that degrade the health of the patient, at least with respect to the kidneys and / or the heart. Thus, the decision-making support system described herein can assist in the identification of kidney health for CKD screening, diagnosis, prevention, and treatment purposes.

[0048] Exemplary Decision-Making Support System Including Exemplary Analyte Sensors FIG. 1 shows an exemplary decision support system 100 for screening, diagnosing, staging, and treating kidney disease, particularly chronic kidney disease (CKD), with respect to a user 102 (individually referred to herein as a user and collectively referred to herein as users) using a continuous analyte monitoring system 104 that includes one or more analyte sensors. The user may be a patient in certain embodiments, or, in some cases, a caregiver of a patient. In certain embodiments, the decision support system 100 includes a continuous analyte monitoring system 104, a display device 107 that executes an application 106, a decision support engine 114, a user database 110, a medical record database 112, a training server system 140, and a decision support engine 114, each of which is described in more detail below.

[0049] As used herein, the term "analyte" is a broad term used in its ordinary meaning that includes, but is not limited to, substances or chemical components in a biological fluid that can be analyzed (e.g., blood, interstitial fluid, cerebrospinal fluid, lymph, or urine). The sample can include naturally occurring substances, artifacts, metabolites, and / or reaction products. Analytes that can be measured by the present device and method include, but are not limited to, potassium, glucose, acarboxyprothrombin; acylcarnitine; adenine phosphoribosyltransferase; adenosine deaminase; albumin; α-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-β-hydroxy-cholic acid; cortisol; creatine kinase; creatine kinase MM isoenzyme; cyclosporin A; cystatin C; d-penicillamine; deethylchloroquine; dehydroepiandrosterone sulfate; DNA (acetylation polymorphism, alcohol dehydrogenase, alpha1-antitrypsin, glucose-6-phosphate dehydrogenase, hemoglobin A, hemoglobin S, hemoglobin C, hemoglobin D, hemoglobin E, hemoglobin F, D-Punjab, hepatitis B virus, HCMV, HIV-1, HTLV-1, MCAD, RNA, PKU, Plasmodium vivax, 21-deoxycortisol); desbutylhalofantrine; dihydropteridine reductase; diphtheria / tetanus antitoxin; erythrocyte arginase; erythrocyte protoporphyrin; esterase D; fatty acid / acylglycine; free beta-human chorionic gonadotropin; free erythrocyte porphyrin; free thyroxine (FT4); free tri-iodothyronine (FT3); fumarylacetoacetase; galactose / gal-1-phosphate;Galactose-1-phosphate uridyltransferase; Gentamicin; Glucose-6-phosphate dehydrogenase; Glutathione; Glutathione peroxidase; Glycolic acid; Glycosylated hemoglobin; Halofantrine; Hemoglobin variant; Hexosaminidase A; Human erythrocyte carbonic anhydrase I; 17-α-hydroxyprogesterone; Hypoxanthine phosphoribosyltransferase; Immunoreactive trypsin; Lactate; Lead; Lipoprotein ((a), B / A-1, β); Lysozyme; Mefloquine; Netilmicin; Phenobarbital; Phenytoin; Phytanic acid / pristanic acid; Progesterone; Prolactin; Prolidase; Purine nucleoside phosphorylase; Quinine; Reverse tri-iodothyronine (rT3); Selenium; Serum pancreatic lipase; Sisomicin; Somatomedin C; Specific antibodies that recognize any one or more of the following (Adenovirus, Antinuclear antibody, Anti-zeta antibody, Arbovirus, Pseudorabies virus, Dengue virus, Medina worm, Taenia solium, Entamoeba histolytica, Enterovirus, Giardia disease, Helicobacter pylori, Hepatitis B virus, Herpes virus, HIV-1, IgE (Atopic disease), Influenza virus, Donovan Leishmania, Leptospira, Measles / mumps / rubella, Mycoplasma pneumoniae, Myoglobin, Trichinella spiralis, Parainfluenza virus, Plasmodium malariae, Poliovirus, Pseudomonas aeruginosa, Respiratory syncytial virus, Rickettsia (Scrub typhus), Schistosoma mansoni, Toxoplasma, Treponema pallidum, Trypanosoma cruzi / Langeli, 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;Other specimens, including but not limited to those containing zinc protoporphyrin, are similarly contemplated. Salts, sugars, proteins, fats, vitamins, and hormones that naturally exist in blood or interstitial fluid can also constitute analytes in certain implementations. Ions are charged atoms or compounds that can include, for example, sodium, potassium, calcium, chloride, nitrogen, or bicarbonate. Analytes can naturally exist in biological fluids, such as metabolites, hormones, antigens, antibodies, ions, etc. Alternatively, analytes can be introduced into the body or 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 an 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, chlorinated hydrocarbons, hydrocarbons); cocaine (crack cocaine); stimulants (amphetamine, methamphetamine, Ritalin, Cylert, Preludin, Didrex, Prestate, Bolarnyl, Sandrex, Pregan); antidepressants (barbiturates, methaqualone, Valium, Librium, Miltown, Serax, Equanil, Tranxene, etc., which are tranquilizers); hallucinogens (fenciclonine, lysergic acid, mescaline, peyote, psilocybin); narcotics (heroin, codeine, morphine, opium, meperidine, Percocet, Percodan, Tussionex, fentanyl, Darvon, Talwin, Lomotil); synthetic narcotics (fentanyl, meperidine, amphetamine, methamphetamine, and analogs of fenciclonine, such as ecstasy); anabolic steroids;and may be drugs or pharmaceutical compositions including, but not limited to, nicotine. Metabolites of drugs and pharmaceutical compositions are also intended analytes. For example, analytes such as ascorbic acid, uric acid, dopamine, norepinephrine, 3-methoxytyramine (3MT), 3,4-dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA), 5-hydroxytryptamine (5HT), and 5-hydroxyindoleacetic acid (5HIAA), as well as other chemical substances generated in the body such as intermediates of the citric acid cycle, can also be analyzed.;

[0050] Analytes measured and analyzed by the devices and methods described herein include potassium, and optionally glucose, creatinine, urea (blood urea nitrogen (BUN)), inulin, dextran, saccharin, iothalamic acid, iohexol, 125I-iothalamate, cystatin C, C-peptide, 51Cr-EDTA, lactate, aspartic acid, polyfructosan, and betaine, although other analytes listed above may also be considered and measured, for example, by analyte monitoring system 104.

[0051] In certain embodiments, the continuous analyte monitoring system 104 is configured to continuously measure one or more analytes and transmit analyte measurements to an electronic medical records (EMR) system (not shown in FIG. 1). The EMR system is a software platform that enables the electronic input, storage, and maintenance of digital medical data. The EMR system is generally used throughout hospitals and / or other care facilities to document clinical information about patients over a long period of time. The EMR system organizes and presents data in a way that, for example, aids clinicians by interpreting health status and providing ongoing care, scheduling, billing, and follow-up. Reports for patient clinical care and / or disease management can also be created using data contained in the EMR system. In certain embodiments, the EMR may communicate (e.g., via a network) with the decision support engine 114 to perform the techniques described herein. In particular, as described herein, the decision support engine 114 can obtain data associated with a user, use the obtained data as input to one or more trained models, and output a prediction. In some cases, the EMR may provide the decision support engine 114 with data that is used as input to one or more models, such as an ML model. Further, in some cases, after making a prediction, the decision support engine 114 may provide the output prediction to the EMR.

[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 a display device 107 for use by an 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 smartwatch, 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 who have an interest in the patient's health (e.g., family members or physicians 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.

[0053] The application 106 is a mobile health application configured to receive and analyze analyte measurements from the analyte monitoring system 104. In particular, the application 106 stores information about the user, including the user's analyte measurements, in the user's user profile 118 for use by a decision support engine 114 for processing and analysis and for providing decision support recommendations or guidance to the user.

[0054] The decision-making 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-making 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-making support engine 114 via a network (e.g., the Internet). In some other embodiments, the decision-making 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-making support engine 114 executes entirely on one or more local devices such as the display device 107. As will be discussed in more detail herein, the decision-making support engine 114 may provide decision-making support recommendations to the user via the application 106. The decision-making support engine 114 provides decision-making support recommendations based on the information included in the user profile 118.

[0055] The user profile 118 may include information collected regarding the user from the application 106. For example, the application 106 provides a set of inputs 128 that includes analyte measurements received from the continuous analyte monitoring system 104 and stored within 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 can obtain additional inputs 128 via manual user input, one or more other non-analyte sensors or devices, other applications running on the display device 107, and the like. 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 respiratory sensor, a thermometer, a peritoneal dialysis machine, a hemodialysis machine, 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 smartwatch), or any other sensor or device that provides related information regarding the user. The inputs 128 of the user profile 118 provided by the application 106 are described in further detail below with respect to FIG. 3.

[0056] The DAM 116 of the decision-making support engine 114 is configured to process a 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, can, in at least some cases, indicate the user's health or condition, such as one or more of the user's physiological state, trends associated with the user's health or condition, and the like. In certain embodiments, the metrics 130 can then be used by the decision-making support engine 114 as an input for providing guidance to the user. As shown, the metrics 130 are also stored in the user profile 118. In certain embodiments, the user's glucose can include glucose level, timestamped glucose level, glucose rate of change, glucose trend, average glucose level, glucose management indicator (GMI), glycemic variability, time in range (TIR), glucose clearance rate, minimum and maximum glucose levels, glucose autocorrelation score, glucose setpoint, and the like.

[0057] 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 obtained from a particular data storage device (e.g., an electronic medical record (EMR), etc.). In certain embodiments, the demographic information 120 can include one or more of the user's age, body mass index (BMI), ethnicity, gender, and the like. In certain embodiments, the disease progression information 122 can include information about the user's disease, such as whether the user has acute kidney injury (AKI), the user's exposure to the risk of developing AKI (e.g., myocardial infarction, rhabdomyolysis, sepsis or other infectious diseases, hypoperfusion due to blood loss, or other diseases of the kidney), or chronic kidney disease (CKD), or whether the user has previously been diagnosed with hyperkalemia, hypokalemia, hyperglycemia, hypoglycemia, and the like. In certain embodiments, the information about the user's disease can also include the length of time after diagnosis, the level of disease control, the level of compliance with disease management therapy, predicted kidney function, other types of diagnoses (e.g., heart disease, obesity), or health metrics (e.g., heart rate, exercise, stress, sleep, etc.).

[0058] In certain embodiments, the medication 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 timestamped and correlated with the user's analyte levels, thereby indicating the impact of the amount, frequency, and type of medication on the user's analyte levels. In certain embodiments, the medication information may include information regarding the consumption of one or more diuretics. Diuretics can be prescribed to patients, for example, for the purpose of treating excessive fluid accumulation caused by congestive heart failure (CHF), liver failure, and / or nephrotic syndrome. For example, CHF is a condition in which the heart is unable to pump blood efficiently to meet the body's oxygen and nutritional requirements. When the heart cannot pump blood efficiently, normal blood circulation is impaired, leading to excessive fluid in the blood. The excessive fluid leaks from the blood vessels and accumulates in the lungs and other tissues. Thus, patients may be prescribed diuretics to help the kidneys flush out the excessive fluid and maintain a normal blood volume.

[0059] The different types of diuretics prescribed to patients can include loop diuretics, thiazide and thiazide-like diuretics, and potassium-sparing diuretics. Loop diuretics inhibit proteins found in a part of the nephron known as the loop of Henle. Loop diuretics can include, for example, furosemide (Lasix), bumetanide (Bumex), torsemide (Demadex), ethacrynic acid (Edecrin). Thiazide diuretics are commonly used to treat hypertension (high blood pressure), but are also used to manage heart failure. Thiazide diuretics inhibit a different protein than that inhibited by loop diuretics, which also aids in mineral reabsorption. Thiazide diuretics can include, for example, chlorothiazide (Diuril), hydrochlorothiazide (Hydrodiuril), metolazone (Zytonix). Finally, potassium-sparing diuretics (such as spironolactone, etc.) are weak diuretics used to increase the amount of fluid passing from the body into the urine, and at the same time prevent excessive potassium from being lost from the body into the urine. As described in more detail below, the decision support system 100 can be configured to use medical information 124 to determine the optimal diuretic to be prescribed to different users. In particular, the decision support system 100 can be configured to identify one or more optimal diuretics for prescription based on the health of the patient when one or more diuretics are prescribed, as well as the condition of the patient being treated.

[0060] In certain embodiments, the pharmaceutical information may include information regarding the consumption of one or more drugs known to damage the kidneys. One or more drugs known to damage the kidneys may include nonsteroidal anti-inflammatory drugs (NSAIDs), such as ibuprofen (e.g., Advil, Motrin) and naproxen (e.g., Aleve), vancomycin, iodinated radiocontrast agents (e.g., referring to any contrast agent used in diagnostic tests), angiotensin-converting enzyme (ACE), such as lisinopril, enalapril, and ramipril, aminoglycoside antibiotics, such as neomycdin, gentamicin, tobramycin, and amikacin, anti-viral human immunodeficiency virus (HIV) pharmaceuticals, zoledronic acid (e.g., Zometa, Reclast), foscarnet, lithium, and the like.

[0061] In certain embodiments, the pharmaceutical information may include information regarding the consumption of one or more drugs known to control complications of kidney disease. One or more drugs known to control complications of kidney disease may include pharmaceuticals for lowering blood pressure and maintaining kidney function, such as ACE inhibitors or angiotensin II receptor blockers, pharmaceuticals for treating anemia, such as supplements of the hormone erythropoietin, pharmaceuticals used to lower cholesterol levels, such as statins, pharmaceuticals used to prevent weak bones, such as calcium and vitamin D supplements, phosphate binders, and the like.

[0062] In certain embodiments, 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, such that the user profile 118 is dynamic. Accordingly, the information in the user profile 118 stored within the user database 110 provides the most up-to-date repository of information related to the user.

[0063] In some embodiments, the user database 110 refers to a storage server operating in a public or private cloud. The user database 110 can be implemented as any type of data store, such as a relational database, a non-relational database, a key-value data store, or a file system including a hierarchical file system. In some exemplary implementations, the user database 110 is distributed. For example, the user database 110 can include a plurality of distributed persistent storage devices. Further, the user database 110 can be replicated such that the storage devices are geographically distributed.

[0064] The user database 110 includes user profiles 118 associated with a plurality of users that interact in a manner similar to the application 106 running on the display device 107 of other users. The user profiles stored in the user database 110 are accessible not only to the application 106 but also to the decision support engine 114. The user profiles in the user database 110 can be accessible to the application 106 and the decision support engine 114 via one or more networks (not shown). As described above, the decision support engine 114, more specifically the DAM 116 of the decision support engine 114, can fetch the input 128 from the user database 110 and then calculate a plurality of metrics 130 that can be stored as application data 126 in the user profile 118.

[0065] In certain embodiments, the user profile 118 stored in the user database 110 may also be stored in the medical record database 112. The user profile 118 stored in the medical 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 record database 112 essentially provides all data related to each user of the application 106, and the data is stored according to an associated timestamp. The timestamp associated with the information stored within the medical record database 112 can identify, for example, when information related to the user was acquired and / or updated.

[0066] Furthermore, the medical record database 112 can maintain time-series data collected about a user over a period of time, including that for users of the continuous analyte monitoring system 104 and the application 106. For example, analyte data for a user who used the continuous analyte monitoring system 104 and the application 106 for five years to manage the user's kidney condition may have time-series analyte data associated with the user maintained over the five-year period.

[0067] Moreover, in certain embodiments, the medical record database 112 may include data for one or more patients who are not users of the continuous analyte monitoring system 104 and / or the application 106. For example, the medical record database 112 may include information (e.g., user profiles) about one or more patients who have not previously been diagnosed with a kidney disease (e.g., CKD), as analyzed by, for example, a health management physician (or other known method), and information (e.g., user profiles) about one or more patients who have previously been diagnosed with kidney diseases (of various types and stages), as analyzed by, for example, a health management physician (or other known method). The data stored in the medical record database 112 may be referred to herein as population data.

[0068] The data associated with each patient stored in the medical record database 112 can provide time series data collected over the patient's disease lifetime. For example, the data can include information about the patient before being diagnosed with a kidney disease and information about the patient during the lifetime of the disease, information about each stage of the kidney disease (e.g., CKD) that has progressed and / or regressed in the patient, and information about other diseases or conditions such as hyperkalemia, hypokalemia, diabetes, heart conditions and diseases, or similar diseases coexisting with the kidney disease. Such information can indicate, over the lifetime of the disease, the patient's symptoms, the patient's physiological state, the patient's potassium level, the patient's glucose level, the patient's creatinine level, the patient's BUN level, the patient's cystatin C level, the patient's C-peptide level, the patient's albumin level, the patient's creatinine level, the patient's inulin level, the patient's dextran level, the patient's saccharin level, the patient's iothalamic acid level, the patient's iohexol level, the patient's 125l-iothalamate level, the patient's 51Cr-EDTA level, the patient's lactate level, the patient's aspartic acid level, the patient's polyfructosan level, the patient's betanin level, the state / condition of one or more organs of the patient, the patient's habits (e.g., activity level, food consumption, etc.), the medications prescribed, etc.

[0069] Although depicted as separate databases for clarity of concept, in some embodiments, the user database 110 and the medical record database 112 can operate as a single database. That is, the historical and current data associated with the users of the continuous analyte monitoring system 104 and the application 106, and the historical data associated with patients who were not previous users of the continuous analyte monitoring system 104 and the application 106 can be stored in a single database. The single database can be a storage server operating in a public or private cloud.

[0070] As described above, the decision support system 100 is configured to screen, diagnose, and stage a user's kidney disease using a continuous analyte monitoring system 104 that includes one or more analyte sensors. In certain embodiments, the continuous analyte monitoring system 104 includes at least a continuous potassium monitor (CPM). In certain embodiments, the decision support engine 114 provides real-time and / or non-real-time decision support regarding kidney disease to a user and / or others, including but not limited to healthcare providers, the user's family, the user's caregivers, researchers, artificial intelligence (AI) engines, and / or other individuals, systems, and / or groups that assist in care or learning from the data. In particular, the decision support engine 114 can be used to collect information related to the user within the user profile 118 stored in the user database 110, predict the presence and / or severity of the user's kidney disease, and / or predict the likelihood that the user will develop a kidney disease within a certain period of time, and provide one or more recommendations for treatment based at least in part on the prediction. The user profile 118 can be accessible to the decision support engine 114 via one or more networks (not shown) for performing such analysis.

[0071] In certain embodiments, the decision support system 100 is designed to predict the risk or likelihood of kidney disease, or the presence and / or severity of kidney disease, in real time (including near real time) or within a specified period of time for the patient. In certain embodiments, to enable such prediction, the decision support engine 114 collects information related to the user within the user profile 118 stored in the user database 110 and is configured to (1) automatically detect and classify abnormal kidney function; (2) evaluate the risk of kidney disease; and (3) perform an analysis to evaluate the presence and stage of kidney disease.

[0072] In certain embodiments, the decision support engine 114 can utilize one or more trained machine learning models that can determine the probability of the presence and / or severity of a renal disease in a user based on information provided by the user profile 118. In the illustrated embodiment of FIG. 1, the decision support engine 114 can utilize a trained machine learning model provided by the training server system 140. Although depicted as a separate server for clarity of concept, in some embodiments, the training server system 140 and the decision support engine 114 may operate as a single server. That is, the model may be trained and used by a single server, or may be trained by one or more servers and deployed for use on one or more other servers. In certain embodiments, the model can be trained on one or more virtual machines (VMs) that are at least partially executed on one or more physical servers in a relational and / or non-relational database format.

[0073] The training server system 140 is configured to train a machine learning model using training data, which can include data associated with one or more patients previously diagnosed with various stages of renal disease (e.g., users or non-users of the continuous analyte monitoring system 104 and / or the application 106), as well as data associated with patients not previously diagnosed with renal disease (e.g., healthy patients). The training data can be stored in the medical record database 112 and can be accessible to the training server system 140 via one or more networks (not shown) for training the machine learning model. The training data may also, in some cases, include user-specific data for a user over time.

[0074] Training data refers to a dataset that is characterized and labeled. For example, the dataset may include a plurality of data records, each of which contains 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, features are individual measurable properties or characteristics. Generally, the features that best characterize the patterns in the data are selected for creating a predictive machine learning model. Data labeling is the process of adding one or more meaningful and useful labels to the data to provide context for learning by a machine learning model.

[0075] As an illustrative example, each relevant feature of a user reflected in the corresponding data record can be a feature used when training a machine learning model. Such features can include age, gender, change (e.g., delta) in analyte level (e.g., potassium level) from a first timestamp to a second timestamp, change (e.g., delta) in kidney disease stage or severity from a first timestamp to a second timestamp, change (e.g., delta) in analyte threshold (e.g., potassium threshold) of a user suffering from kidney disease from a first timestamp to the next timestamp, derivative of a linear system of analyte measurements (e.g., potassium measurement) taken at a specific timestamp, and / or difference in derivatives for determining the rate of change of the slope of increase or decrease in analyte value (e.g., potassium value). Further, the data record is labeled with an indication regarding the kidney disease diagnosis, assigned disease severity, and / or identified risk of kidney disease related to the patient of the user profile.

[0076] Next, the model is trained by the training server system 140 using the characterized and labeled training data. In particular, the features of each data record can be used as input to the machine learning model, and the generated output can be compared to the label associated with the corresponding data record. The model can 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 repeatedly processing each data record corresponding to each medical record patient, in certain embodiments, the model can be repeatedly refined to generate accurate predictions related to the risk, presence, progression, improvement (e.g., regression), and severity of kidney disease in the patient. Further, in certain other embodiments, by repeatedly processing each data record corresponding to each medical record patient, in certain embodiments, the model can be repeatedly refined to generate accurate predictions of the risk and / or presence of CKD.

[0077] 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 can obtain a user profile 118 associated with the user, use the information within the user profile 118 as input to the trained model, and output a prediction. The prediction can indicate the presence and / or severity of the user's kidney disease in real time or within a specific time (e.g., as shown as output 144 in FIG. 1). The output 144 generated by the decision support engine 114 can also provide one or more recommendations regarding treatment based on the prediction. The output 144 can be provided to the user (e.g., through 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 for the purpose of improving the user's health, such as by achieving the recommended treatment.

[0078] 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, to predict the presence and / or severity of a kidney disease in a particular user, a model (e.g., trained using population data) can be deployed for use by the decision support engine 114. Sometime after making a prediction using the model, the decision support engine 114 may be configured to ask the user or a caregiver, physician, etc., whether the predicted presence and / or severity of the kidney disease was confirmed, for example, by other diagnostic methods, and / or the decision support engine 114 may use one or more diagnostic tests to confirm the diagnosis. In some cases, the user's response and / or the results of the diagnostic tests performed can refute the presence of the kidney disease. Accordingly, the model can continue to be retrained and / or personalized using the user's response, diagnostic test results, and / or the user's physiological parameters used as input to the model for personalizing the model for the user.

