Hydration management system
Through computer-implemented multi-data source monitoring and mathematical model evaluation, the problem of hydration status monitoring in dialysis patients is solved, and the continuous monitoring and prediction of hydration status is achieved, the risk of complications is reduced, and the effect of dialysis treatment and quality of life is improved.
Patent Information
- Application Number
- CN202380030356.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-03-25
- Filing Date
- 2023-03-23
- Publication Date
- 2024-12-03
AI Technical Summary
The prior art is difficult to effectively monitor and manage the hydration status of dialysis patients, resulting in complications caused by excessive or insufficient hydration, such as arterial hypertension, hypotension events, pulmonary edema and heart failure, and lacks reliable continuous hydration monitoring tools.
Through a computer-implemented approach, multiple data sources (including patient monitors, interdialysis measurement devices and blood therapy equipment) collect physiological parameters and medical treatment-related parameters, combine multiple parameter combinations in clinical databases, and use mathematical models to evaluate patients using mathematical models. hydration state and output prediction and treatment recommendations.
Continuous monitoring and prediction of hydration status of dialysis patients is achieved, the risk of complications is reduced, the effect of dialysis treatment and the quality of life of patients is improved, personalized water and salt intake advice is provided, and patients' self-care ability is enhanced.
Smart Images

Figure CN119072753A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a computer-implemented method for assessing a patient's physiological state, in particular his hydration state, according to claim 1, a system according to claim 12 and an apparatus according to claim 14 or claim 15 or according to the preamble or general term of each of these claims. Background Art
[0002] Restoring patients to a normal hydration state (HD), also known as the absence of overhydration and underhydration, remains one of the major clinical challenges in dialysis therapy. In fact, only a minority of patients reach the target "dry weight" after removing the prescribed volume of body fluids during dialysis. Abnormal hydration status is associated with a number of complications and comorbidities, such as arterial hypertension, arterial stiffness, intradialytic hypotension (IDH) events, pulmonary and peripheral edema, heart failure and increased mortality.
[0003] Current challenges in this area include active patient participation in fluid management (i.e., including compliance with fluid and salt intake restrictions and physical exercise activation), the lack of reliable instrumentation for continuous hydration monitoring, and / or the complex pathophysiology of complications due to hydration imbalance.
[0004] The main hallmark of advanced chronic kidney disease (CKD) is the inability of the kidneys to maintain fluid homeostasis by eliminating excess water. Fluid overload ultimately leads to severe complications, including peripheral edema, pulmonary congestion, pleural effusions, hypertension, as well as heart failure and increased mortality. Therefore, the main goal of dialysis therapy is to reestablish fluid homeostasis by replacing insufficient renal function. Summary of the invention
[0005] Therefore, one object of the invention may be to propose a computer-implemented method for assessing a patient's physiological state, in particular the hydration state or its complications like hypotension etc. Furthermore, a system comprising an interdialytic and / or intradialytic monitor may be proposed.
[0006] The objects of the invention can be achieved by a computer-implemented method for assessing the physiological state, in particular the hydration state, of a patient having the features of claim 1 , by a system having the features of claim 12 and by a device having the features of claim 14 or claim 15 .
[0007] According to the invention, a computer-implemented method for assessing or for use in assessing the physiological state, in particular the hydration state, of a patient is proposed.
[0008] The method comprises the step of reading in at least one value of at least one first parameter of the patient from a first data source. The at least one first parameter is a physiological parameter of the patient.
[0009] The method comprises as a further step receiving at least one value of at least one second parameter from a second data source. The first data source and the second data source are optionally at least partially different from each other. The second parameter is also related to the patient, or to a medical treatment of the patient, or to a treatment device that has, for example, treated or diagnosed the patient in the past.
[0010] Whenever the term "value" is mentioned in this document, this can be understood as a single value, such as a discrete value, a range of values and / or a continuous development of values. A value is a qualitative or quantitative occurrence of a parameter at a given time, also called a parameter value. For example, although a parameter is something that can be measured, such as a heartbeat, and different results may be obtained when measuring it, such as numbers, values, once measured, the value is a given data, such as 64 beats per minute.
[0011] Values may be collected about parameters describing the patient's entire body, at least a part thereof, or about a patient's body layer (skin, tissue or the like), the patient's function or the patient's state. Values may also be collected by checking or monitoring machine parameters or behavior.
[0012] The method according to the present invention also comprises providing a clinical database of multiple combinations of values comprising multiple parameters. These parameters may comprise the first parameter and the second parameter. In addition, multiple values are obtained from the specific patient and / or from multiple patients, preferably with reference to a patient group or its treatment.
[0013] The method further comprises providing a mathematical model configured or programmed for evaluating at least one read-in value. The evaluation is based on at least one of the plurality of combinations of the parameter values included in the database.
[0014] The method further comprises evaluating at least one read-in value of at least one first physiological parameter. For the evaluating step, the mathematical model is used. The mathematical model is based on at least one of the combinations of parameter values while retrieving at least one of the plurality of combinations of parameter values. In this case, the one or more combinations retrieved from the clinical database may include the received value of at least a second parameter.
[0015] The method further comprises making and outputting a prediction of a future development of at least one physiological state of the patient and / or outputting at least one of the parameter values, in particular the first parameter and / or the second parameter. The prediction may be based on the result of the evaluating step.
