Method for evaluating behavioral response capability and electronic device

By acquiring historical monitoring data in non-motor states and combining it with analyte and acceleration data to assess patients' responsiveness to abnormal analyte events, this approach solves the complex and inefficient assessment problems of existing technologies, enabling quantitative assessment and personalized intervention in natural states.

CN123058720APending Publication Date: 2026-11-10SINOCARE
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611396782.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-09
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing technologies for observing patients’ responses to analyte abnormalities under controlled conditions are complex and inefficient, difficult to replicate frequently in natural living environments, and lack objective evidence for personalized alarm methods.

Method used

By acquiring historical monitoring data of the target object in a non-moving state, and combining the analyte data and acceleration data, abnormal events of the analyte are identified and behavioral response parameters are calculated. The analyte's ability to respond to abnormal events is evaluated, including the response when there is no alarm and the response after an alarm.

Benefits of technology

It enables the objective quantification of patients' behavioral response levels to abnormal analyte events under natural conditions, reduces testing costs and risks, improves long-term assessment efficiency, and provides a basis for personalized intervention.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123058720A_ABST
    Figure CN123058720A_ABST
Patent Text Reader

Abstract

This application discloses a method and electronic device for assessing behavioral response capabilities. The method includes: acquiring historical monitoring data of a target subject in a non-motor state, wherein the historical monitoring data includes analyte data collected by a continuous analyte monitoring device and acceleration data; determining analyte-related abnormal events based on the analyte data and acceleration data; determining behavioral response parameters of the target subject to the analyte-related abnormal events, wherein the behavioral response parameters characterize the degree to which the target subject's response behavior deviates from the target subject's behavioral baseline in a non-motor state when the analyte-related abnormal event occurs; and determining the target subject's behavioral response capability based on the behavioral response parameters. This application solves the technical problem of related technologies that use clamp tests and other methods to observe patients' responses to analyte-related abnormal events under controlled conditions, resulting in complex and inefficient testing processes.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of data processing technology, and more specifically, to a method and electronic device for evaluating behavioral response capabilities. Background Technology

[0002] For assessing a patient’s ability to respond to analyte-related events, related technologies, such as the hyperinsulinemia-hypoglycemia clamp test, are used to observe the subject’s physiological or behavioral response to analyte-related events under artificial control. Such tests require external means to forcibly induce an abnormal state of the analyte, and are monitored and measured by professionals throughout the process. The implementation process is complex, costly, and risky, and is not suitable for frequent repetition in natural living environments. It is also difficult to replace the objective observation of a patient’s long-term actual response to analyte-related events in real daily scenarios.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This application provides a method and electronic device for assessing behavioral response capabilities, which at least solves the technical problem that related technologies use methods such as clamp tests to observe patients' responses to abnormal analyte events under controlled conditions, resulting in complex and inefficient testing processes.

[0005] According to one aspect of the embodiments of this application, a method for evaluating behavioral response capability is provided, comprising: acquiring historical monitoring data of a target object in a non-motion state, wherein the historical monitoring data includes analyte data and acceleration data collected by a continuous analyte monitoring device; determining an analyte abnormal event based on the analyte data and acceleration data; determining behavioral response parameters of the target object to the analyte abnormal event, wherein the behavioral response parameters are used to characterize the degree to which the response behavior of the target object deviates from the behavioral baseline of the target object in a non-motion state when the analyte abnormal event occurs; and determining the behavioral response capability of the target object based on the behavioral response parameters.

[0006] According to another aspect of the embodiments of this application, a device for evaluating behavioral response capability is also provided, comprising: an acquisition module for acquiring historical monitoring data of a target object in a non-motion state, wherein the historical monitoring data includes analyte data collected by a continuous analyte monitoring device and acceleration data; a screening module for determining analyte abnormal events based on the analyte data and acceleration data; a determination module for determining behavioral response parameters of the target object to the analyte abnormal events, wherein the behavioral response parameters characterize the degree to which the target object's response behavior deviates from the target object's behavioral baseline in a non-motion state when the analyte abnormal event occurs; and an evaluation module for determining the behavioral response capability of the target object based on the behavioral response parameters.

[0007] According to another aspect of the embodiments of this application, an electronic device is also provided, including: a memory and a processor, wherein the memory is used to store program instructions; the processor is connected to the memory and is used to execute an evaluation method for implementing the above-mentioned behavioral response capability.