[0079] 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 patient-specific treatment decision or recommendation for preventing the occurrence of one or more kidney disease predictions. For example, in certain embodiments, the output 144 may be a prediction regarding the presence and / or severity of a user's chronic kidney disease (CKD). In certain embodiments, the output 144 may be a prediction regarding the risk that the user has CKD. In certain embodiments, the output 144 may be a prediction regarding the risk that the user has hyperkalemia and / or hypokalemia. In certain embodiments, the output 144 may be a prediction regarding the patient's risk of death. In certain embodiments, the output 144 may be a patient-specific treatment decision or a recommendation for the patient's CKD. In still further embodiments, the output 144 may be a prediction regarding the presence and / or severity of acute kidney injury (AKI), a prediction regarding the risk that the user has AKI, a patient-specific treatment decision or a recommendation for AKI for the patient, and the like. The output 144 stored in the user profile 118 may be continuously updated by the decision support engine 114. Accordingly, the user's previous diagnoses and / or physiological parameters related to kidney disease are initially stored in the user database 110 as the output 144 of the user profile 118 and then passed to the medical record database 112, which can provide an indicator of the progression of the user's CKD / AKI (or other type of kidney disease) over time, and can also provide an indicator regarding the effectiveness of different treatments (e.g., medications) recommended to the user to help stop the progression of the disease.

[0080] In certain embodiments, the user's own historical data can be used to provide decision-making support and insights regarding the user's renal function and / or disease. For example, the user's historical data can 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 two weeks ago can be used as a baseline that can be compared with 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 can be used by the training server system 140 to train a personalized model that can further predict or estimate the future improvement / deterioration of the user's renal function or kidneys based on the user's recent data patterns (e.g., exercise data, food consumption data, etc.).

[0081] In certain embodiments, the model is trained to provide recommendations for lifestyle, exercise, food intake, medications, and other types of decision-making support recommendations so that the user can prevent the onset and / or progression of kidney disease, treat symptoms, and utilize the user's historical data, including how various types of medications, foods, and treatments (e.g., dialysis, etc.) have affected the user's renal function in the past, to improve kidney health and function. In certain embodiments, the model can be trained to predict the causes underlying a particular improvement or deterioration in the patient's renal function. For example, the application 106 can display a user interface with a graph showing the patient's renal function or its score along with a trend line and, for example, retrospectively show what caused the function of the kidneys at a particular point in time (e.g., excessive potassium intake, intake of non-potassium-sparing diuretics, etc.).

[0082] FIG. 2 is FIG. 200 conceptually illustrating an exemplary continuous analyte monitoring system 104 including an exemplary continuous analyte sensor having 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.

[0083] Generally, changes in renal function (e.g., either dysfunction or improvement) can be indicated using real-time or continuous measurements of analyte levels, rates of change, trends, clearance rates, and / or other analyte data measured in interstitial fluid or blood by a continuous analyte monitoring system. Such data can indicate changes in renal function well in advance of conventional renal disease diagnostic tools. Accordingly, continuous analyte monitoring can provide earlier and / or improved screening, diagnosis, prognosis, and / or staging of renal disease compared to conventional diagnosis.

[0084] In certain embodiments, clinical indicators can be used to determine whether a continuous analyte monitoring system, such as continuous analyte monitoring system 104, is needed to evaluate a patient's risk, presence, and / or stage of renal disease. In one example, such clinical indicators include GFR tests that include mGFR and eGFR measurements. GFR tests can generally indicate the presence, risk, or likelihood of renal dysfunction, and thus, after receiving a GFR test, confirmation of the GFR results or screening for renal dysfunction by continuous analyte monitoring may be desirable.

[0085] In yet another example, the clinical indicator can include annual screening. For example, in a patient's annual screening, regardless of the presence of other risk factors for renal disease, it may be desirable to initiate screening for renal dysfunction using a continuous analyte monitor. In certain cases, a patient can be prescribed to use a continuous analyte monitor to screen for renal disease over a period of time (e.g., 2 to 4 weeks or more).

[0086] In yet another example, the clinical indicator can include previous symptoms of renal dysfunction. In the case of a patient who has experienced or has experienced symptoms of renal dysfunction, it may be desirable to initiate screening for renal dysfunction using a continuous analyte monitor. In another example, the clinical indicator can include previous symptoms of potassium imbalance. Generally, symptoms of potassium imbalance can include numbness / tingling, shortness of breath, chest pain, muscle weakness, etc.

[0087] In yet another example, the clinical indicator may include a prescribed or taken medicine. In the case of a patient receiving a specific medicine related to renal dysfunction or a medicine related to the cause of kidney injury, it may be desirable to monitor and / or screen for renal dysfunction. For example, a patient taking a medicine known to cause kidney injury can use a continuous analyte monitor to screen for renal dysfunction that may have been caused by the medicine.

[0088] In yet another example, the clinical indicator may include comorbidities that are often associated with renal dysfunction and / or increase the risk of renal dysfunction. Comorbidities related to kidney disease include cardiovascular disease, obesity, liver disease, hypertension, and / or diabetes.

[0089] In yet another example, the clinical indicator may include patient risk factors that can increase the risk of patients with renal dysfunction and / or disease. Such risk factors for kidney disease may include age, a record of low birth weight, and / or a family history of kidney disease.

[0090] In yet another example, the clinical indicator may include physiological parameters that can increase the risk of patients with renal dysfunction. Such physiological parameters include abnormal potassium levels (e.g., from blood measurements), CPM data indicating that further screening, diagnosis, and / or staging of kidney disease is desirable, and / or CGM data indicating that screening, diagnosis, and / or staging of kidney disease is desirable.

[0091] In yet another example, the clinical indicator may include adverse health events that can increase the risk of patients with renal dysfunction. Such adverse health events may include hyperkalemia, hypokalemia, cardiac events, hyperglycemia, and / or hypoglycemia.

[0092] In certain embodiments, the continuous analyte monitoring system 104 may be utilized as a short-term diagnostic tool (i.e., 10 - 14 days) for screening, diagnosing, and / or staging patients with kidney disease. For example, a trigger event (e.g., a trigger while the analyte sensor is being worn) can indicate the usefulness of a patient wearing an analyte sensor (continuously or discontinuously) for a short period of time to provide screening, diagnosis, and / or staging of kidney disease. In one example, a patient can wear a short-term continuous or discontinuous analyte sensor on the body annually to screen for kidney disease. In another example, a patient can wear a short-term continuous or discontinuous analyte sensor periodically (e.g., every 4 weeks) to monitor for the presence / stage of kidney disease. In yet another example, a patient can wear a short-term continuous or discontinuous analyte sensor to determine the stage of kidney disease.

[0093] In yet another example, a patient can wear a short-term continuous or discontinuous analyte sensor to confirm and / or provide additional data for the clinical diagnosis of kidney disease, or to rule out a potential clinical kidney disease diagnosis. For example, if a patient's clinical creatinine levels may indicate a potential for kidney dysfunction, the patient can wear a short-term continuous or discontinuous analyte sensor to monitor creatinine levels (or other analyte levels) and confirm kidney disease. In another example, a patient can wear a short-term continuous or discontinuous analyte sensor to monitor risk factors associated with kidney disease, such as glucose imbalance, e.g., hypoglycemia, or potassium imbalance, e.g., hyperkalemia. In another example, a patient can wear a short-term continuous or discontinuous analyte sensor during an mGFR test to provide additional insight into the mGFR results.

[0094] In yet another example, a patient can wear a continuous analyte sensor, such as a CPM, whereby the monitor functions periodically as a short-term monitor for screening, diagnosing, and / or staging patients having kidney disease. For example, a diabetic patient using a continuous analyte monitor capable of sensing both glucose and potassium can utilize the continuous potassium monitoring function as a short-term monitoring tool for kidney disease.

[0095] In certain embodiments, the continuous analyte monitoring system 104 can be utilized as a long-term diagnostic tool (i.e., greater than 14 days) for screening, diagnosing, and / or staging patients having kidney disease. For example, patients at high risk of kidney disease and / or adverse events can utilize a continuous analyte sensor to continuously screen, diagnose, and / or stage a patient's kidney disease over weeks, months, etc. For example, a patient having stage 3 kidney disease can utilize a CPM to monitor the kidney disease and / or show deterioration or improvement of kidney function over a long period of time.

[0096] Returning now to FIG. 2, the continuous analyte monitoring system 104 in the illustrated embodiment includes a sensor electronics module 204 and one or more continuous analyte sensors 202 associated with the sensor electronics module 204 (referred to herein individually as continuous analyte sensors 202 and collectively as continuous analyte sensors 202). The sensor electronics module 204 can 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 can also wirelessly communicate (e.g., directly or indirectly) with one or more medical devices, such as medical device 208 (referred to herein individually as medical device 208 and collectively as medical devices 208), and / or one or more other non-analyte sensors 206 (referred to herein individually as non-analyte sensors 206 and collectively as non-analyte sensors 206).

[0097] In certain embodiments, the continuous analyte sensor 202 may include sensors for detecting and / or measuring analytes. The continuous analyte sensor 202 may be a multi-analyte sensor configured to continuously measure two or more analytes, or a single-analyte sensor configured to continuously measure a single analyte as a non-invasive device, a subcutaneous device, a transdermal device, a transdermal device, and / or an intravascular device. In certain embodiments, the continuous analyte sensor 202 may be configured to continuously measure the user's analyte levels using one or more measurement techniques such as enzymatic, chemical, physical, electrochemical, spectrophotometric, polarimetric, calorimetric, electrophoretic, radiometric, immunochemical, etc. In certain aspects, the continuous analyte sensor 202 provides a data stream indicative of the concentration of one or more analytes within the user. The data stream may then include raw data signals that are converted into a calibrated and / or filtered data stream used to provide an estimated analyte value to the user.

[0098] In certain embodiments, the continuous analyte sensor 202 may be a multi-analyte sensor configured to continuously measure multiple analytes within the 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), cystatin C, C-peptide, albumin, inulin, dextran, saccharin, iothalamic acid, iohexol, 125l-iothalamate, 51Cr-EDTA, aspartic acid, polyfructosan, and / or betaine within the user's body.

[0099] 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, for example, BUN levels or lactate levels. Information from each of the multi - analyte sensor and the single - analyte sensor can be combined to provide kidney disease decision - making support using the methods described herein.

[0100] In certain embodiments, the sensor electronics module 204 includes electronic circuitry associated with the measurement and processing of continuous analyte sensor data, including predictive algorithms associated with the processing and calibration of sensor data. The sensor electronics module 204 can be physically connected to the continuous analyte sensor 202 and can be integrated (non - removably attached) with or removably attached to the continuous analyte sensor 202. The sensor electronics module 204 can include hardware, firmware, and / or software that enables the measurement of levels on a display device via the continuous analyte sensor 202. For example, the sensor electronics module 204 can include a potentiostat, a power source for powering 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 fixed to, for example, a printed circuit board (PCB) and can take various 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.

[0101] The display devices 210, 220, 230, and / or 240 are configured to display displayable sensor data, including analyte data, that can be transmitted by the sensor electronics module 204. Each of the display devices 210, 220, 230, or 240 can include a display such as a touch screen display 212, 222, 232, and / or 242 for displaying sensor data to the user and / or receiving input from the user. For example, a graphical user interface (GUI) can be presented to the user for such purposes. In some embodiments, the display device can include other types of user interfaces, such as a voice user interface, instead of or in addition to the touch screen display, for communicating sensor data to the user of the display device and / or receiving user input. The display devices 210, 220, 230, and 240 can be examples of the display device 107 illustrated in FIG. 1 that are used to display sensor data to the user of FIG. 1 and / or receive input from the user.

[0102] In some embodiments, one, some, or all of the display devices are configured to display or otherwise communicate the sensor data when the sensor data is communicated from the sensor electronics module (e.g., in the data packages transmitted to each respective display device) without any additional expected processing that may be required for calibration and real-time display of the sensor data.

[0103] The plurality of display devices can include custom display devices specially designed to display a particular type of displayable sensor data associated with the analyte data received from the sensor electronics module. In certain embodiments, the plurality of 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 plurality of display devices is a smartphone, such as display device 220 representing a mobile phone, configured to use a commercially available operating system (OS) to display graphical representations of continuous sensor data (e.g., including current and historical data). Other display devices can include a display device 230 representing a tablet, a display device 240 representing a smartwatch, a medical device 208 (e.g., an insulin delivery device or a blood glucose meter), and / or a desktop or laptop computer (not shown).

[0104] Because the user interface varies depending on the display device, the content of the data package (e.g., the amount, format, and / or type of data to be displayed, alarms, etc.) can be customized for each specific display device (e.g., programmed to vary depending on the manufacturer and / or end user). Thus, in certain embodiments, multiple different display devices can communicate wirelessly directly with a sensor electronics module (such as, for example, the on-skin sensor electronics module 204 physically connected to the continuous analyte sensor 202) 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 type of alarm customized for each specific display device, the number of alarms customized for each specific display device, the timing of the alarms customized for each specific display device, and / or the threshold levels configured (e.g., for triggering) for each of the alarms are based on the output 144 stored in the user profile 118 for each user (e.g., as noted, the output 144 can indicate the user's current health, the user's kidney condition, the current treatment recommended for the user, and / or the user's physiological parameters when experiencing different stages of kidney disease).

[0105] As noted, the sensor electronics module 204 can communicate with the medical device 208. The medical device 208 can be a passive device in some exemplary embodiments of the present disclosure. For example, the medical device 208 can be an insulin pump for administering insulin to the 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.

[0106] Furthermore, as mentioned, the sensor electronics module 204 may also communicate with other non-analyte sensors 206. The non-analyte sensors 206 may include, but are not limited to, altimeter sensors, accelerometer sensors, temperature sensors, respiration rate sensors, sweat sensors, etc. The non-analyte sensors 206 may also include monitors such as heart rate monitors, ECG monitors, blood pressure monitors, pulse oximeters, calorie intake, and drug delivery devices. The non-analyte sensors 206 may also include a data system for measuring non-patient-specific phenomena such as time, ambient pressure, or ambient temperature, and may include an atmospheric pressure sensor, an outside air temperature sensor, or a clock, timer, or other time measurement value when the sensor was first inserted, or a measurement of the remaining sensor life compared to the insertion time that can be used as calibration for an algorithm model or other data input. One or more of these non-analyte sensors 206 may provide data to the decision support engine 114 further described below. In some embodiments, the 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. 1.

[0107] In certain embodiments, the non-analyte sensors 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 used to transmit sensor data to the sensor electronics module 204 using a common communication 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 used to transmit sensor data to the sensor electronics module 204 using a common communication circuit.

[0108] In certain embodiments, a wireless access point (WAP) may be used to couple one or more of the continuous analyte monitoring system 104, the plurality of display devices, the medical device 208, and / or the non-analyte sensor 206 to each other. 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 among the devices depicted in FIG. 200 of FIG. 2.

[0109] FIG. 3 shows a diagram 300 of exemplary inputs and exemplary metrics calculated based on inputs for use by the decision support system of FIG. 1, according to some embodiments disclosed herein. In particular, FIG. 3 provides a more detailed illustration of the exemplary inputs and exemplary metrics introduced in FIG. 1.

[0110] FIG. 3 shows exemplary input 128 on the left, decision support engine 114 including application 106 and DAM 116 in the center, and metric 130 on the right. In certain embodiments, each of the metrics 130 may correspond to one or more values, such as discrete numerical values, ranges, or qualitative values (e.g., high / medium / low, or stable / unstable). The application 106 obtains the input 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 input 128 may be processed by the DAM 116 and / or the decision support engine 114 to output the metric 130. The input and metric 130 may be used by the decision support engine 114 to provide decision support to a user. For example, the input 128 and metric 130 may be used by the training server system 140 to train and deploy one or more machine learning models for use by the decision support engine 114 to provide decision support regarding kidney disease.

[0111] In some embodiments, starting from the input 128, the food consumption information may include information about one or more of meals, snacks, and / or beverages, such as one or more of quantity, 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 photograph through an application configured to recognize the type and quantity of food, and / or by scanning a barcode or menu. In various examples, the meal quantity may be manually input as one or more of calories, quantity (e.g., "3 cookies"), menu item (e.g., "Royale with Cheese"), and / or food exchange (e.g., 1 fruit, 1 dairy product). In some examples, the meal information may be received via a convenient user interface provided by the application 106. In some examples, the meal information may be provided via one or more other applications synchronized with the application 106, such as one or more other mobile health applications executed by the display device 107. In such examples, the synchronized applications can include, for example, an electronic food diary application or a photo application.

[0112] In certain embodiments, the food consumption information input by the user may be related to the potassium consumed by the user. The potassium for consumption may include any natural or engineered food or beverage containing potassium, such as, for example, an energy drink, avocado, beans, banana, or potato. The food consumption information input by the user may also be related to other analytes, including any of the other analytes described herein.

[0113] In certain embodiments, motion information is also provided as an input. Motion information can be any information surrounding an activity, such as an activity that requires physical movement by the user. For example, motion information can relate to physical movement at low intensity (e.g., taking a few steps) and high intensity (e.g., running 5 miles). In certain embodiments, motion information can be provided by an accelerometer sensor on a wearable device such as, for example, a watch, a fitness tracker, and / or a patch. In certain embodiments, motion information can also be provided through manual user input and / or through a surrogate sensor and prediction algorithm that measures changes in heart rate (or other cardiac metrics). When the user predicts that they are exercising based on their sensor data, the user can be asked to confirm whether exercise is being performed, what type of exercise it is, and / or the level of intense exercise being used during the exercise over a particular period. This data can be used to train the system to learn about the user's exercise patterns in order 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 elements described herein, including time elements such as time and day.

[0114] 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 can also be provided as an input. In certain embodiments, user statistics can be provided through a user interface, by interfacing with an electronic information source such as an electronic medical record, and / or from a measurement device. In certain embodiments, the measurement device can include one or more of, for example, a wireless, e.g., Bluetooth-enabled, scale, and / or a camera that can communicate with the display device 107 to provide user data.

[0115] In certain embodiments, treatment / medical information is also provided as an input. The medical information may include information regarding the type, dosage, and / or timing when one or more medications are taken by the user. As referred to herein, the medical information may include one or more diuretics, one or more drugs known to reduce potassium levels, one or more drugs known to damage the kidneys, one or more drugs known to control the complications of acute or chronic kidney diseases prescribed to the user and / or information regarding one or more medications for treating one or more symptoms of acute or chronic kidney diseases, hyperkalemia, hypokalemia, diabetes, and / or other conditions and diseases 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 an increase or decrease in potassium intake, exercise for at least 30 minutes a day, or an increase in insulin dosage or other medications in order to maintain and / or improve kidney health and / or reduce episodes of hyperkalemia and / or hypokalemia. As another example, a healthcare 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 sweeps unwanted toxins, waste products, and excess fluid from the body by filtering the user's blood. A user having end-stage renal disease (ESRD) (e.g., CKD stage 5) may generally be prescribed dialysis treatment to complement and / or replace the filtration performed by the kidneys, considering that dialysis helps maintain the balance of potassium, phosphorus, and sodium levels in the patient's body. The dialysis treatment may be hemodialysis or peritoneal dialysis. In hemodialysis, the blood is pumped from the user's body to an artificial kidney machine and returned to the body through tubes that connect the user to the machine. In peritoneal dialysis, the inner lining of the user's abdomen acts as a natural filter. Thus, information regarding dialysis treatment for the user may be included in the treatment / medical information. In certain embodiments, the treatment / medical information may be provided through manual user input.

[0116] In certain embodiments, analyte sensor data may also be provided as input, for example, through a continuous analyte monitoring system 104. In certain embodiments, the analyte sensor data may include potassium data (e.g., a user's potassium value) 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 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 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 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) that is part of 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. In certain embodiments, the analyte sensor data may include inulin data measured by at least an inulin sensor (or a multi-analyte sensor configured to measure at least inulin) that is part of the continuous analyte monitoring system 104.In certain embodiments, the analyte sensor data can include dextran data measured by at least a dextran sensor (or a multi - analyte sensor configured to measure at least dextran) that is part of the continuous analyte monitoring system 104. In certain embodiments, the analyte sensor data can include saccharin data measured by at least a saccharin sensor (or a multi - analyte sensor configured to measure at least saccharin) that is part of the continuous analyte monitoring system 104. In certain embodiments, the analyte sensor data can include iotalamic acid data measured by at least an iotalamic acid sensor (or a multi - analyte sensor configured to measure at least iotalamic acid) that is part of the continuous analyte monitoring system 104. In certain embodiments, the analyte sensor data can include iohexol data measured by at least an iohexol sensor (or a multi - analyte sensor configured to measure at least iohexol) that is part of the continuous analyte monitoring system 104. In certain embodiments, the analyte sensor data can include 125I - iothalamate data measured by at least a 125I - iothalamate sensor (or a multi - analyte sensor configured to measure at least 125I - iothalamate) that is part of the continuous analyte monitoring system 104. In certain embodiments, the analyte sensor data can include 51Cr - EDTA data measured by at least a 51Cr - EDTA sensor (or a multi - analyte sensor configured to measure at least 51Cr - EDTA) that is part of the continuous analyte monitoring system 104. In certain embodiments, the analyte sensor data can include aspartic acid data measured by at least an aspartic acid sensor (or a multi - analyte sensor configured to measure at least aspartic acid) that is part of the continuous analyte monitoring system 104. In certain embodiments, the analyte sensor data can include polyfructosan data measured by at least a polyfructosan sensor (or a multi - analyte sensor configured to measure at least polyfructosan) that is part of the continuous analyte monitoring system 104.In certain embodiments, the analyte sensor data can include betaine data measured by at least a betaine sensor (or a multi-analyte sensor configured to measure at least betaine) that is part of the continuous analyte monitoring system 104.

[0117] In certain embodiments, the input can also be received from one or more non-analyte sensors, such as the non-analyte sensor 206 described with respect to FIG. 2. Inputs from such non-analyte sensors 206 can include information related to the user's heart rate, heart rate variability (e.g., the dispersion of time between heart beats), ECG data, respiratory rate, oxygen saturation, blood pressure, or body temperature (e.g., to detect illness, physical activity, etc.). In certain embodiments, the electromagnetic sensor can also detect a low-power radio frequency (RF) field emitted from an object or tool in contact with or in proximity to the object that can provide information about the user's activity or location.

[0118] In certain embodiments, the input received from the non-analyte sensor can include input related to the user's insulin delivery. In particular, input related to the user's insulin delivery can be received via a wireless connection on a smart pen, via user input, and / or from an insulin pump. The insulin delivery information can include one or more of insulin amount, delivery time, etc. Other configurations, such as insulin action time or duration of insulin action, can also be received as input.