[0016] As an alternative or in addition to outputting the results of the prediction and / or evaluation step, recommendations regarding future behavior or treatment of the patient may also be communicated to the patient, medical staff, caregiver and / or physician. This may be included as a further step by the method.
[0017] According to the present invention, a system is proposed.
[0018] The system comprises a first data source and at least one second data source, wherein the first data source and the second data source are optionally at least partially different from each other.
[0019] Furthermore, the system comprises a reading device for reading in at least one value of at least one first parameter of the patient from the first data source. The first parameter is a physiological parameter of the patient.
[0020] The system further comprises receiving means for receiving at least one value of at least one second parameter from the second data source. The second parameter is related to the patient or to a medical treatment of the patient.
[0021] The system also includes a clinical database. In the database, multiple combinations of values of multiple parameters are stored and / or provided. These parameters may include the first parameter and the second parameter, wherein the multiple values are obtained from the patient and / or from multiple patients of a reference patient group.
[0022] The system further comprises a mathematical model for evaluating the at least one read-in value based on at least one of the plurality of combinations of parameter values included in the clinical database.
[0023] Furthermore, the system comprises an evaluation device. The evaluation device is used to evaluate at least one read-in value of the at least one first physiological parameter using the mathematical model. The mathematical model is based on at least one of the combinations of parameter values, while retrieving at least one of the plurality of combinations of parameter values. The one or more combinations retrieved from the clinical database may include the received value of at least the second parameter.
[0024] Based on the result of the evaluation step, the system further comprises an output device for outputting a prediction of the future development of at least one physiological state of the patient or at least one of the patient's parameters, or for suggesting a future behavior or treatment for the patient. The output may be, for example, for the patient, a caregiver or a doctor.
[0025] According to the invention, a device, in particular a handheld device, is proposed. Such a device may be, for example, a tablet computer, a smartphone or the like. The device comprises a control device programmed to execute or trigger the execution of the method according to the invention.
[0026] According to the invention, a device, in particular a handheld device, is proposed, comprising two interfaces. One of these interfaces is programmed or configured to be able to read in at least one value of at least one first parameter of the patient from a first data source, and the other is programmed or configured to be able to receive at least one value of at least one second parameter from a second data source. In this context, the first data source and the second data source are preferably at least partially different from each other.
[0027] Embodiments of the device according to the invention as described above may also comprise features or feature combinations of any device according to the invention as described herein.
[0028] Embodiments according to the present invention may include one or more of the features mentioned above and / or below in any combination, unless a person skilled in the art considers that a particular combination is technically impossible.
[0029] Advantageous developments of the invention are respectively subject matter of the dependent claims and the exemplary embodiments.
[0030] In all the statements above and below, the use of the expressions “may be” or “may have” etc. should be understood as synonyms with “preferably is” or “preferably has” etc., respectively, and are intended to illustrate embodiments according to the present invention.
[0031] Whenever the alternative of "and / or" is introduced herein, those skilled in the art will understand that the "or" contained therein is preferably "either or", and preferably not "and".
[0032] Whenever numerical terms are mentioned herein, those skilled in the art will recognize or understand them as indications of numerical lower limits. Thus, unless this results in an apparent contradiction to those skilled in the art, those skilled in the art will understand that, for example, "one" (or "one / an") includes "at least one". This understanding is also encompassed by the present invention, as numerical terms, such as "one" (or "one / an"), may alternatively mean "exactly one", provided that this is technically apparent to those skilled in the art. Both understandings are encompassed by the present invention and apply to all numerical terms used herein.
[0033] Whenever an embodiment is mentioned herein, it is meant an exemplary embodiment according to the present invention.
[0034] When it is disclosed herein that the subject matter according to the invention comprises one or more features in a certain embodiment, it is also disclosed herein that the subject matter according to the invention also expressly does not comprise this or these features in other embodiments according to the invention, for example in the sense of a disclaimer. Thus, for each embodiment mentioned herein, the opposite embodiment is also disclosed, for example expressed as a negation.
[0035] When method steps are mentioned in this document, the system according to the invention (which may include computing means, control means, closed-loop control means or the like for this purpose) is configured in several embodiments to perform one or more of these method steps in any combination, in particular when these are automatically executable steps, or in order to control corresponding devices accordingly, and the corresponding devices preferably correspond to the designation of the corresponding method steps by their names (for example, "determination" as a method step and "device for determining" for the device), and they can likewise be part of the device according to the invention or be connected to the device in a signal communication manner.
[0036] When reference is made herein to programming or configuration, then these terms may be interchangeable in some embodiments.
[0037] When referring to signal communication or communication connection between two components in this article, this can be understood to mean the connection existing during use. It can also be understood that there is provision for such signal communication (wired, wireless or otherwise implemented), such as by coupling two components, such as by pairing, etc.
[0038] Pairing is a process related to computer networks in order to establish an initial link between computer units for communication purposes. The best known example of this is the establishment of a Bluetooth connection, through which various devices (e.g. smartphones, headphones) are connected to each other. Pairing is sometimes also called bonding.
[0039] When method steps are mentioned in this document, it is provided in several embodiments according to the invention that these steps are performed before or after treatment of the patient, for example during a time when the patient is not connected to the device, for example via an extracorporeal blood circuit or the like, or during a time when the treatment has not yet started or has been completed.