[0008] According to another aspect of the embodiments of this application, a non-volatile storage medium is also provided, the non-volatile storage medium including a stored computer program, wherein the device on which the non-volatile storage medium is located executes the above-mentioned method for evaluating behavioral response capability by running the computer program.

[0009] According to another aspect of the embodiments of this application, a computer program product is also provided, including computer instructions that, when executed by a processor, implement the above-described method for evaluating behavioral responsiveness.

[0010] In this embodiment, historical monitoring data of the target object in a non-motor state is acquired. Abnormal events of the analyte are determined based on the analyte data and acceleration data in the monitoring data. The behavioral response parameters of the target object to the abnormal events of the analyte are determined to characterize the degree to which the response behavior deviates from the behavioral baseline in the non-motor state. Then, the behavioral response ability of the target object is determined based on the behavioral response parameters. This achieves the goal of objectively quantifying the behavioral response level of the target object to the abnormal events of the analyte in a natural state without artificial induction. This achieves the technical effects of reducing testing costs and risks, improving long-term assessment efficiency, and providing a basis for personalized intervention. It also solves the technical problem of related technologies that use clamp tests and other methods to observe the patient's response to the abnormal events of the analyte under controlled conditions, which is complex and inefficient. Attached Figure Description

[0011] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0012] Figure 1 This is a hardware structure block diagram of a computer terminal for evaluating a behavior response capability according to an embodiment of this application.

[0013] Figure 2 This is a flowchart of a method for evaluating behavioral response capability according to an embodiment of this application;

[0014] Figure 3 This is a schematic diagram of the structure of a behavioral response capability assessment device according to an embodiment of this application. Detailed Implementation

[0015] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0016] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0017] To better understand the embodiments of this application, the technical terms involved in the embodiments of this application are explained below:

[0018] Continuous analyte monitoring device: A medical monitoring device that is implanted or worn on the human body surface (such as under the skin or on the skin surface) and can continuously or frequently collect target physiological index data and generate time-series monitoring data at a fixed frequency. In the embodiments of this application, the continuous analyte monitoring device is used to collect analyte concentration data, sensor status information and signal quality markers of the target object in a natural living state (such as a sleep state).

[0019] Analyte: In the fields of biochemistry and medical monitoring, analytes refer to chemical substances, molecules, or physiological indicators present in biological fluids (such as blood, interstitial fluid, urine, or saliva) that need to be quantitatively or qualitatively measured by specific sensors or detection methods. In the embodiments of this application, analytes include, but are not limited to, blood glucose (glucose), lactic acid, ketone bodies, cortisol, and other physiological and biochemical indicators that can be collected in real time through continuous monitoring technology.

[0020] Behavior Response Parameter: Based on the triaxial acceleration characteristics within the observation window, this is a quantitative indicator (such as a weighted combination of body movement enhancement score, posture change score, etc.) calculated relative to the target object's behavioral baseline in a non-motor state. In this embodiment, the behavior response parameter is used to numerically characterize the target object's behavioral activation level when an analyte abnormal event occurs in a non-motor state (such as hypoglycemia during sleep or ketoacidosis upon awakening), directly reflecting the target object's physiological or behavioral intervention intention and ability in response to the analyte abnormal event.

[0021] The widespread adoption of continuous analyte monitoring technology has made long-term, real-time tracking of individual physiological states possible. Assessing a target individual's ability to perceive and respond to analyte-related abnormalities is a universally important indicator for health monitoring, critical illness early warning, and risk management of special populations. Regardless of the type of analyte-induced physiological stress or whether the individual is awake, asleep, or in another non-motor state, objectively quantifying the target individual's autonomous response behavior in natural living scenarios helps to achieve early risk intervention, personalized early warning configuration, and improve the overall efficiency of health management.

[0022] Taking the monitoring of behavioral responses to nocturnal hypoglycemia in diabetic patients as an example, this study illustrates that diabetic patients, especially those with a long disease duration (type 1 diabetes), those receiving insulin therapy (type 2 diabetes), and those with a history of recurrent hypoglycemia, may experience impaired hypoglycemia perception. During sleep, patients find it difficult to actively observe and report hypoglycemic symptoms as they would when awake. Some patients may not wake up promptly or take further action after a nocturnal hypoglycemic episode, thus increasing the risk of persistent and severe hypoglycemia. Therefore, objectively understanding patients' behavioral responses to nocturnal hypoglycemic events in a natural sleep environment is of great significance for hypoglycemia risk management.