[0119] In certain embodiments, the time can also be provided as input, such as the time of day or the time from a real-time clock. For example, in certain embodiments, the input analyte data can be timestamped to indicate the date and time when the analyte measurement was made for the user.

[0120] Any user input of the inputs 128 mentioned above can be through a user interface, such as the user interface of the display device 107 of FIG. 1.

[0121] As described above, in certain embodiments, DAM 116 and / or the decision support engine (e.g., using one or more trained models) determines or calculates a user's metric 130 based on input 128. An exemplary list of metrics 130 is shown in FIG. 3.

[0122] In certain embodiments, the potassium level can be determined from sensor data (e.g., 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 can be one of the non-analyte sensors 206). For example, the potassium level refers to timestamped potassium measurements or values that are continuously generated and stored over time.

[0123] In certain embodiments, the potassium baseline can be determined from sensor data (e.g., potassium measurements obtained from the continuous lactate sensor of the continuous analyte monitoring system 104). The potassium baseline represents the normal (e.g., average) potassium level of the user during a period when significant variations in potassium production are not typically expected. The user's baseline potassium is generally expected to remain constant over time, unless challenged during exercise by actions such as ingestion of potassium or potassium-rich foods, or as a result of changes in kidney health or reduced kidney function.

[0124] Each user may have a different potassium baseline. In certain embodiments, a user's potassium baseline may be determined by calculating the average of the user's potassium levels over a certain amount of time during which significant variation is not expected. For example, a user's baseline potassium may be determined over other periods when the user is sleeping, sitting in a chair, or is prone to sitting and not consuming foods or medications that would reduce or increase potassium levels (e.g., when there are no external conditions that would affect the potassium baseline). In certain embodiments, DAM116 may continuously calculate the potassium baseline, timestamp the calculated potassium baseline, and store the corresponding information in the user's profile 118. In such embodiments, the potassium baseline may be determined based on the average potassium level throughout the user's daily activities.

[0125] In certain other embodiments, DAM116 may use potassium levels measured over a period of time during which the user has been engaged in exercise and / or consumed potassium and / or there are external conditions that affect the potassium baseline for at least a subset of that period. In such embodiments, DAM116 may, in some examples, first identify which measured potassium values should not be used to calculate the potassium baseline by identifying which potassium values were affected by external events such as food consumption, exercise, medications, or other perturbations that interfere with the capture of the potassium baseline measurement. DAM116 may then exclude such measurements when calculating the user's potassium baseline. In some other examples, DAM116 may calculate the potassium baseline by first determining the percentage of the number of potassium values measured during this period that represents the lowest lactate value measured. DAM116 may then take the average of such potassium values to determine the potassium baseline level.

[0126] In certain embodiments, the absolute maximum potassium level can be determined from sensor data (e.g., potassium measurements obtained from the continuous CPM of the continuous analyte monitoring system 104), health / illness metrics (e.g., described in more detail below), and / or disease stage metrics (e.g., described in more detail below). The absolute maximum potassium level represents the maximum potassium level of a user determined to be safe over a period of time (e.g., hourly, weekly, daily, etc.). In certain embodiments, the absolute maximum potassium level can be consistent across all users (e.g., set to 5.5 mmol / L based on current medical guidelines). In certain other embodiments, each patient can have a different absolute maximum potassium level. For example, the absolute maximum potassium level can be lower for a user diagnosed with stage 1 CKD (e.g., normal or high GFR (GFR > 90 mL / min)) than for 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 can change over time. For example, a user can initially be assigned an absolute maximum potassium level based on clinical input. This assigned absolute maximum potassium level can be adjusted over time based on other sensor data, disease stage, co-morbidities, etc. for the patient.

[0127] For example, a user's absolute maximum potassium level can change over time as the user's renal function, kidney disease, and / or one or more other diseases progress and / or improve. In certain embodiments, a first absolute maximum potassium level can be determined for a period in which there are no external conditions affecting the potassium level, and a second absolute maximum potassium level can be determined for a period in which external conditions affecting the potassium level are present (e.g., a period in which the user is consuming potassium, exercising, taking medications that affect the potassium level, etc.).

[0128] In certain embodiments, the absolute minimum potassium level can be determined from sensor data (e.g., potassium measurements obtained from a continuous lactate sensor of the continuous analyte monitoring system 104), health / illness metrics (e.g., described in more detail below), and / or disease stage metrics (e.g., described in more detail below). The absolute minimum potassium level represents the minimum potassium level of a user determined to be safe over a period of time (e.g., hourly, weekly, daily, etc.). In certain embodiments, the absolute minimum potassium level can be consistent across all users (e.g., set based on current medical guidelines). In certain other embodiments, each user can have a different absolute minimum potassium level. For example, the absolute minimum potassium level can be lower for a user diagnosed with stage 1 CKD (e.g., normal or high GFR (GFR > 90 mL / min)) than for 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 can change over time. For example, a user can initially be assigned an absolute minimum potassium level based on clinical input. This assigned absolute minimum potassium level can be adjusted over time based on other sensor data, disease stage, co-morbidities, etc. for the patient.

[0129] For example, a user's absolute minimum potassium level can change over time as the user's renal function, kidney disease, and / or one or more other diseases progress and / or improve. In certain embodiments, a first absolute minimum potassium level can be determined for a period in which there are no external conditions affecting the potassium level, and a second absolute minimum potassium level can be determined for a period in which external conditions affecting the potassium level are present (e.g., a period in which the user is consuming potassium, exercising, taking medications that affect the potassium level, etc.).

[0130] In certain embodiments, potassium thresholds other than the user's absolute maximum and / or minimum potassium levels can be determined from sensor data (e.g., potassium measurements obtained over time from the continuous lactate sensor of continuous analyte monitoring system 104), health / illness metrics (e.g., described in more detail below), and / or disease stage metrics (e.g., described in more detail below). Such potassium thresholds can represent, for example, the maximum or minimum potassium levels determined to be safe during a particular activity, which can vary across different activities. For example, since exercise is known to affect potassium levels, the maximum and / or minimum potassium thresholds for a user during exercise can be different from the maximum and / or minimum potassium thresholds for the user during other activities.

[0131] In certain embodiments, the rate of change of potassium level can be determined from sensor data (e.g., potassium measurements obtained over time from the CPM of continuous analyte monitoring system 104). For example, the rate of change of potassium level refers to the rate indicating how one or more timestamped potassium measurements or values change relative to one or more other timestamped potassium measurements or values. The rate of change of potassium level can be determined over a period of one second or more, one minute or more, one hour or more, one day or more, etc.

[0132] In certain embodiments, the determined potassium level change rate may be marked as "increasing rapidly" or "decreasing rapidly". As used herein, "rapidly" refers to a potassium level change rate that is clinically significant and indicates a tendency for the potassium level to breach the absolute maximum potassium level or the absolute minimum potassium level within a defined period. In other words, the prediction tendency (e.g., generated by the decision support engine 114 using one or more trained models) may, in some cases, indicate, based on the determined potassium level change rate, that the patient is likely to reach the absolute maximum potassium level within a specified period (e.g., 1 hour or 2 hours). Accordingly, such a potassium level change rate may be marked as "increasing rapidly". Similarly, the prediction tendency (e.g., generated by the decision support engine 114 using one or more trained models) may, in some cases, indicate, based on the determined potassium level change rate, that the patient is likely to reach the absolute minimum potassium level within a specified period (e.g., 1 hour or 2 hours). Accordingly, such a potassium level change rate may be marked as "decreasing rapidly".

[0133] In certain embodiments, the potassium baseline change rate may be determined from the potassium baseline determined over time for the user. For example, the potassium baseline change rate refers to the rate indicating how one or more timestamped potassium baselines for the user change with respect to one or more other timestamped potassium baselines for the same user. The potassium baseline change rate may be determined over a period of 1 second or more, 1 minute or more, 1 hour or more, 1 day or more, etc.

[0134] In certain embodiments, the potassium clearance rate can be determined from sensor data (e.g., potassium measurements obtained from the 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 can indicate kidney function. In particular, the slope of the potassium clearance curve during a first period (e.g., after consuming a known amount of potassium) compared to the slope of the potassium clearance curve during a second period (e.g., after consuming the same amount of potassium) can indicate the ability of the kidneys to function, and more specifically, the ability to maintain potassium homeostasis (e.g., the potassium clearance rate can be slower when the user's kidneys are impaired than when the user's kidneys are healthy).

[0135] In certain embodiments, the potassium clearance rate can be determined by calculating the slope between the potassium value at t0 (e.g., during a period when the potassium level is rising) and the user's potassium baseline reached at t1. In certain embodiments, the potassium clearance rate can be calculated over time until the user's increased potassium level reaches some value (e.g., a percentage of the user's potassium baseline) relative to the user's potassium baseline. The potassium clearance rate calculated over time can be timestamped and stored in the user's profile 118.

[0136] In certain embodiments, the rate of increase of the potassium level can be determined from sensor data (e.g., potassium measurements obtained from the CPM of the continuous analyte monitoring system 104) after consumption of potassium (e.g., potassium-containing food). The rate of increase of the potassium level analyzed over time can indicate kidney function. For example, if the user has some kidney dysfunction, the kidneys delay potassium removal, so the potassium level can show a more rapid increase.

[0137] In certain embodiments, the standard deviation of the potassium level (not shown) can be determined from the sensor data. In some examples, the standard deviation of the potassium level can be determined based on the variability of the potassium level compared to the average potassium level over one or more periods.

[0138] In certain embodiments, the potassium trend can be determined based on the potassium level over a particular period. In certain embodiments, the potassium trend can be determined based on the potassium baseline over a particular period. In certain embodiments, the potassium trend can be determined based on the absolute maximum potassium level over a particular period. In certain embodiments, the potassium trend can be determined based on the absolute maximum potassium level over a particular period. In certain embodiments, the potassium trend can be determined based on the rate of change of the potassium level over a particular period. In certain embodiments, the potassium trend can be determined based on the rate of change of the potassium baseline over a particular period. In certain embodiments, the potassium trend can be determined based on the calculated potassium clearance rate over a particular period.

[0139] In certain embodiments, the glucose level can be determined from the sensor data (e.g., blood glucose measurements obtained from the continuous lactate sensor of the continuous analyte monitoring system 104).

[0140] In certain embodiments, the rate of change of the glucose level can be determined from the sensor data (e.g., glucose measurements obtained over time from the continuous glucose monitor (CGM) of the continuous analyte monitoring system 104). For example, the rate of change of the glucose level refers to the rate at which one or more timestamped glucose measurements or values change relative to one or more other timestamped glucose measurements or values. The rate of change of the glucose level can be determined over one second or more, one minute or more, one hour or more, one day or more, etc.

[0141] In certain embodiments, the blood glucose trend can be determined based on glucose levels over a specific period. In certain embodiments, the glucose trend can be determined based on the rate of change of glucose levels over a specific period. In certain embodiments, the glucose trend can be determined based on one or more glucose metrics and / or inputs over a specific period.

[0142] In certain embodiments, glycemic variability can be determined from sensor data (e.g., glucose measurements obtained over time from the continuous glucose monitor (CGM) of the continuous analyte monitoring system 104). For example, glycemic variability refers to the standard deviation of glucose levels over a period of time. Glycemic variability can be determined over a period of 1 minute or more, 1 hour or more, 1 day or more, etc.

[0143] In certain embodiments, the glucose clearance rate can be determined from sensor data (e.g., glucose levels obtained from the continuous glucose sensor of the continuous analyte monitoring system 104) after consumption of a known or estimated amount of glucose. The glucose clearance rate analyzed over time can indicate glucose homeostasis. In particular, the slope of the glucose clearance curve during a first period (e.g., after consumption of a known amount of glucose) compared to the slope of the glucose clearance curve during a second period (e.g., after consumption of the same amount of glucose) can indicate the ability of the kidneys to function, more specifically, the ability to maintain glucose homeostasis (e.g., the glucose clearance rate can be slower when the user's kidneys are impaired than when the user's kidneys are healthy).

[0144] In certain embodiments, the glucose clearance rate can be determined by calculating the slope between the initial high glucose value at t0 (e.g., the highest glucose level during a period of 20 - 30 minutes after glucose consumption) and the subsequent low glucose value at t1. The low glucose value (G L ) is the user's initial high glucose value (G H ) and the baseline glucose value before glucose consumption (G B) can be determined based on. In certain embodiments, G L is G H and G B and the glucose value between, for example, G L = G B + K * (G H - G B ) / 2, where K can be a percentage representing how much the user's glucose level has returned to the user's baseline value. When K is equal to 0, the low glucose value is equal to the baseline glucose value. When K is equal to 0.5, the low glucose value is equal to the average glucose value between the initial glucose value and the baseline glucose value.

[0145] In certain embodiments, the glucose clearance rate can be determined over one or more periods after glucose consumption. The glucose clearance rate can be calculated for each time period to represent the dynamics of the glucose clearance rate after glucose consumption. The glucose clearance rate calculated over time can be timestamped and stored in the user profile 118. Certain metrics can be derived from the timestamped glucose clearance rate such as average, median, standard deviation, percentile, etc. In certain embodiments, a user with kidney disease may have impaired kidney function to metabolize insulin, and the time elapsed from the initial high glucose value to the low glucose value can indicate the ability of the kidneys to function. The time from the initial high glucose value to the low glucose value and the glucose clearance rate can be timestamped and stored in the user profile 118.

[0146] In certain embodiments, insulin sensitivity can be determined using medical history data, real-time data, or a combination thereof, and can 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 can help reduce insulin resistance in the user.

[0147] In certain embodiments, insulin on board can be determined using a known or learned (e.g., from user data) insulin time-action profile that can take into account both non-analyte sensor data inputs (e.g., insulin delivery information) and / or basal metabolic rate (e.g., insulin updates to maintain the body's operations) and insulin use driven by activity or food consumption.

[0148] In certain embodiments, the insulin clearance rate can be determined using historical data, real-time data, or a combination thereof, for example, by calculating the slope between an initial insulin value at t0 (e.g., during a period when the insulin level is rising) and the user's final insulin value at t1.

[0149] In certain embodiments, the albumin level can be determined from sensor data (e.g., creatinine measurements obtained from the continuous analyte monitoring system 104).

[0150] In certain embodiments, the absolute maximum albumin level can be determined from sensor data (e.g., albumin measurements obtained from the continuous albumin sensor of the continuous analyte monitoring system 104), health / illness metrics (e.g., described in more detail below), and / or disease stage metrics (e.g., described in more detail below). The absolute maximum albumin level represents the maximum creatinine level of a user determined to be safe over a period of time (e.g., hourly, weekly, daily, etc.). Each user may have a different absolute maximum albumin level. The user's absolute maximum albumin level can change over time as the user's kidney function, kidney disease, and / or one or more other diseases progress and / or improve.

[0151] In certain embodiments, the rate of change of the albumin level can be determined from sensor data (e.g., albumin measurements obtained over time from the albumin sensor of the continuous analyte monitoring system 104). For example, the rate of change of the albumin level refers to a rate that describes how one or more timestamped albumin measurements or values change relative to one or more other timestamped albumin measurements or values. The rate of change of the albumin level can be determined over a period of one second or more, one minute or more, one hour or more, one day or more, etc. In certain embodiments, the average albumin level can be calculated to determine the rate of change of the user's calculated average albumin level.

[0152] In certain embodiments, the albumin trend can be determined based on albumin levels over a specific period of time.

[0153] In certain embodiments, the creatinine level can be determined from sensor data (e.g., creatinine measurements obtained from the continuous analyte monitoring system 104).

[0154] In certain embodiments, the absolute maximum creatinine level can be determined from sensor data (e.g., creatinine measurements obtained from the continuous creatinine sensor of the continuous analyte monitoring system 104), health / illness metrics (e.g., described in more detail below), and / or disease stage metrics (e.g., described in more detail below). The absolute maximum creatinine level represents the maximum creatinine level of a user 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 can change over time as the user's renal function, kidney disease, and / or one or more other diseases progress and / or improve.

[0155] In certain embodiments, because the rate of increase and / or removal of creatinine may be impaired in users with kidney disease, the rate of change of the creatinine level can be determined from sensor data (e.g., creatinine measurements obtained from the creatinine sensor of the continuous analyte monitoring system 104 over time). For example, the rate of change of the creatinine level refers to the rate that describes how one or more timestamped creatinine measurements or values change relative to one or more other timestamped creatinine measurements or values. The rate of change of the creatinine level can 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 can be calculated to determine the rate of change of the user's calculated average creatinine level. Such measurements may be taken after the user has ingested a known or unknown amount of creatinine, e.g., a creatinine supplement or red meat.

[0156] In certain embodiments, the creatinine trend can be determined based on the creatinine level over a specific period of time.

[0157] In certain embodiments, the urea level can be determined from sensor data (e.g., BUN measurements obtained from the continuous analyte monitoring system 104). In certain embodiments, the BUN trend can be determined based on the BUN levels over a specific period of time.

[0158] In certain embodiments, the rate of change of the BUN level can be determined from sensor data (e.g., BUN measurements obtained over time from the BUN sensor of the continuous analyte monitoring system 104). For example, the rate of change of the BUN level refers to the rate that describes how one or more timestamped BUN measurements or values change relative to one or more other timestamped BUN measurements or values. The rate of change of the BUN level can be determined over a period of 1 second or more, 1 minute or more, 1 hour or more, 1 day or more, etc. In certain embodiments, the average BUN level can be calculated to determine the rate of change of the user's calculated average BUN level.

[0159] In certain embodiments, the BUN clearance rate can be determined by calculating the slope between the initial BUN value at t0 (e.g., during a period when the BUN level is rising) and the user's final BUN value at t0.

[0160] In certain embodiments, the inulin level can be determined from sensor data (e.g., inulin measurements obtained from the continuous analyte monitoring system 104). In certain embodiments, the inulin trend can be determined based on the inulin levels over a specific period of time.

[0161] In certain embodiments, the dextran level can be determined from sensor data (e.g., dextran measurements obtained from the continuous analyte monitoring system 104). In certain embodiments, the dextran trend can be determined based on the dextran levels over a specific period of time.

[0162] In certain embodiments, the saccharin level can be determined from sensor data (e.g., saccharin measurements obtained from the continuous analyte monitoring system 104). In certain embodiments, the saccharin trend can be determined based on the saccharin levels over a particular period of time.

[0163] In certain embodiments, the iotalamic acid level can be determined from sensor data (e.g., iotalamic acid measurements obtained from the continuous analyte monitoring system 104). In certain embodiments, the iotalamic acid trend can be determined based on the iotalamic acid levels over a particular period of time.

[0164] In certain embodiments, the 125l-iotanarate level can be determined from sensor data (e.g., 125l-iotanarate measurements obtained from the continuous analyte monitoring system 104). In certain embodiments, the 125l-iotanarate trend can be determined based on the 125l-iotanarate levels over a particular period of time.

[0165] In certain embodiments, the cystatin C level can be determined from sensor data (e.g., cystatin C measurements obtained from the continuous analyte monitoring system 104). In certain embodiments, the C-peptide trend can be determined based on the cystatin C levels over a particular period of time.

[0166] In certain embodiments, the C-peptide level can be determined from sensor data (e.g., C-peptide measurements obtained from the continuous analyte monitoring system 104). In certain embodiments, the C-peptide trend can be determined based on the C-peptide levels over a particular period of time.

[0167] In certain embodiments, the 51Cr-EDTA level can be determined from sensor data (e.g., 51Cr-EDTA measurements obtained from the continuous analyte monitoring system 104). In certain embodiments, the 51Cr-EDTA trend can be determined based on the 51Cr-EDTA levels over a particular period of time.

[0168] In certain embodiments, lactate levels can be determined from sensor data (e.g., lactate measurements obtained from the continuous analyte monitoring system 104). In certain embodiments, a lactate trend can be determined based on lactate levels over a particular period of time. In certain embodiments, information regarding time in range (TIR) of lactate can also be determined based on lactate levels over time.

[0169] In certain embodiments, aspartic acid levels can be determined from sensor data (e.g., aspartic acid measurements obtained from the continuous analyte monitoring system 104). In certain embodiments, an aspartic acid trend can be determined based on aspartic acid levels over a particular period of time.

[0170] In certain embodiments, polyfructosan levels can be determined from sensor data (e.g., polyfructosan measurements obtained from the continuous analyte monitoring system 104). In certain embodiments, a polyfructosan trend can be determined based on polyfructosan levels over a particular period of time.

[0171] In certain embodiments, betaine levels can be determined from sensor data (e.g., betaine measurements obtained from the continuous analyte monitoring system 104). In certain embodiments, a betaine trend can be determined based on betaine levels over a particular period of time.

[0172] In certain embodiments, health and disease metrics can be determined from physiological sensors (e.g., body temperature), activity sensors, or a combination thereof, based on, for example, one or more of user input (e.g., pregnancy information or known disease or illness information). In certain embodiments, based on the values of the health and disease metrics, for example, the user's state can be defined as one or more of healthy, ill, at rest, or fatigued.

[0173] In certain embodiments, a disease stage metric, such as for a kidney disease, can be determined based on one or more of the user inputs or outputs provided, for example, by the decision support engine 114 illustrated in FIG. 1. In certain embodiments, exemplary disease stages for a kidney disease can 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 can be represented as GFR values / ranges, severity scores, and the like.

[0174] In certain embodiments, a dietary status metric can indicate the user's status regarding food consumption. For example, the dietary status can indicate whether the user is in one of a fasting state, pre-meal state, eating state, post-meal reaction state, or stable state. In certain embodiments, the dietary status can also indicate residual nutrition, such as consumed meals, snacks, or beverages, and can be determined from food consumption information, meal time information, and / or digestibility information, which can be correlated, for example, to the type, amount, and / or order of food (e.g., which food / beverage was eaten first).

[0175] In certain embodiments, a dietary habit metric is based on the content and timing of the user's meals. For example, if the dietary habit metric is on a scale of 0 - 1, in an example, the more the user eats a better / healthier diet, the higher the user's dietary habit metric is relative to 1. Also, in an example, the more the user's food consumption adheres to a specific time schedule or a recommended meal, the closer the user's dietary habit metric is to 1.

[0176] In certain embodiments, medication compliance (not shown) is measured by one or more metrics indicating how committed the user is to their medication regimen. In certain embodiments, the medication compliance metric is calculated based on one or more of the timing of when the user takes the 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, the user's medication compliance can be determined based on user input and / or analyte data received from the analyte monitoring system 104 in a clinical trial where the consumption of the medication and the timing of such medication consumption are monitored.