[0040] The treatment of the patient may at least be completed or may not have even begun when, for example, the patient is not connected to the blood treatment device, for example because their vascular system is not connected in a fluid-connected manner to the extracorporeal blood circuit, or when no blood or fluid is returned to their vascular system, in particular when no blood or fluid is returned at a relevant point in time and / or in the further course of a relevant treatment period.
[0041] The system may cause all or substantially all method steps to be performed, for example via a control device or a closed-loop control device or the like. The method according to the invention may be substantially or completely performed, for example, by the control device or the closed-loop control device. It may be performed in part by the control device or the closed-loop control device. Such method steps may optionally be limited to machine-induced or automatic method steps.
[0042] The term "patient monitor" as used herein may be any device, in particular a handheld or body-carrying item (e.g., a smartwatch, a universal microchip, a subcutaneous microchip, a band, a bracelet, a ring, etc.), a sensor, software, or the like, capable of collecting information about the state (symptoms, physiological measurements, etc.) and the behavior (diet, psycho-physiological state, etc.) of a particular patient, referred to herein as parameter values. In some embodiments, the patient monitor may be in signal connection to the clinical database or ready for such connection. In certain embodiments, the collected information may be at least partially sent to the clinical database and stored therein.
[0043] In some embodiments of the computer-implemented method, the first data source is or includes a patient monitor or an intradialytic measurement device.
[0044] In a number of embodiments, the second data source is or includes a patient monitor, an intradialysis measurement device, a blood treatment device, and / or a reporting tool for receiving input from the patient and / or medical staff. Input may be via a touch screen, keyboard, gestures, language control, or in any other manner.
[0045] In some embodiments, the value of the first parameter and / or the second parameter is measured and / or is being measured by a wearable bioimpedance sensor, a photoplethysmograph, an accelerometer, a blood glucose sensor, a hemoglobin sensor, a sweat sodium (Na) and / or potassium (K) sensor, a calcium (Ca) sensor, a pulse meter, a skin conductance sensor and / or an actigraph. These items may form part of the first data source and / or the second data source.
[0046] In various embodiments, the first parameter and / or the second parameter comprises one or more parameters from the group consisting of:
[0047] -Sodium and potassium intake
[0048] -Muscle sodium content
[0049] - Drinking behavior patterns
[0050] - Water intake
[0051] - Water loss through perspiration and sweat
[0052] - Water balance
[0053] -Lean tissue index (LTI)
[0054] -Adipose tissue index (FAT)
[0055] - Intracellular Water (ICW)
[0056] - Extracellular water (ECW)
[0057] - Hydration status
[0058] -Heart rate
[0059] -Energy intake
[0060] -Energy consumption
[0061] - Sleep quality and duration
[0062] -Physical activity.
[0063] In some embodiments, the at least one second parameter comprises one or more parameters from a group consisting of or including data, wherein the data are preferably retrieved during the dialysis period from a blood treatment device as a second data source, including blood pressure dynamics, heart rate dynamics, blood electrolyte distribution (Na, K, Ca), dialysate electrolyte composition (Na, K, Ca), fluid status and / or administered drugs, preferably retrieved from a blood treatment device as a second data source, a patient monitor and / or a clinical database.
[0064] In various embodiments, the plurality of combinations of values provided in the clinical database include or consist of any combination of values of the set of parameters mentioned herein.
[0065] In some embodiments, said second parameter and / or said parameter retrieved from said clinical database comprises or consists of one or more parameters from the group consisting of clinical parameters, in particular age, gender, comorbidities, biochemical parameters, medications, history of intradialytic events, data collected from electronic chart records, patient's symptoms, behavioral attitudes, beliefs, and intradialytic and / or interdialytic events assessed using a computer adaptive testing application on a patient's smartphone or tablet computer using self-reported results.
[0066] In various embodiments, the output prediction of the future development of at least one physiological state of the patient or at least one of the patient's parameters, which in some embodiments may be a first physiological parameter, or the recommendation of a future behavior or treatment for the patient refers to:
[0067] - Intradialytic hypotension (IDH), specifically if, when and how likely hypotensive episodes are to occur;
[0068] - Maximum tolerable ultrafiltration (UF) rate;
[0069] - Interdialytic weight gain; and / or
[0070] - Pre-dialysis systolic and diastolic blood pressure.
[0071] In some embodiments, the step of suggesting future behavior or treatment for the patient based on the prediction is or includes suggesting specific water and salt intake between dialysis sessions.
[0072] In a number of embodiments, the step of suggesting future behavior or treatment includes or consists of individualized behavior alarms, nudges, and health education modules. Thus, the patient and / or less frequently, medical personnel can receive alarms, for example, via the patient monitor, the first data source, a portable device such as their watch, mobile device, smart phone, smart watch, or the like. In some embodiments, medical information can be provided to the patient or the medical personnel, such as the latest guidelines on nutrition, water intake management, auxiliary measures, and the like.
[0073] In some embodiments, outputting or delivering the recommendations to and / or making them available to the patient or medical personnel may be accomplished, for example, via a web application, such as for smartphones and tablets.
[0074] In various embodiments, the step of suggesting future behavior comprises suggesting or recommending water intake for said patient.
[0075] In some embodiments, the accuracy of the retrieved combination of multiple parameter values associated with the patient is evaluated with respect to the predicted value. It can then be optimized and stored as an optimized version, in particular in the clinical database.
[0076] In various embodiments, suggested future actions or treatments for the patient are stored or collected in the clinical database.
[0077] In some embodiments, the method according to the invention comprises the step of calibrating or optimizing said mathematical model.