[0023] However, the relevant technologies have the following shortcomings in practical applications (taking diabetes as an example):

[0024] (1) Evaluation methods rely on patients' subjective memory: Clinically, the evaluation of hypoglycemia perception ability or impaired hypoglycemia perception mainly uses questionnaires such as the Gold Score, Clarke Questionnaire, and Pedersen method, as well as a review of hypoglycemia history. The above methods are usually classified or scored based on information such as whether the patient can perceive hypoglycemia, past symptoms, and whether they need assistance from others. They are easily affected by recall bias, the frequency of hypoglycemic events, and differences in patients' understanding of symptoms, and are difficult to objectively reflect the actual behavioral response in multiple natural sleep nights.

[0025] (2) Controlled hypoglycemia tests are not suitable for long-term home repetition: In some studies or specialist evaluations, the hyperinsulin-hypoglycemia clamp test can be used to observe the subject's symptoms, hormones, cognition, or arousal response under controlled conditions. This type of test requires induction of hypoglycemia through insulin infusion and continuous measurement and safety monitoring by professionals. The implementation process is complex, costly, and risky, and it is not suitable for frequent repetition in ordinary home environments. It is also difficult to replace long-term observation of the actual response during natural sleep nights.

[0026] (3) Lack of objective behavioral response evaluation based on natural hypoglycemic events: The continuous glucose monitoring system (CGM) used in the relevant technology can continuously output blood glucose data and issue an alarm when the blood glucose value is lower than the set threshold. However, its main function is to detect hypoglycemic state and prompt the patient to take timely action. There is still a lack of methods to quantitatively evaluate the patient's behavioral response when nocturnal hypoglycemic events occur, under the condition that no alarm is actually output and after different alarm methods are actually output.

[0027] (4) Lack of objective basis for personalized alarm method recommendations: The alarm settings of the continuous blood glucose monitoring system used in the relevant technologies are usually determined based on unified rules, user experience or medical staff advice. Different patients may have different nighttime responses to different alarm methods. The relevant technologies lack data basis for individualized evaluation and recommendation of alarm methods based on the patient's historical objective behavioral responses to different alarm methods in multiple natural sleep nights.

[0028] To address the aforementioned technical problems, this application provides corresponding solutions, which are detailed below.

[0029] The behavioral response capability evaluation method provided in this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1 A hardware block diagram of a computer terminal for implementing a method for evaluating behavioral response capabilities is shown. Figure 1 As shown, the computer terminal 10 may include a processor 102, a memory 104 for storing data, and a transmission module 106 for communication functions. The processor 102 may include one or more processors. For ease of explanation, Figure 1The illustration uses multiple processors, including a first processor 102a, a second processor 102b, ..., an nth processor 102n. Processor 102 may include, but is not limited to, processing devices such as microcontroller units (MCUs) or field-programmable gate arrays (FPGAs). In addition, it may include: a display, a keyboard, a cursor control device, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, and a BUS bus. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0030] It should be noted that the aforementioned one or more processors and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10. As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0031] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the behavior response capability evaluation method in this embodiment. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned behavior response capability evaluation method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0032] The transmission module 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission module 106 includes a network interface controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission module 106 may be a radio frequency (RF) module, used for data interaction with external networks or devices via wired and / or wireless network connections.

[0033] The display can be, for example, a touchscreen liquid crystal display (LCD), which allows the user to interact with the user interface of the computer terminal 10. Furthermore, the computer terminal 10 can be connected to a cursor control device and a keyboard via an input / output interface. The cursor control device (e.g., a mouse, trackball, or touchpad) is primarily used to transmit directional information and command selections to the terminal and control the movement of the cursor on the display; the keyboard is primarily used to receive letters, numbers, and other control commands input by the user, thereby enabling interaction between the user and the computer terminal 10.