[0177] In certain embodiments, the activity level metric can indicate the user's activity level. In certain embodiments, the activity level metric can be determined based on input from an activity sensor or other physiological sensor, such as the non-analyte sensor 206 for example. In certain embodiments, the activity level metric can be calculated by the DAM 116 based on one or more of the inputs 128, such as one or more of exercise information, non-analyte sensor data (e.g., accelerometer data), time, user input, etc. In certain embodiments, the activity level can be represented as the user's step rate. The activity level metric can be timestamped so that it can be correlated with the user's lactate level.

[0178] In certain embodiments, the exercise regimen metric (not shown) can indicate one or more of what type of activity the user is engaged in, the corresponding intensity of such activity, how frequently the user is engaged in such activity, etc. In certain embodiments, the exercise regimen metric can be calculated based on one or more of analyte sensor data inputs (e.g., from a lactate monitor, glucose monitor, etc.), non-analyte sensor data inputs (e.g., non-analyte sensor data inputs from an accelerometer, heart rate monitor, respiratory rate sensor, etc.), calendar input, user input, etc.

[0179] In certain embodiments, the body temperature metric can be calculated by the DAM 116 based on input 128, and more specifically, non-analyte sensor data from a temperature sensor. In certain embodiments, the heart rate metric (e.g., including heart rate and heart rate variability) can be calculated by the DAM 116 based on input 128, and more specifically, non-analyte sensor data from a heart rate sensor. In certain embodiments, a respiration metric (not shown) can be calculated by the DAM 113 based on input 128, and more specifically, non-analyte sensor data from a respiratory rate sensor. In certain embodiments, the blood pressure measurement metric (e.g., including blood pressure level and blood pressure trend) can be calculated by the DAM 113 based on input 128, and more specifically, non-analyte sensor data from a blood pressure sensor.

[0180] In certain embodiments, as described in more detail below, physiological parameters related to the user (e.g., potassium level, rate of change of potassium level, glucose level, creatinine level, albumin level, heart rate, blood pressure, etc.) may be stored as metric 130 when the diagnosis, presence, stage (e.g., severity), or risk of kidney disease is confirmed. In certain embodiments, such physiological parameters can be analyzed over time to provide an indication of improvement or worsening of the user's kidney disease. In certain embodiments, the user-specific values of the physiological parameters experienced by the user can be useful inputs for training one or more models designed to evaluate the presence and / or severity of kidney disease in the user. In certain embodiments, one or more personalized models specific to the user can be created using the user-specific values of the physiological parameters experienced by the user to more accurately predict the presence and / or severity of the user's kidney disease.

[0181] Exemplary methods and systems for providing decision-making support regarding kidney disease Figure 4 is a flow diagram showing an exemplary method 400 for providing decision support using a continuous analyte sensor, according to certain illustrative aspects of the present disclosure. For example, method 400 may be implemented using a continuous analyte monitoring system 104 that includes at least a continuous analyte sensor 202, as shown in FIGS. 1 and 2, to provide decision support to a user. Method 400 may provide decision support in real time or within a specified period and may provide, for example, retraining or updating of a machine learning model based on patient input and / or diagnostic tests.

[0182] 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 data and / or population data, to provide kidney disease prediction. The algorithms and / or machine learning models may take into account one or more of the inputs 128 and / or metrics 130 described with respect to FIG. 3 for the patient when providing predictions related to screening, diagnosis, and stage classification of kidney disease.

[0183] One or more of the machine learning models described herein for making such predictions may be trained using population data at least initially. A method for training one or more machine learning models may be described in more detail below with respect to FIG. 5.

[0184] In certain embodiments, as an alternative to using a machine learning model, the decision support engine 114 may use a rule-based model to predict the risk or likelihood that a patient will experience a kidney disease. A rule-based model involves using a set of rules for analyzing data. These rules often follow the pattern of "If X happens then do or conclude Y" and are thus often referred to as "If Statements". In particular, the decision support engine 114 can apply rule statements (e.g., if-then statements) to determine the risk or likelihood that a patient has developed, is experiencing, and / or is at a particular stage of a kidney disease.

[0185] Such rules can be maintained in a reference library by the decision support engine 114. For example, the reference library can maintain ranges of analyte (e.g., potassium) levels and rates of change of analyte levels (and / or other analyte data) and / or other analyte metrics, which can be mapped, for example, to different kidney disease risk stratifications or different stages of kidney disease. In certain embodiments, such rules can be determined based on empirical studies or analyses of patient medical records, such as records stored in the medical history database 112. In some cases, the reference library can be very granular. For example, other factors can be used in the reference library to create such "rules". Other factors can include gender, age, diet, medical history, family medical history, body mass index (BMI), and the like. The increased granularity can provide a more accurate output.

[0186] Returning now to FIG. 4, method 400 may be performed by decision support system 100 and may collect / generate data such as inputs 128 and metrics 130, including, for example, the aforementioned analyte data, patient information, and non-analyte sensor data, and (1) automatically detect and classify abnormal kidney function, (2) assess the risk of kidney disease; and (3) assess the presence and stage of kidney disease. In other words, the decision support system presented herein may provide information useful for directing and improving the care of patients with kidney disease, more specifically chronic kidney disease (CKD), or patients at risk of kidney disease. Method 400 is described below with reference to FIGS. 1 and 2, and their components.

[0187] Generally, real-time or continuous measurements of analyte levels, rates of change, trends, clearance rates, and / or other analyte data, measured in interstitial fluid or blood, can be used to indicate changes in kidney function (e.g., either dysfunction or improvement). Such data can indicate changes in kidney function earlier and / or better than other conventional kidney disease diagnostic tools, such as the creatinine baseline level measured by a glomerular filtration rate (GFR) test, which only changes if there has been a significant loss of kidney function, and the albumin-to-creatinine ratio (ACR) test, electrocardiogram (ECG), etc. Thus, continuous analyte monitoring of one or more analytes, such as potassium, may provide earlier and / or improved screening, diagnosis, prognosis, and / or staging of kidney disease compared to conventional diagnostics.

[0188] Accordingly, at block 402, method 400 begins by continuously monitoring one or more analytes of a patient, such as patient 102 shown in FIG. 1, during one or more periods (e.g., multiple periods) to obtain analyte data. The one or more analytes to be monitored can, in certain embodiments, include at least potassium. Accordingly, the analyte data can contain at least potassium data. Block 402 can be performed, in certain embodiments, by the continuous analyte monitoring system 104 shown in FIGS. 1 and 2, and more specifically, by the continuous analyte sensor 202 shown in FIG. 2. For example, the continuous analyte monitoring system 104 can, in certain embodiments, include a continuous potassium monitor (CPM) 202 configured to measure a patient's potassium level.

[0189] As mentioned, potassium is one of the most important minerals in the body. Potassium helps to regulate fluid balance, muscle contractions, and nerve signals. A high-potassium diet can also help to reduce blood pressure and water retention, protect against stroke, and prevent osteoporosis and kidney stones. Approximately 98% of the potassium in the body is stored inside cells, and the remaining 2% is stored outside cells. Thus, for example, a rapid release of potassium from cells, which can occur as a result of cell injury, cell lysis (e.g., red blood cell (RBC) lysis), and exercise, can dramatically affect extracellular potassium levels (e.g., blood potassium levels).

[0190] Generally, a patient's normal (e.g., baseline) blood potassium level can range from 3.6 to 5.3 millimoles per liter (mmol / L). When a patient's blood potassium level is in the range of 5.3 to 6.0 mmol / L, the user may consider that the patient has an elevated blood potassium level that requires close monitoring. If the blood potassium level exceeds the elevated range, e.g., exceeds 6.0 mmol / L, the condition may be described as "hyperkalemia" or a high blood potassium level. Hyperkalemia can increase the risk of cardiac arrhythmia episodes and even sudden death. Symptoms associated with mild hyperkalemia include muscle weakness, numbness, tingling, nausea, or other abnormal sensations, while symptoms of very high potassium levels include palpitations of the heart, shortness of breath, chest pain, nausea, or vomiting. In more severe cases of hyperkalemia, the patient may experience respiratory failure, sudden cardiac death, or other fatal events. Conversely, if the blood potassium level is lower than normal, e.g., lower than 2.5 mmol / L, the condition may be described as "hypokalemia" or a low blood potassium level. Low potassium levels in patients with untreated kidney disease can result in hypokalemia. Similar to hyperkalemia, severe hypokalemia can result in symptoms of respiratory failure, sudden cardiac death, arrhythmia, or other fatal events.

[0191] The kidneys mainly play a role in maintaining the body's potassium content / distribution by matching potassium intake with potassium excretion. The regulation of renal potassium excretion occurs over several hours, and thus, changes in extracellular potassium concentration are initially buffered by the movement of potassium into and out of skeletal muscle. The regulation of potassium distribution between the intracellular and extracellular spaces is referred to as internal potassium balance. The most important factors that regulate 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 through renal mechanisms that transport potassium and regulate its secretion, reabsorption, and excretion. However, if the kidneys are damaged or lose their ability to function, they may no longer be able to remove excess potassium, and thus, potassium levels can accumulate in the body, potentially causing, for example, hyperkalemia. Therefore, potassium monitoring can be proven useful for detecting and classifying abnormal renal function, evaluating kidney health, assessing the risk of kidney disease or diagnosing and monitoring kidney disease, and / or identifying the stage of kidney disease. In certain examples, potassium monitoring can be useful for assessing the risk of other adverse health events that can result from kidney disease.

[0192] Accordingly, in certain embodiments, a continuous potassium monitoring (CPM) sensor, such as the CPM202, can collect potassium measurements that can be used to generate potassium data including a potassium baseline, a rate of change of potassium, a rate of change of the potassium baseline, a personalized potassium level, an average potassium level, maximum and / or minimum potassium levels, absolute maximum and / or minimum potassium levels, a standard deviation of potassium levels, a potassium clearance rate, a potassium trend, and the like.

[0193] In certain embodiments, the potassium data includes the patient's average potassium level. Typically, as kidney disease (e.g., CKD) progresses in a patient, the patient's average potassium level increases. Thus, CPM can be used to monitor a patient's potassium level over a period of time, including but not limited to real-time potassium levels, potassium change rates, and potassium clearance rates, and these measurements can be averaged over a period of time.

[0194] In some examples, a change in the average potassium level from one period to another may indicate a new or worsening kidney disease in the patient. For example, the average potassium level of a patient during a first initial period (e.g., the first month) can be compared to the average potassium level of the patient during a subsequent second period (e.g., the second month), whereby the difference in the average potassium levels between the first and second periods can suggest a change in the patient's kidney function. In such an example, an increase in the average potassium level between the first and second periods may indicate a new or worsening kidney disease. In another example, the average potassium level of a patient can be monitored over several periods to determine one or more trends (e.g., increasing / decreasing average potassium levels over several periods). In such an example, a trend of increasing average potassium levels may indicate a deteriorating kidney disease.

[0195] Alternatively, in some cases, a change in the average potassium level may indicate an improvement or stabilization of the kidney disease. For example, a situation where the average potassium level of a patient during a first period is similar to the average potassium level of the patient during a second month may indicate a stabilization of the patient's kidney disease. In another example, if the average potassium level of a patient during a first period is higher than the average potassium level of the patient during a second period, such a situation may indicate an improvement in the patient's kidney disease.

[0196] In certain embodiments, in addition to the average potassium level, the standard deviation of the potassium level can be used to indicate the progression of kidney disease. In such embodiments, an increase in the variability of the potassium level, i.e., a higher standard deviation of the potassium level, may indicate a worsening of the patient's kidney disease, while a decrease in the variability of the potassium level, i.e., a lower standard deviation of the potassium level, may indicate an improvement or stabilization of the patient's kidney disease. In some examples, the standard deviation of the patient's potassium level can be determined based on the variability of the patient's potassium level compared to the patient's average potassium level over one or more periods. In such an example where the standard deviation increases from a first period to a subsequent second period, such a situation may indicate a new or worsening of the patient's kidney disease. Alternatively, if the standard deviation decreases from a first period to a subsequent second period, such a situation may indicate an improvement in the patient's kidney disease. Further, similar standard deviations between a first period and a subsequent second period may indicate stabilization of the patient's kidney disease. Together with the average potassium level, the standard deviation of the potassium level can be used to improve and / or correct glomerular filtration rate (GFR) test data.

[0197] In certain embodiments, a CPM, such as CPM202, can also be utilized to collect potassium measurements for generating a potassium clearance rate. To determine the potassium clearance rate, a user can consume or administer (e.g., infuse) a known or estimated amount of potassium. Subsequently, the potassium level, rate of change, etc. of the patient can be monitored after the consumption or administration of potassium to determine the potassium clearance rate. In certain embodiments, the potassium clearance rate can be determined by calculating the slope between an initial potassium value (e.g., after consumption / administration of potassium) and a potassium baseline associated with the user. In certain embodiments, the potassium clearance rate can be calculated over time until the increased potassium level of the patient reaches some value (e.g., % of the patient's potassium baseline) relative to the patient's potassium baseline. Generally, the potassium clearance rate over time (e.g., the potassium clearance rate over one or more periods) can be used to determine changes in renal function. For example, if the potassium clearance rate of a patient during a first period is higher than the potassium clearance rate during a second period, such a situation may indicate that the renal function of the patient has declined (e.g., the patient's kidney disease has worsened).

[0198] In certain embodiments, the potassium level may be personalized to provide context to the patient's collected potassium measurement. Such personalized potassium levels may be used in conjunction with potassium level averages, standard deviations, and / or other analyte data to indicate abnormal kidney function, indicate the presence and / or progression of kidney disease, and / or in certain embodiments, indicate the risk of other adverse health events that may occur as a result of kidney disease. In certain embodiments, personalization of the potassium level involves associating the determined potassium level with one or more changes (e.g., deltas) relative to the patient's potassium baseline (e.g., if the patient does not have kidney disease, is not experiencing symptoms related to kidney disease, or has not participated in any activity that affects potassium) to indicate a high and / or low potassium level for that patient. For example, a patient may have a personalized "high" potassium level established as an increase of 0.3 mmol / L above baseline and a baseline of 5.2 mmol / L, such that when the patient's potassium level reaches 5.6 mmol / L, the patient is determined to have a "high" potassium level. In this example, another patient may not be determined to have a "high" potassium level when such a patient's level reaches 5.6 mmol / L. In certain embodiments, personalization of the potassium level involves associating the determined potassium level with a personalized threshold, such as a personalized threshold for hyperkalemia and / or hypokalemia. In certain embodiments, personalization of the potassium level involves associating the determined potassium level with a personalized rate of change (e.g., increasing rapidly and / or decreasing rapidly). In certain embodiments, personalization of the potassium level involves associating the determined potassium level with the signs and symptoms of, for example, hyperkalemia and / or hypokalemia (or other symptoms / conditions of kidney disease) to determine at what potassium level a patient may experience such a condition.

[0199] In certain embodiments, the potassium level can be individualized based on a patient's activities, such as the patient's daily activities. Generally, a patient's potassium level rises and falls throughout the day as a result of activities performed by the patient. Such activities can include exercise, diet, posture, urination, and the like. For example, when a patient exercises, the patient's potassium level can rise during the performance of the exercise and then fall after the exercise is performed. On the other hand, a patient's potassium level can be low before a meal and may rise after the meal is consumed. Accordingly, the potassium levels during these activities can be used in various ways to derive conclusions about the user's condition and / or to indicate the presence and / or severity of a kidney disease.

[0200] In certain embodiments, the potassium level can be personalized based on non-analyte data. For example, the potassium level can be personalized by associating the determined potassium level with metrics derived from the EKG signal (e.g., QRS interval, peak T wave, P wave duration, PR interval, etc.).

[0201] For example, in certain embodiments, a personalized potassium baseline can be determined based on the average potassium level over one or more of a patient's daily activities. In such examples, potassium levels obtained during activities known to affect potassium levels can be excluded from the determination of the personalized potassium baseline. Exercise is one such activity known to affect potassium levels, and thus potassium levels during exercise can be excluded from the determination of the personalized potassium baseline. In certain embodiments, a personalized activity-specific potassium baseline can be determined based on the average potassium level of the patient during the performance of such an activity. In certain embodiments, the potassium threshold for a particular activity can be adjusted (e.g., personalized) for a patient based on the expected change in the patient's potassium level due to that activity. For example, since exercise is known to affect potassium levels, different potassium thresholds can be used to derive insights regarding the patient's kidney health when the patient is exercising. In certain embodiments, potassium levels during a particular activity can be used to indicate the presence and / or severity of kidney disease. For example, in certain embodiments, potassium levels can be correlated with urine volume, and potassium levels during times of high and / or low urine volume can be used to indicate kidney dysfunction and / or changes in kidney function of the patient.

[0202] Specific activities of a patient can be automatically determined via an analyte monitor (e.g., potassium, glucose, lactate, etc.) and / or a non-analyte monitor (e.g., HR sensor, accelerometer, etc.). Exercise is one activity that can be determined by a combination of non-potassium analyte monitors. For example, exercise may be associated with an increase in analyte levels (e.g., lactate), as well as an increase in non-analyte metrics (e.g., HR). A combination of analyte data and non-analyte data can be used to determine that a patient is exercising during a period of time, and as a result, the potassium level during that period can be annotated or flagged, for example, as an "exercise potassium level". The annotated or flagged potassium level can then be excluded or used in the above determination. For example, the annotated or flagged exercise potassium level can be excluded from an individualized potassium baseline determination. In certain embodiments, other types of monitors (e.g., analyte and / or non-analyte) can be used to automatically determine daily activities that can lead to additional insights including, but not limited to, exercise, diet, posture, bathroom / urination volume.

[0203] In certain embodiments, techniques can be introduced to account for and / or correct inaccurate potassium levels, trends, potassium variability, average values, etc. For example, the potassium level can be corrected over a period of time after the insertion of a continuous potassium sensor, a continuous multi - analyte sensor, or other potassium monitoring device. Due to the risk of hemolysis, it may be necessary to correct the potassium level after sensor insertion. Hemolysis occurs when damage due to sensor insertion, etc., ruptures cells and releases the contents of the cells into the plasma / serum. Since most potassium (98%) is stored intracellularly, hemolysis as a result of sensor insertion increases plasma and serum potassium levels, particularly in the area around the sensor insertion point. Thus, the measured potassium level may be elevated at the time of sensor insertion as a result of hemolysis, and a very high potassium level at the time of sensor insertion may be due to hemolysis. Accordingly, in certain embodiments, the decision - making support system 100 can exclude and / or correct the potassium levels measured during this period after potassium sensor insertion.

[0204] In certain embodiments, the correction of the potassium level measured by the decision - making support system 100 can be based on other analyte data including glucose and / or lactate data. Similar to the potassium level, the lactate level increases with sensor insertion and then normalizes. Thus, a rapid increase and recovery of the lactate level can indicate sensor insertion into the patient's body. Also, since the lactate level and the potassium level can be correlated, the normalization of the lactate level can be used to determine when the potassium level was normalized. Accordingly, when a multi - analyte sensor (lactate and potassium) determines a corresponding increase in the lactate level and the potassium level, the increase in both lactate and potassium may indicate sensor insertion rather than a systemic increase in potassium and / or lactate. In such an example, the decision - making support system 100 can then exclude and / or correct the elevated potassium level and / or lactate level during this period after sensor insertion.

[0205] In certain embodiments, the correction of sensor insertion can be based on a model used to predict the behavior of potassium levels after sensor insertion, such as a machine learning model. Such a model can predict the increase and decrease in the measured potassium levels due to hemolysis after sensor insertion. The model may be trained to predict the corrected potassium level after insertion of the potassium sensor. Deviation from the model for the measured potassium level after sensor insertion may also indicate renal dysfunction. For example, in a healthy patient, the potassium level may increase by X percent after sensor insertion and then take Y hours to recover (e.g., when hemolysis no longer affects the measured potassium).

[0206] In certain embodiments, the same device used to measure potassium levels, such as a multi - analyte sensor, or another device, can measure free hemoglobin and / or observe fluid abnormalities using colorimetric measurements. Based on this information (e.g., free hemoglobin level), the device can, in some cases, refrain from reporting the potassium level and / or warn the user that the potassium level is likely to be inaccurate. In some cases where the exact amount of free hemoglobin can be measured, a correction factor can then be applied to provide a more accurate potassium level for the patient.

[0207] The main analyte for the measurements described herein is potassium, but in certain embodiments, other analytes can be considered alone or in combination (e.g., in combination with potassium). Such analytes can include glucose, albumin, creatinine, lactate, blood urea nitrogen (BUN), inulin, dextran, saccharin, iothalamic acid, iohexol, 125I - iothalamate, C - peptide, 51Cr - EDTA, aspartic acid, polyfructosan, and / or betaine. However, other analytes can also be considered.

[0208] In certain embodiments, using the data of a plurality of analytes, including the data of the above-described analyte, in combination may help to further inform an analysis regarding the risk, presence, and / or stage classification of kidney disease (e.g., chronic kidney disease (CKD)), as compared to the data of a single analyte. For example, monitoring additional types of analytes in addition to potassium, as measured by the continuous analyte monitoring system 104, can provide additional insights into kidney disease-related predictions as compared to the insights derived from potassium alone. Such additional insights may include signs of other health conditions that can contribute to the progression of kidney disease, such as systemic inflammation, decreased systemic homeostasis, and liver disease.

[0209] The additional insights obtained from using combinations of analytes, rather than just a single analyte such as potassium, can improve the accuracy of predictions. For example, the probability of accurately predicting the risk, presence, and / or stage of kidney disease can be a function of the number of analytes measured for a patient. For example, in some instances, the probability of accurately predicting that a patient may have or develop kidney disease using only potassium data (in addition to other non-analyte data) may be lower than the probability of accurately predicting that a patient may have or develop kidney disease using both potassium and glucose data (in addition to other non-analyte data), which may also be lower than the probability of accurately predicting that a patient may have or develop kidney disease using potassium, glucose, and creatinine data (in addition to other non-analyte data) for the analysis.

[0210] Furthermore, using combinations of analytes enables the determination of various ratios related to the analytes (e.g., potassium-to-urea ratio, albumin-to-creatinine ratio, etc.), which can further inform the analysis regarding kidney disease. Such ratios can be determined based on the measured analyte values, analyte thresholds, analyte rates of change, analyte dispersions, analyte clearance rates, and / or any other analyte data related to a combination of analytes.

[0211] Accordingly, in certain embodiments described herein, for example, the analyte combinations measured and collected by one or more sensors (e.g., multi-analyte) for kidney disease-related prediction include at least two or more of potassium, glucose, albumin, creatinine, lactate, blood urea nitrogen (BUN), inulin, dextran, saccharin, iothalamic acid, iohexol, 125l-iothalamate, c-peptide, 51Cr-EDTA, aspartic acid, polyfructosan, and / or betaine. However, other combinations of analytes are also possible.