[0078] In various embodiments, the predictions or recommendations of the output are used to calibrate or optimize the mathematical model. For example, the prediction of the output may be realized as predicted, or may not be realized as predicted. Thus, monitoring the predicted parameters over time can help optimize the mathematical model by adjusting the weights of various parameters, adjusting variables or constants used in the calculation of the predicted values, and so on. Artificial intelligence may be helpful here.
[0079] In some embodiments, if necessary, the patient monitor's measurements, including but not limited to fluid status, blood pressure, and heart rate, can be recalibrated using gold standard measurements that apply complex analytical techniques such as artificial intelligence (AI), neural networks, cutting-edge algorithms, mathematical functions, and integral transforms.
[0080] In various embodiments, the prediction of the output is evaluated using the predicted parameters or measurements of the state in an additional evaluation step. This evaluation step particularly occurs in the future.
[0081] In some embodiments, the prediction is for a specific time point or time window in the future. Then, evaluation measurements can be performed at that time point or time window.
[0082] In various embodiments, the results of the additional evaluation step are used to calibrate or optimize the mathematical model. The calibration or optimization may optionally include changing the weights of various parameters within the mathematical model.
[0083] In some embodiments, the computer-implemented method according to the present invention comprises the step of calibrating or optimizing multiple combinations of values of multiple parameters for multiple patients from a reference patient group on or in said clinical database based on the evaluation step described herein.
[0084] In the clinical database, a data set showing the (predetermined or previously selected) accuracy of the predicted parameters for the patient may be stored. This accuracy may be measured or determined based on measurements by reference sensors or reference methods. Once the method according to the invention has been used for a sufficiently long time, these stored data sets will allow the parameters in question to be determined or predicted with enhanced accuracy, and therefore, the values predicted by the mathematical model based on this enhanced database may themselves be included in the clinical database for further use.
[0085] In various embodiments, the patient's proposed future treatment is or includes treatment parameters that are recommended or requested to be completed, such as ultrafiltration volume, etc.
[0086] In some embodiments, the suggestion of treatment parameters can be sent directly to the blood treatment device via signal communication, so as to be used by the blood treatment device or its control device to perform the next blood treatment session.
[0087] In various embodiments, the suggested future action or treatment for a patient is or includes scheduling an appointment for the patient or setting a time frame for the patient for the next treatment session to occur.
[0088] In some embodiments, the computer-implemented method according to the invention comprises the step of correcting a parameter value measured or provided by said first data source by measurement of said parameter via said second data source.
[0089] In various embodiments, the principles of the present invention can be used in the absence of a blood treatment device, for example, in patients with kidney disease who are not yet receiving dialysis.
[0090] In some embodiments, the second data source is not a blood treatment device.
[0091] In various embodiments, the second data source of the system according to the invention is or comprises a patient monitor, in particular a smartphone or a smart watch and / or a wearable device.
[0092] In some embodiments, the system does not include a ring having an inner surface with an effective inner diameter configured to be wearable around a finger of the user. The computer-implemented method does not include using such a ring or a system having such a ring.
[0093] In various embodiments, the system does not include a processing device configured to determine a first power consumption level of the system. The computer-implemented method does not include using a processing device so configured.
[0094] In some embodiments, the system does not include a device or wearable device having a rigid shell including a bottom wall including the first channel, the second channel, and the third channel. The computer-implemented method does not include using such a device or wearable device.
[0095] In various embodiments, the term "combination" may be replaced by the terms collection, item, data set, or record. They may be considered synonyms in those embodiments.
[0096] Some or all embodiments according to the invention may include one, several or all of the advantages mentioned above and / or below.
[0097] Complications during hemodialysis (HD) sessions remain one of the major challenges in the management of dialysis patients. Intradialytic hypotension (IDH) is one of the most common adverse events in HD treatment. Due to the lack of a clear definition of IDH, the incidence of IDH varies greatly in different studies, ranging from less than 10% to nearly 70% of HD courses. Improper management of IDH may lead to adverse outcomes, including increased risk of death, hospitalization, and vascular access thrombosis. The present invention can advantageously minimize or eliminate this risk.
[0098] Even patients on peritoneal dialysis (PD), which theoretically allows for a more flexible treatment schedule, are not immune to complications associated with overhydration. In fact, a large proportion of PD patients suffer from symptomatic fluid retention. When overhydration occurs, hospitalization rates are much higher than those observed in the general PD population. In multiple observational studies, patient nonadherence to dietary restrictions and prescribed PD prescriptions is a major predictor of fluid retention.
[0099] Although multiple treatment algorithms have been proposed for both in-center and home dialysis, appropriate fluid management remains challenging. Data-driven AI models based on real-time patient behavior monitoring, sensor measurements, and prediction of physiological parameter trajectories and suggesting optimal treatment strategies, while enhancing patient self-care capabilities while monitoring dietary patterns and fluid status during dialysis intervals, will significantly improve patient outcomes.
[0100] Therefore, a possible advantage of the present invention may be that it provides an enhanced strategy for IDH prevention by predicting the occurrence of these events as an example of the patient's physiological state or the development of this state. Here, it is advantageous to use an AI-based classification model to predict the risk of IDH in the near future, especially in the next HD treatment.
[0101] By a feature selection strategy based on a trade-off between prediction accuracy and model complexity, a suitable set of input variables can be identified, for example, including 35 variables. All variables can be variables routinely collected in clinical practice, or they can be automatically recorded from blood treatment equipment, so that the clinical staff does not need additional work to use the present invention.
[0102] Another advantage of the present invention may be that, based on the evaluation step forming part of the method according to the present invention, a new or adapted personalized digital therapy program may be installed, thereby making it possible to prevent IDH in a specific patient.
[0103] Bioimpedance spectroscopy (BIS) technology, such as BCM (Body Composition Monitor, such as that distributed by Fresenius Medical Care, Germany), is a non-invasive and accurate method to measure body impedance at multiple electrical frequencies and estimate hydration status values of fitted model functions. In addition, BCM helps to study the connection between fluid status and blood pressure. Although there have been great improvements in HS control through BCM technology, the limitation of this method is the lack of continuous measurement, as it is time consuming, requires specific equipment, and can only be used by trained professionals before dialysis. Therefore, fluid dynamics and intake during the dialysis interval are not monitored. By continuously monitoring not only the patient's hydration status but also the vital signs and habits of the present invention, complication management and the patient's quality of life can be advantageously and continuously improved.
[0104] The present invention enables continuous hydration monitoring to be integrated into a clinical decision support system (CDSS), which is designed to advantageously activate the patient's self-care potential in hydration management and advantageously enables physicians to personalize dialysis prescriptions for dialysis-dependent CKD (DD-CKD: DialysisDependent CKD) and advanced non-dialysis-dependent CKD (stage 5, NDD-CKD: Advanced Non-DialysisDependent CKD).
[0105] Due to the monitoring and decision support systems included in the present invention, hydration management for DD-CKD patients and Stage 5 NDD-CKD patients (ie, medication dosage recommendations, dietary recommendations, behavior change interventions, etc.) can be advantageously optimized so as to be customized for the specific patient.
[0106] The framework of augmenting continuous monitoring measurements (via wearable devices) with clinical information can be advantageously applied to a large number of clinical problems of a diverse nature, such as adaptive medication administration and chronic disease management.
[0107] All advantages achievable by the method according to the present invention can in certain embodiments be achieved by the device, apparatus or system according to the present invention, and vice versa. BRIEF DESCRIPTION OF THE DRAWINGS
[0108] In the following, with reference to the accompanying drawings, the present invention is described based on preferred embodiments of the present invention. However, the present invention is not limited to these embodiments. In the accompanying drawings, the same reference numerals represent the same or similar elements, values, etc., and the following applies:
[0109] Figure 1The system according to the invention is shown in a first embodiment; and
[0110] Figure 2 An exemplary workflow of the method according to the present invention is shown. DETAILED DESCRIPTION
[0111] Figure 1 A system 100 according to the invention is shown in a first embodiment.
[0112] Management of the patient's hydration status (HS) preferably involves continuous (24 hour / 7 day) monitoring of the patient during and between dialysis sessions. The following examples relate to such continuous interdialytic fluid monitoring and management.
[0113] Can be used as Figure 2 The interdialytic and / or intradialytic monitoring values underlying the described method are realized by integrating information collected from multiple, partially selectable data sources. The transmission of data or information is indicated by arrows, respectively.
[0114] exist Figure 1 In the example of , the first data source is implemented as a patient's smartwatch 200. It can capture and / or collect intradialytic and intradialytic values of the following parameters:
[0115] 1. Sodium and potassium intake
[0116] 2. Muscle Sodium Content
[0117] 3. Drinking behavior patterns
[0118] 4. Water intake
[0119] 5. Water loss through perspiration and sweat
[0120] 6. Water balance
[0121] 7. Lean tissue index (LTI)
[0122] 8. Adipose tissue index (FAT)
[0123] 9. Intracellular Water (ICW)
[0124] 10. Extracellular Water (ECW)
[0125] 11. Hydration status
[0126] 12. Heart rate
[0127] 13. Energy intake
[0128] 14. Energy consumption
[0129] 15. Sleep quality and duration
[0130] 16. Physical activity
[0131] Typically, the values obtained using the first data source may be collected for the patient's entire body, at least a portion thereof, or with respect to a body layer (skin, tissue, or the like) of the patient.
[0132] Alternatively or additionally, for example during dialysis, the information may be obtained from Figure 1 In the example of the second data source 200 implemented as a blood treatment device 210, the information can be collected. For example, the information can include values from a body composition monitor (BCM), vital signs, laboratory tests and results, administered medications / drugs, blood pressure dynamics, heart rate dynamics, blood electrolyte distribution (e.g., Na, K, Ca), dialysate electrolyte composition (e.g., Na, K, Ca), and fluid status. This information can be used to compare with the information from the first data source 200 to complete the information or calibrate the information.
[0133] Alternatively or additionally, information may be collected by a reporting tool 220, such as a patient's smartphone or tablet, as a second data source. It may be particularly helpful to capture personal perceptions and impressions, such as (but not limited to) pain, nausea, sweating, perceived coldness or heat, etc. This may alternatively or additionally apply to the patient's symptoms, behavioral attitudes, and / or interdialytic events, which may be assessed with a computer adaptive testing application on the patient's smartphone 220 or tablet as a reporting tool or as a further data source. This information may again be used to complete the information from the first data source or calibrate the first data source.