[0212] In certain embodiments, at block 402, the continuous analyte monitoring system 104 may continuously monitor the glucose level of a patient over multiple periods. In certain embodiments, since the kidney plays an important role in glucose regulation in addition to its role in potassium homeostasis, the measured glucose concentration can be used in conjunction with the potassium level to determine the risk, presence, and / or stage classification of kidney disease (e.g., CKD). For example, the kidney can raise blood glucose levels by producing glucose via gluconeogenesis and releasing glucose into the bloodstream. The kidney can also lower blood glucose levels by filtering glucose from the blood. However, most of the filtered glucose is then reabsorbed in the proximal renal tubules of the kidney as an energy source. Additionally, glucose levels typically have higher variability in the body than potassium levels (e.g., glucose levels vary more than potassium levels), and glucose has a higher "healthy" or normal range than potassium, so monitoring and analyzing glucose data can provide additional insights into kidney disease-related predictions compared to insights derived from potassium alone.

[0213] In certain embodiments, the measured glucose concentration can be used in conjunction with urine glucose measurements to determine the risk, presence, and / or stage classification of kidney disease. For example, the measured glucose concentration at which glucose appears in urine glucose measurements can assist in determining a kidney glucose threshold that may indicate kidney health. By monitoring the glucose concentration in conjunction with urine glucose measurements over time to determine the kidney glucose threshold over time, the decision support engine 114 can determine kidney health, kidney disease risk, and / or the progression of kidney disease over time.

[0214] In particular, glucose is a monosaccharide (e.g., a simple sugar). Glucose can not only be ingested, but can also be produced in the body from proteins, fats, and carbohydrates. An increase in glucose stimulates insulin release. Insulin causes cells to take up glucose and potassium as fuel. Thus, insulin stimulates the uptake of potassium and glucose 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 the glucose level in the patient's body is high, excessive insulin can be produced, which can thereby cause intracellular movement of potassium. On the other hand, when a patient's glucose level decreases and the rate of change of the glucose level in the patient's body is low, insulin secretion can be less. Low insulin can lead to restricted access of cells to glucose and potassium, and thus extracellular glucose and potassium levels can increase.

[0215] Insulin is partially removed from circulation by the kidneys. Thus, as kidney function declines, insulin is removed more slowly, and the release of insulin may have a more pronounced or prolonged decrease in glucose because the released insulin cannot be removed as quickly, so patients with kidney dysfunction are at high risk of hypoglycemia because in such patients insulin has higher activity and can cause a decrease in glucose levels below healthy concentrations.

[0216] Furthermore, patients suffering from renal dysfunction may have impaired gluconeogenesis in the kidneys, and thus may have a reduced ability to counteract a decrease in glucose levels. Here too, gluconeogenesis is the production of glucose from precursor molecules (e.g., lactate, glycerol, and / or amino acids) and is carried out by the liver and kidneys. Gluconeogenesis is one mechanism for maintaining glucose homeostasis in the body, the purpose of which is to prevent low blood glucose levels (i.e., hypoglycemia). However, when renal function declines, gluconeogenesis in the kidneys decreases, and thus the ability of the kidneys to respond to a drop in blood glucose is limited.

[0217] Furthermore, hyperglycemia (i.e., high blood glucose) is known to promote the progression of kidney disease. For example, it has been noted that diabetic patients with hypoglycemia have a slower progression of kidney disease than diabetic patients with hyperglycemia. However, due to the increased risk of severe hypoglycemia and death, best practice clinical guidelines suggest that patients with kidney disease should maintain blood glucose levels slightly higher than those that are clinically beneficial for reducing this risk.

[0218] Renal dysfunction significantly affects glucose homeostasis, and thus kidney disease is common in diabetic patients. In fact, diabetic patients are at high risk of kidney disease. However, diabetic patients may miss warning signs of renal dysfunction because changes in glucose imbalance or glucose control can be entirely, and wrongly, attributed to their diabetes rather than as a sign of kidney disease. In certain instances, patients with both diabetes and kidney disease may attribute all changes in glucose control to their diabetes instead of signs of the presence or progression of kidney disease.

[0219] Accordingly, glucose and related insulin can be indicators of kidney function. For example, if the glucose in the bloodstream increases over time, the kidneys may filter excessive blood. As time passes, this extra work places more pressure on the nephrons, and as a result, the nephrons often lose their ability to biofilter, thereby impairing kidney function. Therefore, the evaluation of glucose levels over time can provide insights into the overall health of a patient's kidneys, which can help determine whether the patient has kidney disease and the stage of the kidney disease, or whether the patient is likely to develop kidney disease in the future. Accordingly, in addition to the CPM202, the continuous analyte monitoring system 104 can include a continuous glucose monitor sensor (CGM), or a multi-analyte sensor configured to monitor both potassium and glucose to collect glucose data, and the glucose data includes glucose levels, timestamped glucose levels, glucose rate of change, glucose trend, glucose average, glucose management index, glycemic variability, time in range (TIR), glucose clearance rate, minimum and maximum glucose levels, glucose autocorrelation features, glucose setpoints, insulin clearance, and / or changes in glucose data.

[0220] In certain embodiments, glucose data collected by a CGM may be utilized in combination with A1C measurements to indicate the onset of renal dysfunction and / or the progression of a renal disease. A1C is a measure of the glycosylation of hemoglobin found in red blood cells (RBCs) as determined from a blood sample. More specifically, A1C is the percentage of glycated hemoglobin based on the assumed RBC half-life. Thus, A1C summarizes the duration of high blood glucose levels over the lifespan of the RBC. In patients without renal dysfunction, A1C measurements typically correlate with the monitored blood glucose levels. Accordingly, if the A1C measurement does not correlate with the monitored blood glucose levels (e.g., glucose time in range), the patient may be diagnosed with renal dysfunction and / or CKD. A decreasing correlation over time between the monitored glucose levels and the A1C measurement may indicate the onset of a renal disease and / or prompt further testing to confirm the health of the kidneys.

[0221] In some examples, CGM monitored blood glucose levels may be reported as a measurement known as the glucose management indicator (GMI). GMI is determined based on the average (e.g., mean) glucose level of a patient over a period of time. A decision support system, such as decision support system 100, can utilize the following relationship to calculate the GMI for a desired period based on the measured glucose levels of the patient during the period. GMI (%) = 3.31 + 0.02392 * [Average Glucose (mg / dl)]

[0222] Here too, for patients with normal and healthy kidneys (e.g., without kidney disease), GMI correlates with the clinically measured A1C value. However, in patients suffering from kidney disease (e.g., CKD), since kidney dysfunction reduces the half-life of RBCs, GMI does not correlate with the clinically measured A1C value. As a result of the shorter RBC half-life, the clinically measured A1C determined based on the assumed RBC half-life is lower than the patient's actual A1C level for patients with kidney disease. Otherwise, such A1C levels would correlate with higher glucose measurements in patients without CKD. Furthermore, A1C levels are further affected by erythropoietin treatment that may be prescribed for patients on dialysis due to CKD. Generally, erythropoietin treatment lowers the clinically measured A1C level below the patient's actual A1C level. Therefore, A1C measurement is less reliable as an indicator of the actual blood glucose level in patients suffering from kidney disease. A1C measurement is particularly unreliable in stage 3B kidney disease and more advanced conditions. However, although clinical A1C may lead to an underestimation of the actual blood glucose level in patients with kidney disease, since GMI is based on the actual blood glucose level, GMI measurement is very accurate even in patients with kidney disease. Therefore, as described above, patients with healthy kidneys have corresponding GMI and clinical A1C measurement values, while patients with kidney dysfunction show inconsistent measurement values for GMI and clinical A1C. Therefore, this discrepancy can be used as an indicator of kidney dysfunction. Therefore, GMI and clinical A1C measurements can be used to screen, diagnose, and / or stage kidney disease in conjunction with other diagnostic tools, such as potassium levels measured by CPM. Therefore, in certain embodiments, the decision support system 100 can warn the patient of the discrepancy between the GMI and the clinical A1C measurement value (via user input, EMR, etc.) and recommend further investigation of potential RBC dysfunction.

[0223] In certain embodiments, GMI and clinical A1C measurements can also be utilized to determine, in conjunction with renal disease, the presence or likelihood of other conditions associated with the affected RBC half-life (e.g., disorders of hemolysis). For example, in addition to renal disease, disorders of hemolysis are associated with the occurrence of oncological processes, the presence of advanced liver disease, and other conditions. Thus, in certain embodiments, the comparison of GMI measurements and clinical A1C measurements can be repeated for a patient over multiple periods (e.g., weeks, months, years) in order to passively monitor for disorders of hemolysis and thereby indicate the presence and / or likelihood of certain conditions. In such embodiments, if there is a risk that hemolysis is impaired, decision support alerts and / or recommendations can be provided to the patient, e.g., by decision support system 100. Further, if a patient has renal disease, changes in hemolysis can indicate a worsening of the renal disease or other serious conditions. Decision support recommendations for such patients can include alerts regarding the risk / change of hemolysis disorder, the risk of renal disease or worsening of renal disease, and / or the need for further investigation, which can include additional renal disease diagnostic triggers as discussed below.

[0224] In certain embodiments, the glucose metric can be a minimum and / or maximum glucose level. For example, the minimum and / or maximum glucose level can be based on glucose levels over, e.g., a day or a week.

[0225] In certain embodiments, the glucose metric represents an average glucose level that can be an average of two or more timestamped glucose levels. In certain embodiments, the average glucose level can be calculated based on glucose levels and other inputs 128 such as food consumption information, thereby determining the average glucose using the corresponding glucose levels and food consumption information (e.g., if timestamps are overlapping). The average glucose level can be calculated over a period (e.g., a day) and compared to the average glucose level of the next day.

[0226] In certain embodiments, the glucose data collected by the CGM and monitored by the decision support system 100 for screening, diagnosing, and / or staging kidney disease includes glycemic variability. Glycemic variability generally can include, in addition to time in range (TIR) data, the standard deviation of glucose levels over a period of time. TIR refers to one or more periods during which a patient's glucose level is within a particular desired range (e.g., a healthy range). The kidney's mechanisms that counter changes in glucose levels (i.e., gluconeogenesis and insulin clearance) are impaired by kidney injury and / or disease, so glucose homeostasis is disrupted in patients with kidney disease. In such patients, blood glucose levels have greater variability, which can result in an increase in glycemic variability. Thus, disruption of glucose homeostasis evidenced by an increase in glycemic variability can be an indicator of the presence and / or severity of kidney disease. For example, if the glycemic variability of a patient with unknown kidney dysfunction is higher during a subsequent second period compared to an initial first period, such a situation can indicate kidney dysfunction causing disruption of the patient's glucose homeostasis. In another example, if the glycemic variability of a patient with known kidney dysfunction (e.g., at CKD stage 3b) is higher during a subsequent second period compared to an initial first period, such a situation can indicate, for example, a decline in kidney function from CKD stage 3b to CKD stage 4. In patients with kidney disease, blood glucose can have greater variability due to kidney dysfunction. Greater glycemic variability results in an increase in glucose fluctuations. High glycemic variability can be due to higher and / or longer elevated glucose levels, as well as lower and / or longer decreased glucose levels.

[0227] In certain embodiments, the glucose metric indicative of glucose variability can be a setpoint metric. For example, the decision support engine 114 can determine a setpoint based on an estimation of the “mode” of the patient's glucose values (e.g., the glucose value that most frequently appears in the setting of glucose values). The setpoint can be determined, for example, based on population data of medical histories and / or glucose data of the patient's medical history. Based on the calculated setpoint, the glucose metric can further indicate the time within a range of glucose levels within a range of setpoint values.

[0228] In certain embodiments, the glucose metric can demonstrate a pattern or trend of glucose levels obtained at different times (e.g., separated by 5 minutes, separated by 10 minutes, etc.). The glucose metric can be an autocorrelation score that demonstrates the similarity of patterns and trends between glucose levels obtained at different times. The autocorrelation score can be a numerical value between 1.0 and 0.0, where 1.0 indicates that the time-series glucose levels are correlated (e.g., the patterns of glucose levels obtained at different times are very similar and / or the same), and 0.0 indicates that the time-series glucose levels are not correlated (e.g., the patterns of glucose levels obtained at different times are not similar and / or not the same).

[0229] In certain embodiments, the glucose data collected by CGM and monitored by the decision support system 100 for screening, diagnosing, and / or staging kidney disease may include glucose clearance rate. For example, when kidney function declines due to chronic kidney disease (CKD), the patient's glucose clearance rate typically changes. Thus, in certain embodiments, the decision support system 100 can compare glucose clearance rates taken at different times to determine whether the glucose clearance rate has changed over time. In such embodiments, the decision support system 100 can continuously (1) utilize historical data including glucose levels over time; (2) continuously engage the user in a glucose tolerance challenge; and / or (3) automatically detect glucose consumption and determine the glucose clearance rate therefrom to monitor the glucose clearance rate.

[0230] For example, at an initial time A, the decision support system 100 can determine a first glucose clearance rate for a patient based on glucose measurements provided by CGM. Then, at a subsequent time B, the decision support system 100 can determine a second glucose clearance rate for the patient that has decreased compared to time A, such that at time B, the patient's blood glucose level has remained elevated over a long period (e.g., above the glucose baseline) but then decreases to a level that is lower over a long period (e.g., below the glucose baseline). This change in glucose clearance rate can, as described above, indicate the presence or progression of kidney dysfunction and thus kidney disease. The decision support system 100 can alert the patient to the change and recommend, for example, consultation with a healthcare provider and / or administration of additional kidney function tests. Further, the decision support system 100 can incorporate the new (at time B) glucose clearance rate into predicted glucose levels, such as for optimizing future recommendations for the patient.

[0231] In certain embodiments, the glucose clearance rate can be analyzed in conjunction with or adjusted based on insulin data such as the insulin clearance rate, onboard insulin, administered insulin, and insulin sensitivity. Such insulin data may be based on insulin measurements provided by a continuous insulin sensor, a multi-analyte sensor, or other devices. In certain embodiments, the insulin clearance can be determined with reference to the glucose clearance rate. For example, the decision support system 100 can first determine the patient's glucose clearance rate based on the consumption of X amount of glucose by the patient. Next, the decision support system 100 can determine the patient's glucose-insulin clearance rate based on the consumption of X amount of glucose and Y amount of insulin by the patient. The difference between the glucose clearance rate and the glucose-insulin clearance rate for the same amount (X) of glucose can reveal the insulin clearance rate for the patient.

[0232] 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. Consequently, glucose metabolism as well as intracellular potassium movement may be impaired. As a result, the patient's pancreas produces more insulin to help glucose and insulin enter the patient's cells.

[0233] Insulin resistance can have different effects on glucose metabolism compared to potassium metabolism. Furthermore, the effects of insulin resistance on glucose metabolism and potassium metabolism can vary among different patients. In particular, a patient with a first insulin resistance may require X doses of insulin to reduce the extracellular potassium level by Y, while another patient with a second insulin resistance may require Z doses of insulin to reduce the extracellular potassium level by Y. For example, diabetic patients with insulin resistance may have a higher risk of hyperkalemia and may require higher insulin concentrations when using insulin for hyperkalemia management.

[0234] In other words, a patient's glucose level can 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 (which can measure potassium in the interstitial fluid), but the effect of insulin resistance on potassium metabolism can cause less than the expected amount of potassium to move into cells even when the patient's glucose and / or insulin is elevated. Therefore, accurately predicting a patient's potassium level and better understanding the true health of the patient's kidneys may require understanding the effects of insulin resistance not only on glucose metabolism but also on potassium metabolism for the patient. For example, in some cases, the glucose level, in addition to the patient's insulin resistance, can help understand whether the patient's abnormal potassium level is actually due to a decline in the patient's kidney health or some other reason such as an increase in insulin resistance that can be caused by heart disease, liver disease, obesity, etc.

[0235] In certain embodiments, other analyte data (e.g., C-peptide) may also be used in combination with and / or instead of insulin data, in combination with glucose data. For example, C-peptide measurements can be used to indicate endogenous insulin levels for determining the production of endogenous insulin, as well as the dosage of exogenous insulin levels in which a patient is being administered insulin. In certain embodiments, glucose, C-peptide, insulin and / or other data (analyte and / or non-analyte) may be used to indicate (1) levels of endogenous insulin; (2) exogenous insulin being administered; (3) insulin (endogenous / exogenous) clearance rate; (4) changes in glucose metabolic rate based on insulin amount; and / or (5) urinary glucose clearance based on estimated metabolic rate and insulin concentration dependency rate.

[0236] In another example, at block 402, the continuous analyte monitoring system 104 can continuously monitor a patient's creatinine level over one or more periods. In particular, creatinine is a waste product produced by muscle from the breakdown of a compound called creatine. Creatinine is removed from the body by the kidneys, which filter almost all of the creatinine from the blood and excrete it into the urine. Thus, creatinine levels can provide insight into kidney health and function. Thus, a patient who experiences high measured extracellular potassium levels and is expected to have impaired kidney function (e.g., considering that excessive potassium is not being filtered from the body) may also be expected to experience high measured creatinine levels (e.g., considering that a damaged kidney is likely unable to remove creatinine from the blood). Thus, in certain embodiments where creatinine levels are monitored in combination with potassium levels, the measured creatinine level can be used to assign a confidence level to the patient's measured potassium level, where the confidence level indicates the level of certainty that the measured potassium level for the patient reflects the patient's actual potassium level. For example, if the measured potassium and creatinine levels are high, a higher reliability level can be assigned to the measured potassium level for the patient. Further, the assumption that the patient's kidneys are damaged can be strengthened, thereby increasing the likelihood that the patient's kidneys are not functioning properly and increasing the likelihood that the patient actually has a kidney disease.

[0237] In certain embodiments, the creatinine clearance rate can be used in combination with an mGFR test. Creatinine clearance measurements can be useful for distinguishing changes in the secretory and filtration functions of the kidneys, as both secreted and filtered creatinine can be measured: creatinine is secreted in the proximal tubules of the kidneys and further filtered through its glomeruli. Thus, the difference between the mGFR measurement and the creatinine clearance rate over the same period can be used to determine the secretory function of the kidneys. Changes in secretory function can indicate the health of the tubules and the risk or presence of tubulointerstitial fibrosis.

[0238] In another example, at block 402, the continuous analyte monitoring system 104 can continuously monitor a patient's albumin level over one or more periods for screening, diagnosing, and / or staging kidney disease. Albumin levels generally do not vary significantly throughout the day, whether in blood or interstitial fluid, and the normal range of albumin is 3.4 to 5.4 g / dL, with a turnover period of approximately 25 days. Due to the relative stability of albumin levels in the body, a rapid change in albumin levels can indicate kidney dysfunction. Thus, in certain embodiments, the decision support system 100 can determine an albumin change rate and / or albumin variability based on the measured albumin levels. For example, an increase in the variability of urine albumin levels (e.g., a higher standard deviation) over a one-day period can indicate the presence and / or progression of kidney disease.

[0239] In certain embodiments, albumin measurements can be used in combination with creatinine measurements for screening, diagnosing, and / or staging kidney disease. For example, real-time or continuous measurements of creatinine levels and / or change rates can be provided over minutes, hours, and / or days in interstitial fluid or blood and used in combination with a baseline albumin level to indicate kidney health. Since the change rates of albumin and creatinine occur on different timescales, abnormal changes in albumin levels and creatinine levels over a given period (e.g., 24 hours), and / or the ratio, change rate, or trend between albumin levels and creatinine levels can be used to indicate kidney dysfunction and / or changes in kidney function.

[0240] In certain embodiments, albumin measurements can be used in combination with creatinine and potassium measurements for screening, diagnosing, and / or staging kidney disease. More specifically, potassium, creatinine, and albumin measurements can be used in combination to confirm that fluctuations in the levels of one or more of these analytes are due to kidney disease.

[0241] In yet another example, in block 402, the continuous analyte monitoring system 104 can continuously monitor a patient's urea level (e.g., blood urea nitrogen (BUN) level) over one or more periods. Urea is synthesized in the liver and removed by the kidneys. In particular, the liver produces ammonia containing 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, which is a chemical waste product. Urea moves from the liver to the kidneys through the bloodstream. Healthy kidneys filter urea and remove other waste products from the blood, and the filtered waste products exit the body through urine. Thus, urea levels can provide insight into kidney health and function.

[0242] A decrease in urea synthesis can indicate liver disease rather than kidney dysfunction, but a decrease in urea synthesis is typically only seen in end-stage liver disease, and thus liver disease can be excluded as a cause with low potential for changes in urea levels. Thus, in the absence of end-stage liver disease, changes in a patient's urea level can reliably indicate changes in urea clearance by the kidneys. Thus, for users without end-stage liver disease, the decision support system 100 can utilize urea levels to determine the urea clearance rate, which can be used as an indicator of the risk, presence, and / or progression of kidney disease. In certain embodiments, the urea clearance rate can be used in conjunction with the creatinine clearance rate as an alternative to mGFR measurement. For example, urea clearance and creatinine clearance can be monitored for one hour and then the clearance rate can be calculated. Such a determination can be as effective as mGFR measurement. However, to enhance effectiveness, urea clearance and creatinine clearance can be monitored for more than 24 hours.

[0243] Furthermore, in certain embodiments, the urea reabsorption rate can be used as a surrogate for a patient's hydration or renal blood flow, thereby indicating the progression and stage of kidney disease.

[0244] In yet a further embodiment, the measured urea level can be used to assign a confidence level to the patient's measured potassium level. Thus, a patient who experiences a high level of measured extracellular potassium and is assumed to have impaired kidney function may also be expected to be experiencing a high level of measured extracellular potassium (e.g., considering that a damaged kidney is likely unable to filter urea and remove other waste products from the blood). Thus, if the measured urea level is also high, a higher confidence level can be assigned to the potassium level measured for the patient. Further, the assumption that the patient's kidney is damaged can be strengthened, thereby increasing the likelihood that the patient's kidney is not functioning properly and increasing the likelihood that the patient is actually experiencing (or at risk of) kidney disease.

[0245] In yet another example, at block 402, the continuous analyte monitoring system 104 can continuously monitor the patient's C-peptide level over one or more periods. In certain embodiments, the measured C-peptide level can be used to determine the patient's endogenous insulin level and to provide additional context regarding the potassium measurements collected by the continuous analyte sensor 202.

[0246] In particular, C-peptide is a substance that is created when insulin is produced and released into the body. Since there is currently no way to measure insulin in the body, measuring the level of C-peptide in the blood can indicate how much insulin is being produced by the pancreas. Often, C-peptide is measured to show the difference between the insulin produced by the body and the insulin injected into the body. Understanding the patient's insulin level can provide the patient with insight into why the patient's measured potassium level indicates that it is lower or higher than normal, in order to better evaluate the patient's kidney health.