[0134] Alternatively or additionally, data may be retrieved, for example, from a clinical database 250 in which, for example, patient history, treatment course, patient parameters, collective data, and the like may be collected and stored. In addition, clinical parameters may be collected from electronic chart records, including but not limited to age, sex, comorbidities, biochemical parameters, medications, intradialytic event history.
[0135] The first and / or second data sources 200, 210, 220, 250 may be or include, for example, but not limited to, a wearable bioimpedance sensor, a photoplethysmograph, an accelerometer, a blood glucose sensor, a hemoglobin sensor, a sweat sodium (Na) and potassium (K) sensor, a calcium (Ca) sensor, a pulse meter, a skin conductance sensor, and an actigraph, and the like.
[0136] The values retrieved and collected by the means mentioned herein and as described herein, in particular the values of the first and / or second parameters, are fed into a computing device 500 configured to include and run a mathematical model, for example in an evaluation device 530. Thus, these values may be or include raw values or pre-processed values, in particular calibrated values. Figure 1 As shown in the example of , a reading device 510 and a receiving device 520 may be provided for this purpose. These devices 510, 520, 530 may be part of the computing device 500, such as Figure 1 As shown, or may be separate devices, possibly in separate housings.
[0137] exist Figure 1 In an embodiment, the computing device 500 also includes an output device 540 or is in signal communication with the output device 540, through which the results obtained by the evaluation device can be processed for output, as described below.
[0138] In some embodiments, the first data source 200 and / or the second data source 210, 220 may be prepared and / or adapted to receive and / or display notifications, behavioral suggestions, or the like. Alternatively or additionally, it may be prepared to issue an alert or to react to the evaluation result of the first parameter as described herein (see Figure 2 ). The response of the second data source may in particular be or include an adjustment of, for example, a treatment parameter for the specific patient, in particular during their next blood treatment.
[0139] Figure 2 An exemplary workflow of the method according to the present invention is shown. Figure 1 's reference numerals.
[0140] In method step M1, first and second data sources 200, 210, 220, 250 and a computing device 500 are provided. In this case, the computing device 500 is adapted and / or provided to facilitate or control a computer-implemented method according to the invention for assessing a patient's physiological state, in particular hydration state.
[0141] Method step M2 represents reading in at least one value of at least one first parameter of the patient from the first data source 200. Thus, the first parameter is preferably a physiological parameter of the patient. The at least one value may be a specific value, a value range, a course of change of the value of a specific parameter over time or the like.
[0142] Method step M3 represents receiving at least one value of at least one second parameter from a second data source 210, 220, the second parameter being related to the patient or to the medical treatment of the patient. Preferably, such values are currently measured or otherwise collected and may not (yet) be stored in a database for later reference or use. Thus, for example, during a dialysis session, relevant individual and clinical information is collected, for example, by sensors of the blood treatment device 210. In addition, the patient can interactively add further information related to relevant events occurring during the dialysis session, for example via the reporting tool 220.
[0143] In some embodiments, such values may have been measured or collected in the past.
[0144] Method step M4 represents providing a clinical database 250 comprising a plurality of combinations of values of a plurality of parameters, which may include a first parameter and a second parameter. The plurality of values are obtained from the patient and / or from a plurality of patients of a reference patient group.
[0145] By means of method steps M2 to M3, a wide range of intra-dialytic and intra-dialytic information about the first parameter or the second parameter can be measured or otherwise collected by frequently or even constantly monitoring the patient, for example via a patient monitor, or more generally, via a first data source, or frequently monitoring the second parameter, for example via the treatment device 210, or more generally, via a second data source. This information can be stored as values in the clinical database 250.
[0146] Method step M5 represents providing a mathematical model for evaluating at least one read-in value based on at least one of a plurality of combinations of said parameter values comprised by the clinical database 250. The mathematical model may be executed or hosted by the computing device 500.
[0147] Method step M6 represents using a mathematical model to evaluate at least one read-in value of at least one first physiological parameter based on at least one of the combinations of parameter values while retrieving at least one of the multiple combinations of parameter values. The one or more combinations retrieved from the clinical database 250 may include, but are not limited to, the received value of at least a second parameter. In fact, the value of the second parameter may have been pre-collected and stored in the clinical database 250 before the value of the first parameter is available and therefore received. However, the value of the second parameter may be measured or collected at approximately the same time that the value of the first parameter has been selected. Therefore, the value of the second parameter may have been recently obtained when evaluated and therefore has not (yet) been stored.
[0148] Depend on Figure 2Method step M7 in the embodiment indicates that some or all of the captured and collected values, such as the first or second parameter, i.e. the physiological parameter and optionally the behavioral parameter, can be evaluated using, for example, signal integration. This may include (re)calibration of the data when necessary using gold standard measurements that apply complex analysis techniques such as artificial intelligence (AI), neural networks, cutting-edge algorithms, mathematical functions and integral transformations.
[0149] Step M7 may be based on a formula such as formula (1):
[0150] M gold_standard =f(M patient_monitor , individual_characteristics)+ε (1)
[0151] in,
[0152] f is an arbitrary function obtained by any analysis method, and the analysis method includes but is not limited to machine learning technology, mathematical approximation technology, mathematical transformation technology (such as Laplace transform, wavelet...), and cutting-edge algorithms.
[0153] M gold_standard is any relevant gold standard measurement. For example, gold standard measurements of hydration status may include different techniques including but not limited to isotope total body water (TBW), bioimpedance spectroscopy (BIS) techniques including but not limited to BCM (body composition monitor), multi-frequency whole body (BIA).