[0247] In yet another example, at block 402, the continuous analyte monitoring system 104 can continuously monitor lactate levels. Lactate levels may be related to glucose, insulin, and potassium metabolism. Lactate levels can also be used to detect the consumption of food, exercise, rest, and / or stress. Thus, additional insights obtained from lactate levels can improve the analysis of other analyte data and trends by correlating such analyte measurement criteria with different body states (e.g., food consumed, exercise, rest, stress, etc.).

[0248] In certain embodiments, at block 402, the continuous analyte monitoring system 104 can continuously monitor one or more analytes that indicate a disorder of either the kidney filtration function or the kidney secretion function in a patient having kidney disease, or that can distinguish between the kidney filtration function and the kidney secretion function. For example, the continuous analyte monitoring system 104 can continuously monitor one or more analytes that can be used as a marker for, or in combination with, a glomerular filtration rate (GFR) test to determine the kidney filtration function. Ideal GFR markers include small analytes that are retained in the vasculature, not bound to proteins, and freely filtered across the glomeruli of the kidney. Since such markers are not reabsorbed, secreted, or metabolized by the kidney, the measured GFR is equal to the urinary clearance of the marker after intravenous injection into the patient. Further, ideal GFR markers are also generally recognized as generally regarded as safe (GRAS), include relatively high oral bioavailability, and are removed by the kidney without intervention of other metabolic pathways or interactions.

[0249] An example of a nearly ideal GFR marker for continuous monitoring by the continuous analyte monitoring system 104 to inform analysis of the risk, presence, and / or progression of kidney disease is inulin. Inulin is an analyte typically measured in an mGFR test. Thus, continuous inulin measurements can be used in conjunction with, or instead of, a standard mGFR test to improve mGFR results. For example, in certain embodiments, rather than subjecting a patient to a standard mGFR test (where the patient must stay at a clinic for several hours and provide several blood samples), one or more continuous analyte sensors 202 of the continuous analyte monitoring system 104 can include a continuous inulin sensor for continuously monitoring inulin levels over a given period for analysis by the decision support system 100. In this example, patient convenience is improved because monitoring of inulin via the continuous inulin sensor is less time-consuming and invasive than a conventional mGFR test. Further, utilization of the continuous inulin sensor 202 reduces the loss of time to rapid decline in the early stages, allows for a longer sampling period, and thereby can reduce the effects on mGFR measurements caused by time (during sampling), patient posture, and the patient's diet. Further, utilization of continuous potassium measurements in combination with continuous inulin measurements for an mGFR test can improve the reliability and analysis of the measured inulin clearance.

[0250] Another example of a potential marker that can be continuously monitored by a continuous analyte monitoring system 104 for use with a GFR test to inform the analysis of kidney disease is dextran. High molecular weight dextran (such as 150 kDa and above) remains in the blood and is not filtered by the kidneys. Thus, such high molecular weight dextran can be monitored to determine plasma volume as its measurement can be utilized to quantify plasma volume distribution based on the principle of dilution. On the other hand, low molecular weight dextran (such as 5 kDa) distributes into the interstitial space and is then filtered by the kidneys but is not further metabolized or distributed intracellularly. Thus, the plasma concentration of low molecular weight dextran decreases as a function of time and depends on both renal clearance and redistribution into the vasculature and extracellular fluid, so low molecular weight dextran can be monitored to determine the renal filtration rate. However, dextran must be administered intravenously or subcutaneously (i.e., without oral bioavailability), which limits its usefulness as a GFR marker.

[0251] Yet another example of a potential GFR marker for continuous monitoring by a continuous analyte monitoring system 104 to inform the analysis of kidney disease is saccharin. Saccharin is a small molecule with good oral bioavailability (about 98%) and thus enables non-clinical GFR measurements (e.g., at home). Saccharin is removed by the kidneys without reabsorption and minimal protein binding. Since saccharin has no other metabolic pathways, it can be monitored to exhibit various kidney functions over time, thereby enabling the diagnosis and staging of kidney dysfunction over time. However, saccharin is also secreted by the kidneys, which can confound single-point GFR measurements. Nevertheless, saccharin can be a reliable compound for GFR estimation based on continuous measurements.

[0252] In addition to, or instead of, the GFR markers described above, to provide additional insights into the analysis of the risk, presence, and / or progression of kidney disease, the continuous analyte monitoring system 104 can monitor non-radioactive markers such as iothalamic acid, iohexol, and polyfructosan, and other GFR markers including radioactive markers such as 125I-iothalamic acid, and 51Cr-EDTA.

[0253] In certain embodiments, for example, one or more of the algorithms and / or models described herein for predicting the risk, presence, and / or progression of kidney disease can be configured to use inputs from one or more sensors that measure one or more of the plurality of analytes described above. The parameters and / or thresholds of such algorithms and / or models can be varied based at least in part on the number of analytes being measured for input, to reflect the knowledge obtained from each of the other analytes being measured.

[0254] In certain embodiments, in addition to continuously monitoring one or more analytes of a patient at block 402 above, one or more analytes can be monitored / measured by a urine test. For example, chemical analysis or photographic analysis can be used to analyze exogenous analytes present in urine. Exogenous analytes include analytes that are not naturally found in the body and can be effectively removed from the body by a healthy kidney in a short period of time. Although not naturally found in the human body, these analytes can be naturally occurring, such as aspartic acid and betanin. Aspartic acid is a sulfur-containing compound that can be detected by a point-of-care electrochemical assay for a certain volume of urine that varies over time compared to a known ingestion time point and is easily excreted from the body during the consumption of asparagine. Betanin is another naturally occurring analyte commonly found in beets. At high concentrations, betanin can change the color of urine to blood red, and this color change can be easily observed by the naked eye or by electronic means at a certain time point compared to a known ingestion time point. Measuring the user's response to a known ingestion can also be a means of understanding renal function clearance over time, or the brightness of the urine can be set based on the concentration of the urine excreted at a specific time point. By comparing this urine brightness to a normalization curve, the difference between the red color expected to be cleared during the test period and the actual red color cleared can be determined. The difference between the actual red color and the expected red color can help determine the functional clearance rate of the kidneys and renal function.

[0255] For example, at block 402, a patient can consume a known amount of betanin and then capture the amount or concentration of betanin later found in the patient's urine. In one iteration, the patient can take a photograph of a serving of beets before consuming the beets. Then, the patient can take a photograph of the urine to capture the color change caused by betanin and input such a photograph into the decision support system 100. The decision support engine 114 can then analyze the photograph of the serving of beets to determine the initial consumption amount of betanin and further analyze the photograph of the user's urine to determine the concentration of betanin present in the patient's urine.

[0256] Similarly, aspartic acid contains a heterocyclic disulfide functional group that can be chemically detected in urine. One method of chemical detection involves the use of a urine test strip that includes a pad or reagent configured to react with a specific concentration of aspartic acid to change color. Another method of chemical detection involves the use of chemical reactants in toilet water, which also change color when exposed to aspartic acid. Thus, as in the example above, at block 402, the patient can capture a photo of the serving before consuming the aspartic acid and then, after urination, capture a photo of the test strip (including the chemical reactants) or the toilet water. Such photos are input into the decision support system 100 and analyzed by the decision support engine 114 to determine the initial consumption amount of aspartic acid present in the patient's urine as well as the concentration of aspartic acid.

[0257] Furthermore, in certain examples, an initial amount of aspartic acid or betaine can be consumed in the form of a capsule, pill, or beverage, such that the initial amount of any analyte being cleared is known. Thereafter, based on the resulting urine concentration, the analyte clearance of aspartic acid or betaine can be determined. Comparison of the expected clearance rate and expected detection time of the analyte with the actual clearance rate and actual detection time can indicate renal function and / or the need for further investigation of renal function.

[0258] In optional block 404, method 400 continues by optionally monitoring non-analyte sensor data using one or more non-analyte sensors or devices (such as, for example, non-analyte sensor 206 and / or medical device 208 of FIG. 2) during one or more periods.

[0259] As previously mentioned, the non-analyte sensors 206 and devices can 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 respiratory sensor, a thermometer, a pulse oximeter, an impedance sensor, 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 smartwatch), or any other sensor or device that provides relevant information about the user. One or more of these non-analyte sensors 206 and devices can provide data to the aforementioned decision support engine 114. In some embodiments, a user, such as a patient, can manually input data for processing by the decision support engine 114.

[0260] One or more specific metrics of the metric 130 shown in FIG. 3 can be calculated using measurement data from each of these additional sensors. Further, as shown in FIG. 3, one or more of the metrics 130 calculated from non-analyte sensor or device data can include body temperature, heart rate (including heart rate variability), respiratory rate, and the like. In certain embodiments described in more detail below, one or more of the metrics 130 calculated from non-analyte sensor or device data can be used to further inform an analysis related to kidney disease prediction.

[0261] In certain embodiments, one or more non-analyte sensors and / or other devices can be worn by a user to assist in detecting periods of increased physical movement by the user. Such non-analyte sensors and / or devices can include an accelerometer, an ECG sensor, a blood pressure sensor, a heart rate monitor, an impedance sensor, a dialysis machine, and the like. In certain embodiments, data measured and collected from periods of increased physical activity and periods of sedentary activity by the user can be used to analyze, for example, potassium levels, glucose levels, lactate levels, etc. between each of these specified periods to inform kidney disease prediction.

[0262] In certain embodiments, one or more non-analyte sensors and / or devices that can be worn by a patient can include a temperature sensor. The temperature sensor can be worn, for example, to assist in correcting potassium levels measured to predict the risk, presence, and / or progression of a patient's kidney disease. In particular, there is a correlation between body temperature and potassium release in the human body.

[0263] For example, at higher body temperatures, the body sweats and potassium is excreted through the sweat. Thus, higher body temperatures can induce lower potassium levels. In some cases where a patient's measured potassium level appears to be lower than normal for that patient, the patient's lower-than-normal potassium level may be due to an impaired tubular response to vasopressin (ADH) and the inability to concentrate urine, which in some cases can lead to the excretion of large volumes of dilute urine containing potassium. However, such decreased potassium levels may actually be due to excessive sweating, for example, excessive potassium exiting the patient's body. In other words, a patient's high body temperature can affect the amount of extracellular potassium measured by a continuous potassium sensor, such as a CPM202. Thus, in some aspects, a secondary sensor, such as a temperature sensor, can be used to indicate that a dynamic and sudden change in potassium can be the result of sweating (e.g., related to exercise, hot and / or humid weather, etc.) as compared to a negative health event.

[0264] As another example, in some cases, potassium levels measured in a patient who is initially experiencing hypothermia (e.g., which occurs when the body loses heat faster than it can produce heat, causing a dangerously low body temperature) may appear lower than the normal potassium levels associated with the patient. Such decreased potassium levels may be due to the patient experiencing the onset of hypothermia. In particular, hypothermia can cause an initial decrease in extracellular potassium levels. Hypothermic hypokalemia is related more to intracellular shift than to actual net loss. Intracellular shift is caused by various factors such as enhanced function of Na+K+ATPase, beta - adrenergic stimulation, pH, and membrane stabilization in deep hypothermia.

[0265] However, as hypothermia progresses in a patient, irreversible cell damage can occur. In particular, the body can experience a lack of enzymes that function at low temperatures and a disruption of active transport. Thus, as hypothermia progresses, the measured potassium level in the patient can increase from a level lower than normal to a level higher than normal. In other words, the patient's low body temperature can affect the amount of extracellular potassium measured by CPM202 over time. Therefore, monitoring the patient's body temperature can help to inform the measured potassium level in the patient so that it can be corrected before the measured potassium level predicts kidney disease.

[0266]

[0267] In certain embodiments, one or more non-analyte sensors and / or devices that can be worn by a patient may include a blood pressure sensor. Blood pressure measurements collected from the blood pressure sensor can be used to provide additional insight into the patient's kidney health. In particular, kidney disease and high blood pressure are closely related. Typically, as blood pressure rises, kidney function declines. Thus, a patient who is hypothesized to have kidney dysfunction as indicated by high levels of measured extracellular potassium (e.g., excessive potassium that is not being filtered from the body) may also be expected to experience high blood pressure levels. Thus, if the patient's blood pressure level is also high, the hypothesis that the patient's kidneys are damaged can be strengthened, thereby increasing the likelihood that the patient's kidneys are not functioning properly and that the patient is actually suffering from kidney disease.

[0268] In certain embodiments, one or more non-analyte sensors and / or devices that can be worn by a patient may include an ECG sensor and / or a heart rate monitor. As is known in the art, an ECG device is a device that measures the electrical activity of the heart. In certain embodiments, the heart rate measurements and heart rate variability information collected from the ECG sensor and / or heart rate monitor can be used in combination with the CPM to better inform the assessment of kidney health. In particular, the patient's potassium level measured using the CPM can be used to detect hyperkalemia or hypokalemia. The patient's ECG measurements may be essential in combination with the CPM measurements to provide a holistic assessment of the physiological significance of hyperkalemia or hypokalemia.

[0269] In block 406, method 400 continues by processing analyte data from one or more periods and, in certain embodiments, other non-analyte sensor data, to determine at least one analyte trend or analyte rate of change for the patient. Block 406 may be performed, in certain embodiments, by decision support engine 114.

[0270] As described above, the trend or rate of change of an analyte indicates the change in one or more timestamped metrics, measurements, or values of the analyte relative to one or more other timestamped measurements or values of the analyte. In certain embodiments, the machine learning models described herein used to provide kidney disease-related predictions can include not only the patient's analyte levels, but also one or more features related to the trend of the analyte levels and the rate of change of the patient's analyte levels. For example, an exemplary machine learning model can include weights applied to features related to one or more trends or rates of change of, for example, potassium levels. Thus, in certain embodiments, prior to use of the machine learning model, at least one rate of change of potassium levels for the patient may need to be calculated for input into the model.

[0271] Furthermore, in certain embodiments, the rule-based models described herein used to provide kidney disease-related predictions can include not only rules related to the patient's analyte levels, but also one or more rules related to the trend of the patient's analyte levels and the rate of change of the analyte levels. For example, a reference library used to define one or more rules of a rule-based model can maintain, for example, a range of potassium levels and a range of rates of change of potassium levels, which can be mapped to different stages of kidney disease. Thus, prior to using the rule-based model, it may be necessary to calculate at least one rate of change of potassium levels for the patient for input into the model. Additional features such as the rate of change of potassium levels added to the model may, in some cases, enable a more accurate prediction of the patient's kidney disease risk, presence, and / or progression.

[0272] In block 408, method 400 continues by generating a kidney disease prediction, which can include (1) the likelihood that the patient is experiencing (or will experience) abnormal kidney function, (2) the risk of kidney disease, and (3) the presence and / or stage of the patient's kidney disease using (a) at least one analyte trend or analyte rate of change of the patient (e.g., determined at block 402) and (b) a trained model or one or more rules (e.g., a rule-based model).

[0273] Different methods for generating renal disease predictions can be used by the decision support engine 114. In particular, in certain embodiments, the decision support engine 114 can use a rule-based model to provide real-time decision support for the risk assessment, diagnosis, and stage classification of renal diseases. As previously mentioned, a rule-based model involves using a set of rules for analyzing data. In particular, the decision support engine 114 can apply rule statements (e.g., if, then statements) to evaluate 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.

[0274] For example, one rule can be related to the current patient's absolute maximum potassium level or based on the change in the absolute maximum potassium level over time. Another rule can be related to the current patient's absolute minimum potassium level or based on the change in the absolute minimum potassium level over time. Another rule can be based on the change in the patient's potassium baseline over time. Another rule can be related to the patient's potassium level change rate, whether the potassium level change rate 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 change rate over time. Another rule can be related to glucose metrics, insulin metrics, creatinine metrics, BUN metrics, albumin metrics, dextran metrics, inulin metrics, saccharin metrics, iothalamic acid metrics, iohexol metrics, 125l-iothalamate metrics, 51Cr-EDTA metrics, lactate metrics, aspartic acid metrics, polyfructosan metrics, betaine metrics, and / or C-peptide metrics, as described with respect to FIG. 3 or based on the change in such metrics over time.

[0275] Another rule may relate to a patient's glucose response or lack thereof to biochemical hypoglycemia (e.g., less than 70 mg / dL), whether or not circulating insulin can be measured. Another rule may relate to whether a patient experiences an acute or rapid increase in creatinine concentration due to acute kidney injury (AKI) (e.g., blood loss, vomiting, diarrhea, heart failure, etc.). Another rule may relate to a patient's potassium clearance rate following known or estimated potassium consumption. For example, in patients with renal dysfunction, the clearance rate may be slower than that of healthy controls.

[0276] Another rule may relate to the absolute maximum potassium level after known or estimated potassium consumption. For example, the absolute maximum potassium level following known or estimated potassium consumption may be greater in patients with renal dysfunction than in those of healthy controls. Another rule may relate to the correlation between a patient's absolute maximum potassium level and the patient's potassium clearance rate. For example, an increase in the absolute maximum potassium level accompanied by reduced potassium clearance may be observed as kidney 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 together with each other or in combination with one or more other rules when using a rule-based model.

[0277] Such rules may be maintained in a reference library by the decision support engine 114. For example, the reference library may maintain ranges of analyte levels and / or rates of change that may be mapped to different severities of kidney disease. In certain embodiments, such rules may be determined based on empirical studies and analysis of historical patient records from the historical record database 112.

[0278] In certain embodiments, as an alternative to using rule-based models, AI models such as machine learning models can be used to provide real-time decision-making support for the risk assessment, diagnosis, and stage classification of kidney diseases. In certain embodiments, the decision support engine 114 can deploy one or more of these machine learning models to perform screening, diagnosis, stage classification, and risk stratification of a patient's kidney disease. Risk stratification can refer to the process of assigning a health risk status to a patient and using the risk status assigned to the patient to direct and improve care.

[0279] In particular, the decision support engine 114 can obtain information from the user profile 118 associated with the patient, which is stored in the user database 110, characterize the information about the patient stored in the user profile 118 in terms of one or more features, and use these features as inputs to such a model. Alternatively, the information provided by the user profile 118 can be characterized by another entity and then the features can be provided to the decision support engine 114 to be used as inputs to the ML model. In certain embodiments, features related to the patient can be used as inputs to one or more models to assess the risk, presence, and / or severity of the patient's kidney disease.

[0280] In certain embodiments, features associated with a patient can be used as input into one or more of the models for risk stratifying the patient to identify whether the patient is at high or low risk of developing a kidney disease (e.g., CKD). In certain embodiments, features associated with a patient can be used as input into one or more of the models to identify risks associated with the patient's current kidney 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 can be used as input into one or more of the models to perform any combination of the above-described functions. Details regarding how one or more machine learning models can be trained to provide real-time decision support for kidney disease risk assessment, diagnosis, and / or stage classification are discussed further in connection with FIG. 5.

[0281] As described above, in certain embodiments, at block 506, in addition to analyte data, non-analyte sensor data can be used by decision support engine 114 to generate a prediction of a patient's kidney disease. For example, data provided by 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, a pulse oximeter, an impedance sensor, a peritoneal dialysis machine, a hemodialysis machine, 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 smartwatch), or any other sensor or device that provides relevant information about the user can be used as input into such machine learning models and / or rule-based models to predict the risk, presence, and / or severity of a user's kidney disease.

[0282] The decision support engine 114 can generate a kidney disease prediction based on a continuous analysis of patient data (e.g., analyte data and optionally non-analyte data) collected over one or more periods using a machine learning model and / or a rules-based model. Analysis of data collected about a patient over various periods can provide insights into whether the patient's kidney health and / or disease is improving or deteriorating. For example, patients previously diagnosed with chronic kidney disease using the models discussed herein can continue to be continuously monitored (e.g., collected continuously for the patient) to determine whether the disease is worsening or improving, etc. As an example, comparison of analyte data (glucose levels, timestamped glucose levels, glucose baseline, absolute maximum glucose level, absolute minimum glucose level, rate of change of glucose level, glucose metrics (e.g., glucose setpoint metric, glucose autocorrelation score, etc.), TIR, average glucose, GMI, and / or glycemic variability) and / or other sensor data over multiple months can indicate the progression of the patient's disease. For example, the decision support engine 114 can determine that a patient is at risk of kidney disease if the patient's absolute minimum glucose level begins to decrease over time, especially when the patient is sleeping.

[0283] For example, the decision support engine 114 can provide the likelihood that a patient is experiencing abnormal kidney function or is at risk of kidney disease based on the patient's timestamped glucose levels. For example, users at risk of developing kidney disease may experience, over time, higher daytime glucose levels, a greater number of daytime hyperglycemic events, higher postprandial glucose levels, lower nighttime glucose levels, and / or a greater number of nighttime hypoglycemic events, which can demonstrate a worsening of glycemic control and thus the presence of kidney disease.

[0284] In certain embodiments, a low minimum glucose level over time and / or a high maximum glucose level over time can demonstrate that a patient has developed and / or is experiencing a worsening of kidney disease. For example, a patient with kidney disease can be expected to experience hyperglycemic spikes after dinner and / or when the patient goes to bed. If a patient develops a hyperglycemic spike after dinner and / or when the user goes to bed over time, the decision support engine 114 can determine that the patient has developed and / or is at risk of developing kidney disease.

[0285] In certain embodiments, a lower autocorrelation score over time (e.g., less than 0.5) can demonstrate a worsening of kidney disease. In particular embodiments, a higher maximum glucose level and a lower minimum glucose level can demonstrate a worsening of kidney disease. In particular embodiments, a higher rate of glucose change over time can demonstrate that a patient has developed and / or is experiencing a worsening of kidney disease. In particular embodiments, a higher average glucose level over time and / or a higher standard deviation of glucose levels over time can demonstrate that a patient has developed and / or is experiencing a worsening of kidney disease.

[0286] In certain embodiments, the decision support engine 114 can use set point metrics to determine the patient's kidney disease stage. For example, as kidney disease worsens, the variability of glucose measurements increases, and as a result, the glucose level time within the range, particularly within the set point range, may become shorter. As the glucose levels within the set point range become fewer, the decision support engine 114 can determine that the user's kidney disease is progressing.

[0287] In some cases, method 400 continues in block 410 where the decision support engine 114 generates one or more recommendations for treatment, at least partially based on the disease prediction generated in block 408. In particular, the decision support engine 114 may provide recommendations for the treatment or prevention of kidney disease, such as lifestyle recommendations, pharmaceutical recommendations, service intervention recommendations, or other recommendations for managing kidney health. The decision support engine 114 may then output such recommendations for treatment to the user (e.g., through application 106).

[0288] In certain embodiments, one or more recommendations generated by the decision support engine 114 include alerts regarding normal or abnormal analyte levels, analyte thresholds, analyte rate of change, analyte clearance rate, and / or analyte variance; risk of developing future kidney disease (e.g., CKD) (e.g., screening for kidney disease risk); risk of current presence of kidney disease (e.g., diagnosis of kidney disease); prediction regarding current stage of kidney disease (e.g., stage classification); risk of adverse events and / or death (e.g., risk stratification); risk of adverse health events (e.g., cardiac events, hyperkalemia and / or hypokalemia); risk of kidney health adverse events (e.g., hyperkalemia and / or hypokalemia); recommendations for additional diagnostic tests; recommendations for the treatment or prevention of kidney disease, including diet, pharmaceutical, lifestyle, alarm / alert, and service intervention recommendations; and / or risk of other health conditions other than kidney disease (e.g., liver disease).

[0289] Recommendations for the treatment or prevention of kidney disease may, in some cases, be based, for example, on the determined optimal balance of a 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 a patient's potassium and insulin levels, which is then used to form one or more recommendations for the patient regarding diet, lifestyle changes, treatment, insulin dosage, and / or pharmaceuticals.

[0290] In certain embodiments, the 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. In certain embodiments, the dietary recommendations may include recommending to the patient to consume potassium-containing foods at specific times, such as before insulin administration or before exercise. Monitoring potassium consumption helps ensure that excessive levels of potassium are not pushed into cells (e.g., due to excessive insulin) while also ensuring that the consumed potassium can be removed by the patient's kidneys.

[0291] In certain embodiments, the dietary recommendations may include recommendations for the patient to increase their potassium consumption. For example, if excessive insulin and / or one or more diuretics are causing a significant drop in measured extracellular potassium (e.g., which can lead to hypokalemia), the patient may be recommended to increase their potassium consumption.

[0292] In certain embodiments, lifestyle recommendations (e.g., exercise recommendations and / or exposure to a sweating-promoting environment such as a sauna) may include recommendations for the patient to increase physical activity daily, weekly, etc., considering that an increase in physical activity can be one way to remove excess potassium from the body, e.g., through sweating. For example, for patients with hyperkalemia, one or more models can be used to determine when such patients should engage in physical activity based on the patient's potassium level, renal function, and / or current insulin production / injection. The patient can be recommended to follow a modified physical activity schedule or engage in additional rest based on the determined schedule around when the patient should engage in physical activity. In certain embodiments, the exercise recommendations can be made to optimize sweating while restricting the patient's body movement to reduce potassium, while preventing an increase in potassium due to the patient's exercise activity. For example, the recommendation can be to spend time stimulating sweating without requiring the patient's physical movement. In certain embodiments, other analyte data or non-analyte data, such as heart rate or respiratory rate data, can be used in combination with potassium data to provide such exercise recommendations.

[0293] In certain embodiments, the treatment recommendations may include recommendations for dialysis for the patient. In certain embodiments, determining the optimal balance of the patient's potassium and insulin levels can help inform whether dialysis is a recommended treatment for the patient. In particular, dialysis is a treatment for kidney failure that sweeps unwanted toxins, waste products, and excess fluid from the body by filtering the patient's blood. Dialysis helps maintain the balance of potassium, phosphorus, and sodium levels in the patient's body. Thus, understanding the optimal balance of potassium and insulin in the body can help inform such treatments for which dialysis is recommended.

[0294] In certain embodiments, service intervention recommendations may include recommendations for a patient to seek medical attention. For example, in certain embodiments, service intervention recommendations may indicate to the patient, or another individual interested in the patient's health, that the patient needs to go to the emergency room immediately and / or needs to contact the patient's healthcare provider. In certain other embodiments, service intervention recommendations may automatically alert the patient's healthcare provider regarding the patient's condition with respect to a physician's intervention. In certain other embodiments, service intervention recommendations may alert healthcare providers to send assistance to the patient, such as by triggering an ambulance service or emergency medical service to provide the patient with emergency pre-hospital treatment and stabilization and / or to transport the patient to definitive care. In certain embodiments, the decision support engine 114 may make service intervention recommendations based on the patient's ability to seek medical assistance and / or the patient's access to medical relief.

[0295] In certain embodiments, insulin dosing recommendations may include recommendations for combination dosages such as insulin / glucose (e.g., to prevent hypoglycemia). In certain embodiments, one or more algorithms may be used to determine the combination dosages recommended for a patient. In particular, insulin can be used to treat hyperkalemia such that increased insulin intake can be used to reduce extracellular potassium. In CKD patients with diabetes who are receiving insulin and have elevated potassium levels, an algorithm can be created and used for administration of insulin to avoid or treat hyperkalemia while also preventing hypoglycemia. For example, the algorithm can calculate the amount of insulin necessary to reduce the potassium level by a value X (e.g., X is a value greater than zero), as well as the amount of glucose and the timing of glucose consumption to prevent hypoglycemia. Further, insulin dosing recommendations can be individualized for each patient to account for differences in insulin resistance across patients. For example, insulin resistance changes the ability of insulin to push potassium intracellularly, and thus patients with higher insulin resistance may require different insulin dosages compared to another patient with a different level of insulin resistance. In certain aspects, individualized insulin dosing recommendations can be modified over time in a patient as insulin resistance increases or decreases.

[0296] In certain embodiments, a medication recommendation may include recommending to a patient to take a new medication if the patient has not previously taken a similar medication. In certain embodiments, a medication recommendation may include a recommendation to discontinue taking a previously prescribed medication by the patient, and in some cases, may recommend an alternative medication for the patient's consumption. In certain embodiments, a medication recommendation may include recommending to a patient to take a lower or higher dosage of a previously prescribed medication. In certain embodiments, a medication recommendation may include a recommendation for titration of a previously prescribed dosage or timing of dosage for a patient in order to determine the patient's optimal dosage (e.g., while monitoring the user's kidney and heart health). In certain embodiments, recommendations regarding medications may be generated to reduce the risk of adverse health events.

[0297] In certain embodiments, the decision support engine 114 may determine that CKD is progressing in a patient and may correlate such progression with medications previously prescribed to the patient. The decision support engine 114 may make this determination based on the entered medication consumption information for the patient (in combination with other factors). In certain embodiments, determining an optimal balance of the patient's potassium and insulin levels, and understanding the interactions between the patient's potassium level, glucose level, insulin, and one or more types of medications, may help inform which medications (including dosage and frequency) are most suitable for the patient.

[0298] In certain embodiments, a medication recommendation may include a recommendation for a patient to take a potassium binder rectally as an enema. In certain embodiments, a medication recommendation may include recommending to a patient to take an oral potassium binder such as Valtassa.

[0299] In certain embodiments, the medication recommendation may include recommendations to the patient to discontinue the use of glucose-lowering medications (e.g., sulfonylureas) or to titrate the glucose medication to a lower dose. In certain embodiments, the medication recommendation may be based on a decrease in renal function that results in lower blood glucose levels. To prevent dangerous hypoglycemic events, the medication recommendation can instruct the patient to discontinue and / or titrate the glucose-lowering dosing in response to a decrease in renal function and a decrease in blood glucose levels.

[0300] In certain embodiments, the medication recommendation may include recommendations for the patient to administer a particular type and / or dosage of diuretic. As previously mentioned, medications such as diuretics can be prescribed to patients for the purpose of treating excessive fluid accumulation caused by congestive heart failure (CHF), liver failure, and / or nephrotic syndrome. The type of diuretic prescribed can include loop diuretics, thiazide and thiazide-like diuretics, and / or potassium-sparing diuretics. Each of the identified diuretic types can be prescribed to the patient for different purposes. Thus, in certain embodiments, the decision support engine 114 can determine the optimal diuretic for prescription based on the patient's health and the condition of the patient being treated. For example, a patient with CHF (without kidney disease) may typically be prescribed a thiazide-like diuretic. In particular, thiazide-like diuretics can help to clear excessive fluid caused by CHF; however, in some cases, such diuretics can dehydrate the patient. Dehydration of a patient with impaired kidneys can further damage the patient's kidneys. In particular, dehydration can clog the kidneys with muscle protein (myoglobin). Thus, prescribing a thiazide-like diuretic to a patient who is experiencing only CHF and not kidney disease may not cause significant harm to the patient's kidneys. However, if the patient is experiencing kidney disease, such diuretics may not be optimal for prescription. Thus, another diuretic type may be considered.

[0301] In certain embodiments, the medication recommendation may include a recommendation to the patient to avoid medications that may expose the patient to the risk of high or low potassium levels.

[0302] Furthermore, in certain embodiments, the decision support engine 114 may determine the optimal diuretic for prescription by also considering the possible side effects that the prescribed diuretic may have on other organs of the patient. By considering the effects that different medications have on other organs, the decision support engine 114 may assist the patient in managing the condition of other organs within the patient's body. In certain embodiments, the CPM202 may assist in making this determination, at least with respect to the patient's kidneys. For example, the CPM202 may be used to monitor the effects of medications prescribed to the patient for the purpose of treating CHF in order to better determine its side effects on the patient's renal function. If it is observed that the potassium level has decreased 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, it may be concluded that such dehydration may have an adverse effect on the patient's kidneys. Accordingly, a new diuretic may be considered for prescription.

[0303] In certain embodiments, the alarm / alert recommendations can include recommendations regarding the addition of a new type of alarm / alert, the removal of an existing alarm / alert, an increase or decrease in the frequency of an existing alarm / alert, and / or a change in an existing threshold level for an existing alarm / alert configured for a device used by a patient. As noted, in certain embodiments, the type of alarm / alert 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 level configured for each of the alarms (e.g., for triggering) are based on the patient's current health, the patient's kidney condition, the current treatment 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 kidney disease prediction generated at block 408.

[0304] In certain embodiments, when the decision support is based at least on GMI and / or clinical A1C measurements, one or more recommendations generated can include a risk of developing kidney disease; a risk of the presence of kidney disease; a risk of health-harmful events such as hypoglycemia and hyperglycemia; and / or a recommendation to seek additional diagnostic tests for kidney disease. Such recommendations may, in certain embodiments, be based on a difference, change, and / or other discrepancy between the GMI measurement and the clinical A1C measurement.

[0305] In certain embodiments, when the decision support is based at least on glucose clearance, glucose-insulin clearance, and / or insulin clearance, one or more recommendations generated can include a risk of kidney disease; a risk of health-harmful events such as hypoglycemia and hyperglycemia; a risk of insulin-based harmful health events such as hypoglycemia and hyperglycemia; and / or a recommendation to seek additional diagnostic tests for kidney disease.

[0306] In one embodiment, when the decision-making support is based at least on potassium data, one or more recommendations generated may include the risk of kidney disease; the stage of kidney disease; the risk of death due to kidney disease; and / or the risk of adverse health events such as hypokalemia, hyperkalemia, cardiac events, etc.

[0307] In certain embodiments, when the decision-making support is based at least on potassium data and glucose data, one or more recommendations generated may include the risk of kidney disease; the stage of kidney disease; the risk of death due to kidney disease; and / or the risk of adverse health events such as hypokalemia, hyperkalemia, hypoglycemia, hyperglycemia, cardiac events, etc.

[0308] In certain embodiments, the treatment recommendations may include recommendations to perform a renal function challenge to further inform renal health assessment and renal disease prediction. The renal function load test may include administering a significant amount of an analyte or analyte precursor (e.g., fructose that can be metabolized to lactate and glucose), such as those described herein, and monitoring the clearance / metabolism of such analyte. For example, a known or estimated amount of an analyte can be administered to a patient, preferably orally, and then the analyte levels are monitored to determine the clearance rate of the analyte and / or the peak analyte level (which is higher in a diseased kidney compared to a healthy kidney). Generally, the clearance rate can be determined by calculating the slope between the initial analyte value and the baseline value. In certain embodiments, the baseline value can be determined using the patient's historical data. In certain embodiments, the baseline value can be determined from sensor data. The baseline represents the normal analyte level of the patient during a period in which no significant variation in analyte levels is expected, and each user may have a different baseline. The baseline can also be determined, in part, based on other measurements related to other analytes (e.g., lactate indicates exercise, glucose indicates food consumption, etc.) and / or non-analyte data (e.g., HR indicating exercise, time indicating circadian rhythm). In certain embodiments, the treatment recommendations may also include recommendations for the amount of analyte to administer for the renal function challenge. During or after the performance of the renal function challenge, the decision support engine 114 can generate alerts regarding the determined analyte clearance rate, changes in the analyte clearance rate, recommendations for additional renal disease tests, the risk of renal disease, the risk of progression or regression of renal disease, and / or the risk of renal disease stage. In further embodiments, the treatment recommendations may further include recommendations to repeat the renal function challenge at different times to determine changes in the kidney's ability to remove the analyte and / or to verify the analyte clearance rate. A decrease in the analyte clearance rate may indicate a decrease in renal function.

[0309] 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 kidney disease. The algorithms and / or machine learning models may take into account one or more of the inputs 128 and / or metrics 130 (e.g., including analyte levels and / or analyte trends) described with respect to FIG. 3 for the patient to determine optimal recommendations for the prevention and / or management of the patient's kidney disease. In certain embodiments, the model may be able to see different patterns of analyte measurements collected for the patient to guide the patient in the management of their disease. Again, the recommendations generated by the model, and thus the decision support engine 114, may be based on analyte levels, analyte thresholds, analyte rates of change, analyte variances, analyte clearance rates, and / or other analyte data.

[0310] After generating one or more recommendations, at block 412, method 400 continues by sending an indication (e.g., an alert, alarm, or other type of notification) to the user regarding kidney disease-related predictions (e.g., the presence of abnormal kidney function, a prediction regarding the risk of kidney disease; and / or the presence and / or stage of kidney disease) and / or the generated recommendations (e.g., alerts and / or alarms regarding the risk, presence, and / or severity of kidney disease, recommendations regarding treatment, etc.). In certain embodiments, the indication is sent to the patient via the application 106 and the indication is displayed to the user on a display device 107 such as a smartphone or other computing device. In certain embodiments, the indication is sent to the healthcare provider in addition to, or instead of, the patient.

[0311] In certain embodiments, any one or more components or devices of the decision support system 100 can include a "share / follow" function to alert, warn, provide recommendations to, and share history and / or predictive data with a patient's healthcare professional, clinician, and / or other caregiver. For example, such a "share / follow" function can be included in one or more continuous analyte sensors 202 of the continuous analyte monitoring system 104 and / or application 106 that are executed on the display device 107. In certain embodiments, such decision support alerts, warnings, and / or recommendations can be tailored to a patient's healthcare professional, clinician, and / or caregiver rather than the patient. In certain embodiments, such support alerts, warnings, and / or recommendations may be automatically provided to a patient's healthcare professional, clinician, and / or other caregiver. In certain embodiments, a patient can request the decision support system 100 to provide such support alerts, warnings, and / or recommendations to a patient's healthcare professional, clinician, and / or other caregiver, for example, via an interface of the display device 107 associated with the patient or patient interaction with the continuous analyte sensor 202. Support alerts, warnings, and / or recommendations can generally be provided to a patient's healthcare professional, clinician, and / or other caregiver via wired / wireless communication and / or other means of communicating data.

[0312] In certain embodiments, the machine learning model deployed by the decision support engine 114 includes one or more models trained by a training server system 140, as illustrated in FIG. 1. FIG. 5 further details techniques for training a machine learning model deployed by the decision support engine 114 to generate predictions associated with kidney disease, according to certain embodiments of the present disclosure.

[0313] In certain embodiments, method 500 is used to train a model to generate predictions related to kidney disease as an output. Preventive measures related to kidney disease include: (1) predictions regarding the presence of abnormal kidney function; (2) predictions regarding the risk of kidney disease; (3) predictions regarding the presence and / or stage of kidney disease. In certain embodiments, the predictions related to kidney disease may further include predictions regarding the risk of health adverse events for a patient (e.g., the user shown in FIG. 1), and / or predictions regarding the optimal treatment for the patient. In certain embodiments, the output generated by the model includes a determination of the variation between the modeled analyte data and the expected analyte data. In certain embodiments, the output generated by the model can correct and / or validate the measured analyte data.

[0314] Method 500 begins at block 502 by retrieving data from a medical record database, such as medical record database 112 shown in FIG. 1, by a training server system, such as training server system 140 shown in FIG. 1. As referred to herein, medical record database 112 can provide up-to-date information and medical history information for users of a continuous analyte monitoring system and a connected mobile health application, such as users of continuous analyte monitoring system 104 and application 106 illustrated in FIG. 1, as well as a repository of data for one or more patients who are not, or were not previously, users of continuous analyte monitoring system 104 and / or application 106. In certain embodiments, medical record database 112 can include one or more data sets of medical history patients who do not have kidney disease or have various stages of kidney disease (e.g., CKD).

[0315] The retrieval of data from the medical record database 112 by the training server system 140 may include, at block 502, the retrieval of all or any subset of the information maintained by the medical record database 112. For example, if the medical record database 112 stores information about 100,000 patients (e.g., non-users and users of the continuous analyte monitoring system 104 and the 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.

[0316] As an illustrative example, integration with an in-house or cloud-based medical record database through, for example, Fast Healthcare Interoperability Resources (FHIR), a web application programming interface (API), Health Level 7 (HL7), and / or other computer interface languages may enable the aggregation of medical history records for baseline assessment in addition to the aggregation of anonymized patient data from a cloud-based repository.

[0317] As an illustrative example, at block 502, the training server system 140 may retrieve information about 100,000 patients with various stages of kidney disease stored in the medical record database 112 to train a model that predicts the risk, presence, and / or severity of kidney 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 record database 112. Each user profile 118 may include information such as that discussed with respect to FIG. 3.

[0318] Next, the training server system 140 uses the information in each of the records to train an artificial intelligence or ML model (referred to herein as an "ML model" for simplicity). Examples of the types of information included in a patient's user profile were provided above. The information in each of these records can be characterized (e.g., manually or by the training server system 140) to yield features that can be used as input features for training the ML model. For example, a patient record can include or be used to generate features regarding the patient's age, the patient's gender, the patient's occupation, the patient's analyte levels over time, the rate of change and / or trend of the patient's analyte levels over time, physiological parameters associated with different stages of the patient's kidney disease over time, and / or any information provided by the input 128 and / or the metric 130. The features used to train the machine learning model can vary in different embodiments.

[0319] In certain embodiments, each medical history patient record retrieved from the medical history database 112 is further associated with a label indicating whether the patient was healthy or experienced any variation of kidney disease, the kidney disease diagnosis and / or stage of chronic kidney disease (CKD) previously determined for the patient, the kidney disease risk assessment, treatment, and / or similar metrics. What a record is labeled with depends on what the model is being trained to predict.

[0320] In block 504, method 500 continues by the training server system 140 training one or more machine learning models based on features and labels associated with the medical history patient records. In some embodiments, the training server does this by providing the features as inputs to the model. This model can be a new model initialized with random weights and parameters, or can 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 can indicate a diagnosis and / or stage of the patient's kidney disease, a risk assessment associated with the patient developing kidney disease, and an assessment regarding the improvement or worsening of the patient's existing kidney disease. In certain embodiments, the output can indicate the level of risk that the patient may develop kidney disease in the future.

[0321] In certain embodiments, the training server system 140 compares this generated output with the actual labels associated with the corresponding medical history patient records in order to calculate a loss based on the difference between the actual and the generated results. This loss is then used to improve one or more internal weights and parameters of the model (e.g., via backpropagation) such that the model learns to more accurately predict the presence and / or severity of kidney disease (or its recommended treatment).

[0322] One of various machine learning algorithms can be used to train the model described above. For example, one of a supervised learning algorithm, a neural network algorithm, a deep neural network algorithm, a deep learning algorithm, etc. can be used.

[0323] In block 506, the training server system 140 deploys the trained model to make predictions associated with kidney disease at runtime. In some embodiments, this includes transmitting some indication (e.g., a weight vector) of the trained model 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 evaluate, in real time, the presence and / or severity of a user's kidney disease using the application 106, provide treatment recommendations, and / or make other types of predictions as discussed above. In certain embodiments, the training server system 140 may continue to train the model in an “online” fashion by using input features and labels associated with new patient records.

[0324] Furthermore, a similar method for the training shown in FIG. 5 using historical patient records can also be used to train the model using patient-specific records to create a more personalized model for making predictions associated with kidney disease. For example, a model trained using historical patient records deployed for a particular user can be further retrained after deployment. For example, the model can be retrained after it has been deployed for a particular patient to create a more personalized model for that patient. The more personalized model can make more accurate kidney disease-related predictions for the patient based on the patient's own data (not just historical patient record data) including the patient's own analyte (e.g., potassium) thresholds.

[0325] FIG. 6 is a block diagram showing a computing device 600 configured to diagnose, stage, treat, and evaluate the risk of kidney disease according to certain embodiments disclosed herein. Although depicted as a single physical device, in embodiments, computing device 600 may be implemented using virtual devices and / or across several devices such as a cloud environment. As illustrated, computing device 600 includes a processor 605, a memory 610, a storage 615, a network interface 625, and one or more I / O interfaces 620. In the illustrated embodiment, processor 605 fetches and executes programming instructions stored in memory 610 and stores and fetches application data present in storage 615. Processor 605 generally represents a single CPU and / or GPU, multiple CPUs and / or GPUs, a single CPU and / or GPU having multiple processing cores, and the like. Memory 610 is generally included to represent random access memory (RAM). Storage 615 can be any combination of a disk drive, flash-based storage device, etc., and can include fixed and / or removable storage devices such as a fixed disk drive, removable memory card, cache, optical storage, network attached storage (NAS), or storage area network (SAN).

[0326] In some embodiments, an I / O device 635 (such as a keyboard, monitor, etc.) can be connected via an I / O interface 620. Further, via a network interface 625, the computing device 600 can be communicatively coupled to one or more other devices and components such as a user database 110 and / or a medical record database 112. In certain embodiments, the computing device 600 is communicatively coupled to other devices via a network that can include the Internet, a local network, etc. The network can include a wired connection, a wireless connection, or a combination of a wired connection and a wireless connection. As shown, the processor 605, the memory 610, the storage 615, the network interface 625, and the I / O interface 620 are communicatively coupled by one or more interconnects 630. In certain embodiments, the computing device 600 represents a display device 107 associated with a user. In certain embodiments, as discussed above, the display device 107 can include a user's laptop, computer, smartphone, etc. In another embodiment, the computing device 600 is a server executed in a cloud environment.

[0327] In the embodiment shown, the storage 615 includes a user profile 118. The memory 610 includes a decision support engine 116 that itself includes a DAM 116. The decision support engine 114 is executed by the computing device 600 to perform the operations of method 400 of FIG. 4 and the operations of method 500 of FIG. 5 to provide decision support in the form of risk assessment and treatment of kidney disease (e.g., CKD).

[0328] As described above, the continuous analyte monitoring system 104 described in connection with FIG. 1 can be a multi-analyte sensor system that includes a multi-analyte sensor. FIGS. 7-11 depict exemplary multi-analyte sensors used to measure multiple analytes.