[0154] M patient_monitor Any relevant measurement obtained from the dialysis room and intradialytic phases using any patient monitor, including but not limited to hydration status, blood pressure, heart rate, energy balance, sleep duration and quality.
[0155] individual_characteristics represents a vector of patient-specific information derived from continuous intradialytic and interdialytic monitoring during and between blood treatment sessions using, for example, the patient monitoring device 200 and / or the blood treatment apparatus 210 .
[0156] ε is a vector describing the approximation error.
[0157] In method step M8, based on the results of the evaluation step described herein, the future development of at least one of the at least one physiological state or parameters of the patient is predicted, which in some embodiments may also be a first (physiological) parameter. Thus, the evaluation can give an outlook or forecast of how the first parameter may develop in the near or foreseeable future. However, the evaluation may not only predict the future course of the first parameter. Instead, the outlook or forecast may, for example, clarify the question of whether and when, for example, a hypotensive episode will occur, whether during dialysis (during a treatment session) or between dialysis sessions (between two treatment sessions).
[0158] In this method step, for example, the following predictions or forecasts can be made:
[0159] a. Intradialytic hypotension (IDH) and maximum tolerable ultrafiltration (UF) rate;
[0160] b. Interdialytic weight gain;
[0161] c. pre-dialysis systolic and diastolic blood pressure;
[0162] d. Water and salt intake between dialysis rooms;
[0163] e. Based on the predictions from a. to c., the maximum tolerable water and salt intake between dialysis sessions can be calculated;
[0164] Indicated by reference mark f., based on the mathematical model and predicted evaluation results as described herein, an individualized behavioral alert, supervision and health education module based on behavioral and physiological function monitoring during the dialysis interval is calculated in order to achieve the maximum tolerable water and salt intake calculated in e.
[0165] In method step M9, the features calculated in or for steps a. to f. are made available to the patient, for example by displaying them on, for example, a patient monitor 210, i.e., a patient's smartwatch, or a reporting tool 220, i.e., a patient's tablet. In this case, the personalized behavioral alert, supervision, and health education modules are preferably sent to the patient via a web application optimized for multiple platforms. Optionally and advantageously, the web application is optimized for the patient's smartphone 200 and / or tablet 240 that has been used with the method according to the invention.
[0166] Alternatively or additionally, the features calculated in or for steps a. to f. may be made available to a caregiver or physician to be able to adjust further treatment plans or treatment parameters for a particular patient. This may also include suggesting at least one time point for a subsequent treatment session.
[0167] In a particular embodiment, all or at least part of the data collected by the data sources 200, 210, 220 for intradialytic and / or interdialytic measurements may be collected in a clinical database 250. This also applies to the predictions implemented in steps a. to f. This information in the clinical database 250 may be used to steadily and continuously improve the accuracy of the mathematical model on which the method according to the invention is based.
[0168] Reference numerals list
[0169] 100 Systems
[0170] 200 First data source; e.g., dialysis room or intradialysis measurement device
[0171] (Here exemplarily in the form of a patient's smart watch)
[0172] 210 Second data source; intradialysis measurement device
[0173] (Here exemplarily in the form of a blood treatment device)
[0174] 220 Secondary data sources; reporting tools
[0175] (Here exemplarily in the form of a tablet for the patient)
[0176] 250 Clinical Databases
[0177] 500 Computing Devices
[0178] 510 Reading device
[0179] 520 Receiving Device
[0180] 530 Evaluation Device
[0181] 540 Output device
[0182] M1 to M9 method steps
[0183] a. to f. Forecasting and evaluation
Claims
1. A computer-implemented method for assessing a patient's physiological state, in particular his / her hydration state, The following steps are involved: - reading in at least one value of at least one first parameter of the patient from a first data source (200), The first parameter is a physiological parameter; - receiving at least one value of at least one second parameter from a second data source (210, 220), the second parameter being related to the patient or the medical treatment of the patient, the first data source (200) and said second data sources (210, 220) are at least partially different from each other; - providing a clinical database (250) comprising a plurality of combinations of values of a plurality of parameters, including the first parameter and the second parameter, wherein the plurality of values are obtained from the patient and / or from a plurality of patients of a reference patient group; - providing a mathematical model for evaluating at least one read-in value based on at least one of the plurality of combinations of parameter values comprised by the clinical database; - evaluating at least one read-in value of at least one first physiological parameter using the mathematical model, the evaluating being based on at least one of the combinations of parameter values while retrieving at least one of the plurality of combinations of parameter values, wherein the one or more combinations retrieved from the clinical database (250) include the received value of at least one second parameter and optionally also the read-in value of at least one first parameter; and - based on the result of the step of evaluating, outputting a prediction of a future development of at least one physiological state of the patient or at least one parameter of the patient, such as at least one first parameter, or suggesting a future behavior or treatment for the patient; wherein the first data source (200) is or includes a patient monitor or a dialysis room measurement device; and Therein, the second data source (210, 220) is or includes a patient monitor, an intradialysis measurement device, a blood treatment device and / or a reporting tool for receiving input from a patient and / or medical staff.
2. A computer-implemented method according to any one of the preceding claims, in, One or more values of the first parameter and / or the second parameter are measured and / or are being measured by a wearable bioimpedance sensor, a photoplethysmograph, an accelerometer, a blood glucose sensor, a hemoglobin sensor, a sweat sodium (Na) and / or potassium (K) sensor, a calcium (Ca) sensor, a pulse meter, a skin conductance sensor and / or an actigraph.