[0329] As used herein, the terms "analyte measurement device", "analyte monitoring device", "analyte sensing device", and / or "multianalyte sensor device" are broad terms and are given their ordinary and customary meaning to one of ordinary skill in the art (and are not limited to a special or customized meaning), and refer to, but are not limited to, devices and / or systems that are responsible for the detection of a particular analyte or combination of analytes or the conversion of signals associated therewith. For example, these terms may refer to, but are not limited to, devices responsible for the detection of a particular analyte or combination of analytes. In one embodiment, the device includes a sensor coupled to a circuit arranged within a housing and configured to process signals associated with analyte concentration into information. In one embodiment, such a device and / or system can use a biological recognition element combined with a conversion (detection) element to provide specific quantitative, semi-quantitative, qualitative, and / or semi-qualitative analytical information.

[0330] As used herein, the terms "biosensor" and / or "sensor" are broad terms and are given their ordinary and customary meaning to one of ordinary skill in the art (and are not limited to a special or customized meaning), and refer to, but are not limited to, part of an analyte measurement device, analyte monitoring device, analyte sensing device, and / or multianalyte sensing device that is responsible for the detection of a particular analyte or combination of analytes or the conversion of signals associated therewith. In one embodiment, a biosensor or sensor generally comprises a body and a working electrode, a reference electrode, and / or a counter electrode coupled to the body and configured to form a surface that provides a signal during an electrochemical reaction. One or more membranes can be affixed to the body and cover the electrochemically reactive surface. In one embodiment, such a biosensor and / or sensor can use a biological recognition element combined with a conversion (detection) element to provide specific quantitative, semi-quantitative, qualitative, semi-qualitative analytical signals.

[0331] As used herein, the terms "sensing moiety", "sensing membrane", and "sensing mechanism" are broad terms and are given their ordinary and customary meaning to one of ordinary skill in the art (and are not limited to a special or customized meaning), and refer to, but are not limited to, a biosensor and / or a part of a sensor that is responsible for the detection of a specific analyte or combination of analytes or the conversion of a signal associated therewith. In one embodiment, the sensing moiety, sensing membrane, and / or sensing mechanism generally comprises an electrode configured to provide a signal during an electrochemical reaction with one or more membranes covering an electrochemically reactive surface. In one embodiment, such sensing moiety, sensing membrane, and / or sensing mechanism can use a biological recognition element combined with a conversion (detection) element to provide specific quantitative, semi-quantitative, qualitative, semi-qualitative analysis information.

[0332] As used interchangeably herein, the terms "biomimetic membrane" and "biomimetic layer" are broad terms and are given their ordinary and customary meaning to one of ordinary skill in the art (and are not limited to a special or customized meaning), and refer to, but are not limited to, a permeable membrane (which can include multiple domains) or layer that functions as a biodefense interface between a host tissue and an implantable device. The terms "biomimetic" and "biodefense" are used interchangeably herein.

[0333] As used herein, the term "cofactor" is a broad term and is given its ordinary and customary meaning to one of ordinary skill in the art (and is not limited to a special or customized meaning), and non-limitingly refers to one or more substances whose presence contributes to or is required for the analyte-related activity of an enzyme. Analyte-related activity can include, but is not limited to, any one or a combination of binding, electron transfer, and chemical conversion. Cofactors include coenzymes, non-protein chemical compounds, metal ions, and / or metal-organic complexes. Coenzymes include co-substrate families and prosthetic groups.

[0334] As used herein, the term "continuous" is a broad term and is given its ordinary and customary meaning to one of ordinary skill in the art (and is not limited to a special or customized meaning), and refers to a portion, domain, coating, or layer that is uninterrupted or seamless, but is not limited thereto.

[0335] As used herein, the phrases "continuous analyte sensing" and "continuous multi-analyte sensing" are broad phrases and are given their ordinary and customary meaning to one of ordinary skill in the art (and are not limited to a special or customized meaning), and refer to a period during which monitoring of analyte concentration is performed continuously, intermittently, and / or periodically (but regularly), for example, every about 1 second or less to about 1 week or more, but is not limited thereto. In further embodiments, monitoring of analyte concentration is performed every about 2, 3, 5, 7, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, or 60 seconds to about 1.25, 1.50, 1.75, 2.00, 2.25, 2.50, 2.75, 3.00, 3.25, 3.50, 3.75, 4.00, 4.25, 4.50, 4.75, 5.00, 5.25, 5.50, 5.75, 6.00, 6.25, 6.50, 6.75, 7.00, 7.25, 7.50, 7.75, 8.00, 8.25, 8.50, 8.75, 9.00, 9.25, 9.50, or 9.75 minutes. In further embodiments, monitoring of analyte concentration is performed every about 10, 20, 30, 40, or 50 minutes to about 1, 2, 3, 4, 5, 6, 7, or 8 hours. In further embodiments, monitoring of analyte concentration is performed every about 8 hours to about 12, 16, 20, or 24 hours. In further embodiments, monitoring of analyte concentration is performed every about 1 day to about 1.5, 2, 3, 4, 5, 6, or 7 days. In further embodiments, monitoring of analyte concentration is performed every about 1 week to about 1.5, 2, 3 weeks or more.

[0336] As used herein, the term "coaxial" should be broadly interpreted to include a sensor architecture having elements aligned along a common axis around a core that can be configured to have a circular, elliptical, triangular, polygonal, or other cross-section, such elements can include electrodes, insulating layers, or other elements that can be circumferentially located around a core layer such as a core electrode or a core polymer wire.

[0337] As used herein, the term "coupled" is a broad term and is given its ordinary and customary meaning to one of ordinary skill in the art (and is not limited to a special or customized meaning), and refers to two or more system elements or components configured to be attached to at least one of electrical, mechanical, thermal, operable, chemical, or otherwise, but is not limited thereto. For example, if an element is covalently, communicatively, electrostatically, thermally, mechanically, magnetically, or ionically associated with, or physically captured, adsorbed, or absorbed by another element, the element is "coupled". Similarly, as used herein, the phrases "operatively connected", "operatively linked", and "operatively coupled" can refer to one or more components coupled to another component in a manner that facilitates the transmission of at least one signal between the components. In some embodiments, the components are part of the same structure and / or integrated with each other, such as when they are covalently, electrostatically, mechanically, thermally, magnetically, ionically associated with, or physically captured or absorbed (i.e., "directly coupled" as if there were no intervening elements). In other embodiments, the components are connected via remote means. For example, one or more electrodes can be used to detect an analyte in a sample and convert the information into a signal, which can then be transmitted to an electronic circuit. In this example, the electrode is "operatively linked" to the electronic circuit. As used herein, the phrase "removably coupled" can refer to two or more system elements or components configured to be attached and removed electrically, mechanically, thermally, operably, chemically, or otherwise, without damaging any of the coupled elements or components, or so configured.As used herein, the phrase "permanently coupled" refers to two or more system elements or components that are configured or attached to be electrically, mechanically, thermally, operably, chemically, or otherwise attached, or are attached, but cannot be separated without damaging at least one of the coupled elements or components, and are covalently, electrostatically, ionically associated, or physically trapped or absorbed without being physically separated.

[0338] As used herein, the term "discontinuous" is a broad term and is given its ordinary and customary meaning to one of ordinary skill in the art (and is not limited to a special or customized meaning), and refers to, but is not limited to, a severed, interrupted, or separated portion, layer, coating, or domain.

[0339] As used herein, the term "distal" is a broad term and is given its ordinary and customary meaning to one of ordinary skill in the art (and is not limited to a special or customized meaning), and refers to, but is not limited to, a region that is relatively far spaced from a reference point such as an origin or an attachment point.

[0340] As used herein, the term "domain" is a broad term and is given its ordinary and customary meaning to one of ordinary skill in the art (and is not limited to a special or customized meaning), and refers to, but is not limited to, a region of a membrane system that can be a layer, a homogeneous or inhomogeneous gradient (e.g., an anisotropic region of a membrane), or a portion of a membrane that is capable of sensing one, two, or more analytes. The domains discussed herein can be formed as a single layer, as two or more layers, as a pair of bilayers, or as combinations thereof.

[0341] As used herein, the term "electrochemically reactive surface" is a broad term and is given its ordinary and customary meaning to one of ordinary skill in the art (and is not limited to a special or customized meaning), and refers to, but is not limited to, the surface of an electrode at which an electrochemical reaction occurs. In one example, this reaction is Faradaic and results in charge transfer between the surface and its environment. In one example, hydrogen peroxide produced by an enzyme-catalyzed reaction of an analyte oxidized on the surface results in a measurable electron current. For example, in the detection of glucose, glucose oxidase produces hydrogen peroxide (H2O2) as a byproduct. H2O2 reacts with the surface of the working electrode to produce two protons (2H + ), two electrons (2e - ), and one oxygen molecule (O2), which produces the detected electron current. At the counter electrode, a reducible species, such as O2, is reduced at the electrode surface to balance the current generated by the working electrode.

[0342] As used herein, the term "electrolysis" is a broad term and is given its ordinary and customary meaning to one of ordinary skill in the art (and is not limited to a special or customized meaning), and non-limitingly refers to, but is not limited to, the electrochemical oxidation or reduction (collectively "redox") of a compound, either directly or indirectly, by one or more enzymes, cofactors, or mediators.

[0343] As used herein, the terms "implanted", "being implanted", "embedded", or "implantable" are broad terms and are given their ordinary and customary meaning to one of ordinary skill in the art (and are not limited to a special or customized meaning), and refer to an object (e.g., a sensor) inserted or configured to be inserted subcutaneously (i.e., within the adipose layer between the skin and muscle), intradermally (i.e., penetrating the stratum corneum and located within the epidermal or dermal layer of the skin), or transdermally (i.e., penetrating, entering, or passing through intact skin), which can result in, but is not limited to, a sensor having an in vivo portion and an ex vivo portion. The term "implanted" also encompasses an object configured to be inserted subcutaneously, intradermally, or transdermally, whether or not it has itself been inserted.

[0344] As used herein, the terms "interferent" and "interfering species" are broad terms and are given their ordinary and customary meaning to one of ordinary skill in the art (and are not limited to a special or customized meaning), and refer to an effect and / or species that interferes with the measurement of an analyte of interest in a sensor and produces a signal that does not accurately represent the analyte measurement, but is not limited thereto. In one example of an electrochemical sensor, an interfering species is a compound that produces a signal that is not analyte-specific due to a reaction on the electrochemically active surface.

[0345] As used herein, the term "in vivo" is a broad term and is given its ordinary and customary meaning to one of ordinary skill in the art (and are not limited to a special or customized meaning), and includes, but is not limited to, a portion of a device (e.g., a sensor) adapted for insertion into and / or presence within the body of a host.

[0346] As used herein, the term "ex vivo" is a broad term and is given its ordinary and customary meaning to one of ordinary skill in the art (and is not limited to a special or customized meaning), and includes, but is not limited to, a portion of a device (e.g., a sensor) adapted to remain and / or exist outside of the host organism.

[0347] As used herein, the terms and phrases "mediator" and "redox mediator" are broad terms and phrases and are given their ordinary and customary meaning to one of ordinary skill in the art (and are not limited to a special or customized meaning), and refer to any chemical compound or aggregate of compounds capable of direct or indirect electron transfer between an analyte, analyte precursor, analyte surrogate, analyte reducing enzyme or analyte oxidizing enzyme, or cofactor, and an electrode surface held at a potential, but are not limited thereto. In one example, the mediator receives electrons from or transfers electrons to one or more enzymes or cofactors and / or exchanges electrons with the sensor system electrode. In one example, the mediator is a transition metal coordination organic molecule capable of reversible oxidation and reduction reactions. In other examples, the mediator can be an organic molecule or metal capable of reversible oxidation and reduction reactions.

[0348] As used herein, the term "membrane" is a broad term and is given its ordinary and customary meaning to one of ordinary skill in the art (and is not limited to a special or customized meaning), serving as a protection for the exposed electrode surface from the biological environment, a diffusion resistance (limitation) for an analyte, a matrix for a catalyst (e.g., one or more enzymes) to enable an enzymatic reaction, a limitation or blocking of interfering species, providing hydrophilicity at the electrochemically reactive surface of a sensor interface, serving as an interface between the host tissue and an implantable device, regulating the host tissue response via drug (or other substance) release, and combinations thereof, but not limited thereto. Structures configured to perform functions including, but not limited to, those described above are referred to. As used herein, the terms "membrane" and "matrix" are meant to be interchangeable.

[0349] As used herein, the phrase "membrane system" is a broad phrase and is given its ordinary and customary meaning to one of ordinary skill in the art (and is not limited to a special or customized meaning), and can be composed of two or more domains, layers, or layers within a domain, typically composed of a material with a thickness of several microns or more, permeable to oxygen, and optionally permeable to, for example, glucose or another analyte, but not limited thereto. In one embodiment, the membrane system includes an enzyme that enables an analyte reaction to occur, whereby the concentration of the analyte can be measured.

[0350] As used herein, the term "plane" should be broadly construed to describe a sensor architecture having a substrate that includes at least a first surface and an opposing second surface, and comprising a plurality of elements disposed on one or more surfaces or edges of the substrate. The plurality of elements can include a conductive layer, an insulating layer, or elements configured to operate as a circuit. The plurality of elements may or may not be electrically or otherwise coupled. In one embodiment, the plane includes one or more edges separating the opposing surfaces.

[0351] As used herein, the term "proximal" is a broad term and is given its ordinary and customary meaning to one of ordinary skill in the art (and is not limited to a special or customized meaning), and refers to, but is not limited to, the spatial relationship between various elements as compared to a particular reference point. For example, some embodiments of a device include a membrane system having a biointerface layer and an enzyme domain or enzyme layer. If the sensor is considered the reference point and the enzyme domain is located closer to the sensor than the biointerface layer, the enzyme domain is more proximal to the sensor than the biointerface layer.

[0352] As used herein, the phrases "sensing portion", "sensing membrane", and "sensing mechanism" are broad phrases and are given their ordinary and customary meaning to one of ordinary skill in the art (and are not limited to a special or customized meaning), and refer to, but are not limited to, a biosensor and / or a part of a sensor that is responsible for the detection of a particular analyte or combination of analytes or the conversion of a signal associated therewith. In one embodiment, the sensing portion, sensing membrane, and / or sensing mechanism generally comprises an electrode configured to provide a signal during an electrochemical reaction with one or more membranes covering an electrochemically reactive surface. In one embodiment, such a sensing portion, sensing membrane, and / or sensing mechanism can use a biological recognition element combined with a conversion (detection) element to provide specific quantitative, semi-quantitative, qualitative, semi-qualitative analytical information.

[0353] During the general operation of an analyte measurement device, biosensor, sensor, sensing region, sensing portion, or sensing mechanism, a biological sample, such as blood or interstitial fluid, or components thereof, after passing directly or through one or more membranes, comes into contact with an enzyme, such as glucose oxidase, DNA, RNA, or a protein or aptamer, such as one or more periplasmic binding proteins (PBPs) having one or more analyte binding regions or variants or fusion proteins thereof, and each region is capable of specifically or reversibly binding and / or reacting with at least one analyte. The interaction between the biological sample or its components and the analyte measurement device, biosensor, sensor, sensing region, sensing portion, or sensing mechanism results in a conversion of a signal that enables a qualitative, semi-qualitative, quantitative, or semi-quantitative determination of the analyte level in the biological sample, such as glucose, ketones, lactate, potassium, etc.

[0354] In one embodiment, the sensing region or sensing portion can comprise at least a portion of a conductive substrate or at least a portion of a conductive surface, such as a wire (coaxial) or conductive trace, or a substantially planar substrate including a substantially planar trace, and a membrane. In one embodiment, the sensing region or sensing portion can comprise a non-conductive body, a working electrode, a reference electrode, and a counter electrode (optional) that form an electrochemically reactive surface at one location on the body and an electrical connection at another location on the body, and a sensing membrane attached to the body and covering the electrochemically reactive surface. In some embodiments, the sensing membrane further comprises an enzyme domain, such as an enzyme domain, and an electrolyte phase, such as a free-flowing liquid phase containing an electrolyte-containing fluid further described below. These terms are broad enough to include the entire device or only the sensing portion thereof (or something in between).

[0355] In another embodiment, the sensing region can include one or more peripheral membrane-binding proteins (PBPs) that contain the variant or fusion protein, or an aptamer having one or more analyte-binding regions, and each region can specifically and reversibly bind to at least one analyte. Changes in the aptamer or mutations in the PBP can contribute to or change the long-term stability of the protein, including one or more binding constants, thermal stability, in order to bind the protein to a special encapsulation matrix, membrane or polymer, or to attach a detectable reporter group or "label" to indicate changes in the binding region, or to convert a signal corresponding to one or more analytes present in the biological fluid. Specific examples of changes in the binding region include, but are not limited to, changes in the hydrophobic / hydrophilic environmental change, three-dimensional conformational change, change in the orientation of amino / nucleic acid side chains in the binding region of the protein, and redox state of the binding region. Such changes to the binding region provide for the conversion of a detectable signal corresponding to one or more analytes present in the biological fluid.

[0356] In one embodiment, the sensing region determines the selectivity between one or more analytes such that only the analyte to be measured results in (converts) a detectable signal. This selection can be based on any chemical or physical recognition of the analyte by the sensing region if the chemical composition of the analyte does not change, or if the sensing region causes or catalyzes an analyte reaction that changes the chemical composition of the analyte.

[0357] As used herein, the term "sensitivity" is a broad term and is given its ordinary and customary meaning to those of ordinary skill in the art (and is not limited to a special or customized meaning), and refers to the amount of signal (e.g., in the form of current and / or voltage) produced by a given quantity (unit) of the measured analyte, but is not limited thereto. For example, in one embodiment, the sensor has a sensitivity (or slope) of about 1 to about 100 picoamperes of current per mg / dL of analyte.

[0358] The terms "signal medium" or "transmission medium" should be interpreted to include any form such as a modulated data signal, a carrier wave, etc. The term "modulated data signal" means a signal in which one or more of its characteristics are set or changed so as to encode information in the signal.

[0359] As used herein, the terms "convert" or "conversion" and their grammatical equivalents are broad terms and are given their ordinary and customary meaning to one of ordinary skill in the art (and are not limited to a special or customized meaning), and refer to, but are not limited to, optical, electrical, electro-chemical, acoustic / mechanical, or colorimetric techniques and methods. Electrochemical properties include current and / or voltage, inductance, capacitance, impedance, and potential. Optical properties include absorbance, fluorescence / phosphorescence, fluorescence / phosphorescence decay rate, wavelength shift, dual wave phase modulation, bioluminescence / chemiluminescence, reflectivity, light scattering, and refractive index. For example, the sensing region converts the recognition of an analyte into a semi-quantitative or quantitative signal.

[0360] As used herein, the term "transducing element" is a broad term when used herein and is given its ordinary and customary meaning to one of ordinary skill in the art (and is not limited to a special or customized meaning), and refers to, but is not limited to, an analyte recognition moiety that can directly or indirectly facilitate a detectable signal transduction corresponding to the presence and / or concentration of a recognized analyte. In one example, the transducing element is one or more enzymes, one or more aptamers, one or more ionophores, one or more capture antibodies, one or more proteins, one or more living cells, one or more oligonucleotides, and / or one or more DNA or RNA moieties. The transdermal continuous multi-analyte sensor can be used in vivo over various lengths of time. The continuous multi-analyte sensor system discussed herein can be a transdermal device in that a portion of the device can be inserted into the soft tissue below the host's skin while a portion of the device remains on the surface of the host's skin. In one aspect, to overcome short-term noise or problems associated with other sensor functions, one example uses a material that promotes the formation of a fluid pocket around the sensor, such as a porous biocompatible interfacial membrane or matrix that creates a space between the sensor and the surrounding tissue. In some examples, the sensor comprises a spacer adapted to provide a fluid pocket between the sensor and the host tissue. This spacer, such as a biocompatible material, matrix, structure, etc., as described in more detail elsewhere herein, is thought to...

Claims

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

2. The continuous analyte sensor comprises: A substrate; A working electrode arrayed on the substrate; and A reference electrode arrayed on the substrate, wherein the analyte measurement generated by the continuous analyte sensor corresponds to an electromotive force at least partially based on a potential difference generated between the working electrode and the reference electrode. The monitoring system according to claim 1.

3. The continuous analyte sensor includes a continuous potassium sensor, The analyte measurement includes a potassium measurement. The monitoring system according to claim 1.

4. A memory including executable instructions; One or more processors configured to communicate with the memory, receive potassium data associated with the potassium measurement from the sensor electronics module, process the potassium data to determine at least one potassium trend, and generate a kidney disease prediction based on the at least one potassium trend of the patient. The monitoring system according to claim 3.

5. The kidney disease prediction indicates at least one of: The risk of future kidney disease in the patient; The current presence of kidney disease in the patient; The severity of kidney disease in the patient; or The level of improvement or deterioration of the kidney disease in the patient. The monitoring system according to claim 4.

6. The severity of the kidney disease corresponds to the stage of chronic kidney disease. The monitoring system according to claim 5.

7. The method further comprises generating one or more recommendations for the treatment or prevention of kidney disease based at least in part on the kidney disease prediction. The monitoring system according to claim 4.

8. The one or more recommendations include at least one of: Recommendations for improving lifestyle habits; Medical recommendations; Intervention recommendations; or Recommendations for additional diagnostic tests. The monitoring system according to claim 7.

9. The one or more recommendations include a recommendation to perform a renal function stress test. The monitoring system according to claim 7.

10. The one or more recommendations include: Abnormal analyte levels; Abnormal analyte change rates; ​ ​ ​ An abnormal analyte clearance rate, or The monitoring system according to claim 7, comprising an alert or alarm indicating at least one of abnormal analyte dispersion.

11. The continuous analyte sensor further includes a continuous glucose sensor, The analyte measurement values further include glucose measurement values, The processor is further configured to receive glucose data associated with the glucose measurement values from the sensor electronics module, The renal disease prediction is further based on the glucose data, and the monitoring system according to claim 4.

12. Further includes one or more non-analyte sensors, and the processor Is configured to receive non-analyte sensor data generated for the patient using the one or more non-analyte sensors, and the renal disease prediction is generated based on the non-analyte sensor data. The monitoring system according to claim 4.

13. 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, impedance, or a respiratory sensor. The monitoring system according to claim 12.

14. The renal disease prediction is generated using a model trained based on population data including records of historical patients with various stages of renal disease. The monitoring system according to claim 4.

15. 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, illness information, or renal disease stage information regarding the patient, The renal disease prediction is generated based on at least one of the food consumption information, the activity level information, the medical information, the health and illness information, the disease information, or the renal disease stage information regarding the patient. The monitoring system according to claim 4.

Images