3. A computer-implemented method according to any one of the preceding claims, in, The first parameter and / or the second parameter include one or more parameters from the group consisting of: -Sodium and potassium intake -Muscle sodium content - Drinking behavior patterns - Water intake - Water loss through perspiration and sweat - Water balance -Lean tissue index (LTI) -Adipose tissue index (FAT) - Intracellular Water (ICW) - Extracellular water (ECW) - Hydration status -Heart rate -Energy intake -Energy consumption - Sleep quality and duration - physical activity, and / or Wherein, the at least one second parameter comprises one or more parameters from the group consisting of blood pressure dynamics, heart rate dynamics, blood electrolyte distribution (Na, K, Ca), dialysate electrolyte composition (Na, K, Ca), fluid status and / or administered drugs, or one or more parameters from the group consisting of blood pressure dynamics, heart rate dynamics, blood electrolyte distribution (Na, K, Ca), dialysate electrolyte composition (Na, K, Ca), fluid status and / or administered drugs, preferably retrieved from a blood treatment device (210), a patient monitor (200) and / or a clinical database (250) as a second data source during a dialysis period.
4. The computer-implemented method according to at least one of the preceding claims, in, the second parameters and / or the parameters retrieved from the clinical database (250) comprising or consisting of one or more parameters from the group consisting of clinical parameters, in particular age, sex, comorbidities, biochemical parameters, medications, history of intradialytic events, data collected from electronic chart records, patient's symptoms, behavioral attitudes, perceptions and intradialytic and / or interdialytic events assessed using a computer adaptive testing application for self-reported results on the patient's smartphone or tablet; and / or Wherein the output is a prediction or recommendation of future behavior or treatment for the patient and is related to or consists of: - Intradialytic hypotension (IDH); - Maximum tolerable ultrafiltration (UF) rate; - Interdialytic weight gain; and / or - Pre-dialysis systolic and diastolic blood pressure.
5. The computer-implemented method according to at least one of the preceding claims, in, Advising the patient of future behavior or treatment steps is or includes advising specific water and salt intake between dialysis sessions.
6. The computer-implemented method according to at least one of the preceding claims, in, The accuracy of the retrieved combination of multiple parameter values associated with the patient is evaluated relative to the predicted value, optimized and stored, in particular as an optimized version in the clinical database (250).
7. The computer-implemented method according to at least one of the preceding claims, in, The method comprises the steps of calibrating or optimizing the mathematical model, and / or Therein, the output predictions or recommendations are used to calibrate or optimize the mathematical model.
8. The computer-implemented method according to at least one of the preceding claims, in, As needed, measurements from the first data source (200), including but not limited to fluid status, blood pressure and heart rate can be recalibrated using gold standard measurements applying sophisticated analytical techniques such as artificial intelligence (AI), neural networks, cutting-edge algorithms, mathematical functions and integral transforms.
9. A computer-implemented method according to any one of the preceding claims, in, using the predicted parameters or measurements of the state in an additional evaluation step to evaluate the predictions of the output; and / or Therein, the results of the additional evaluation step are used to calibrate or optimize the mathematical model.
10. A computer-implemented method according to any one of the preceding claims, in, The proposed future treatment of the patient is or includes a recommendation or request for treatment parameters to be completed, such as ultrafiltration volume.
11. A computer-implemented method according to any one of the preceding claims, in, The recommended future action or treatment for the patient is or includes prioritizing the patient or setting a time frame for the next treatment period for the patient.
12. A system (100), include: - a first data source (200); - reading means (510) for reading in at least one value of at least one first parameter of the patient from said first data source (200), said first parameter being a physiological parameter; - a second data source (210, 220), said first data source (200) and said second data source (210, 220) being at least partially different from each other; - receiving means (520) for receiving from said second data source (210, 220) at least one value of at least one second parameter, said second parameter being related to the patient or to a medical treatment of the patient; - a clinical database (250) comprising a plurality of combinations of values of a plurality of parameters, including said first parameter and said second parameter, wherein the plurality of values are obtained from said patient and / or from a plurality of patients of a reference patient group; - a mathematical model for evaluating at least one read-in value based on at least one of said plurality of combinations of parameter values comprised by said clinical database; - evaluation means (530) for evaluating at least one read-in value of at least one first physiological parameter based on at least one of said combinations of parameter values while retrieving at least one of said plurality of combinations of parameter values using said mathematical model, wherein the one or more combinations retrieved from said clinical database (250) include the received value of at least a second parameter; and - output means (540) for outputting, based on the result of said step of evaluating, a prediction of a future development of at least one physiological state of the patient or at least one of the patient's parameters, or The researcher makes recommendations for the patient's future behavior or treatment.
13. The system (100) according to claim 12, in, The second data source (210, 220) is or comprises a patient monitor, in particular a smartphone or a smart watch and / or a wearable device.
14. A device, in particular a handheld device, comprising control means programmed to carry out or trigger the execution of a method according to any one of claims 1 to 11.
15. A device, in particular a handheld device, in particular a device according to claim 14, comprising two interfaces, which are programmed or configured to be able to read in at least one value of at least one first parameter of a patient from a first data source (200) and to receive at least one value of at least one second parameter from a second data source (210, 220), the first data source (200) and the second data source (210, 220) being at least partially different from each other.