[0034] It should be noted here that, in some optional embodiments, the above... Figure 1 The computer terminal shown may include hardware elements (including circuitry), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware and software elements. It should be noted that... Figure 1 This is only one instance of a specific particular instance, and is intended to illustrate the types of components that may exist in the aforementioned computer terminal.

[0035] In the above operating environment, this application provides an embodiment of a method for evaluating behavioral response capabilities. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0036] Figure 2 This is a flowchart of a method for evaluating behavioral response capability according to an embodiment of this application, such as... Figure 2 As shown, the method includes the following steps:

[0037] Step S202: Obtain historical monitoring data of the target object in a non-moving state. The historical monitoring data includes analytical data and acceleration data collected by the continuous analytical monitoring device.

[0038] In step S202 above, historical monitoring data refers to a multimodal data set collected and stored by the continuous analyte monitoring device in a time-series manner during the period the target object wears the device, used to assess the target object's ability to respond to abnormal analyte events. In some embodiments of this application, historical monitoring data may include analyte data (such as blood glucose concentration, sensor status, and signal quality markers) and acceleration data (such as triaxial acceleration values ​​and their derived dynamic acceleration characteristics) collected by the continuous analyte monitoring device. It may also further include alarm records, which record in detail the alarm event information actually output by the device during monitoring, such as alarm trigger time, alarm method (such as sound, vibration, screen prompt, etc.), alarm intensity, duration, and number of repetitions.

[0039] Among them, analyte data refers to time-series data collected by electrochemical or optical sensors in continuous analyte monitoring devices, reflecting changes in the concentration of specific chemical substances (such as glucose, lactic acid, ketone bodies, etc.) in the body fluids of the target object. Acceleration data refers to time-series data collected by triaxial accelerometers integrated into continuous analyte monitoring devices or associated wearable devices, representing the acceleration of the device's motion in three-dimensional space, used to quantify the intensity of the target object's behavioral activities, posture changes, and body movement frequency.

[0040] It should be noted that the non-motor state refers to the target object's daily physiological state of being at rest, sleeping, resting, or with very low metabolic consumption, which is different from the state of strenuous exercise or high-intensity physical activity. This is to eliminate the interference of large acceleration signals generated by high-intensity exercise on the establishment of behavioral baseline, and to ensure that the behavioral features extracted subsequently mainly reflect the physiological responses in the resting or micro-movement state. This allows for a more accurate assessment of the target object's autonomous perception and response to abnormal events of the analyte, rather than passive behavioral changes caused by exercise.

[0041] In some embodiments of this application, the sampling time of analyte concentration collected by the device and the acceleration sampling time of the accelerometer can be mapped to a unified time axis through the clock synchronization mechanism inside the continuous analyte monitoring device. Specifically, taking hypoglycemia monitoring during sleep periods as an example, blood glucose monitoring data, triaxial acceleration data, and alarm records are acquired from multiple nighttime sleep periods of the same subject. The above data are mapped to a unified time axis according to a common device clock or timestamp, and resampling or window aggregation is performed according to preset time units. The triaxial accelerometer and blood glucose sensor are housed within the same continuous blood glucose monitoring device, worn on the subject's body along with the continuous blood glucose monitoring device, and output acceleration values ​​in three mutually perpendicular directions. (t), (t), (t) Blood glucose monitoring data should at least include sampling time and blood glucose value, and may also include blood glucose trend, signal quality markers, and sensor status information. Triaxial acceleration data should at least include sampling time and... (t), (t), (t). Alarm logs should include at least the actual time of alarm output, and may also include alarm type, alarm intensity, duration, and number of repetitions.

[0042] Step S204: Determine the abnormal events of the analyte based on the analyte data and acceleration data.

[0043] In step S204 above, an analyte abnormality event refers to a physiological state change in which the concentration of a specific chemical substance (such as glucose, lactic acid, etc.) in the body fluid of the target object deviates from the normal physiological range, reaches or exceeds a preset clinical threshold, and lasts for a certain period of time. For example, in the context of diabetes management, it manifests as a hypoglycemic event (blood glucose below 3.9 mmol / L) or a hyperglycemic event.

[0044] In some embodiments of this application, analyte anomaly events can be determined through the following steps: acquiring analyte data and acceleration data of the target object in a non-motion state; identifying analyte anomaly candidate events with analyte concentrations below a preset anomaly threshold based on the analyte data; extracting multi-stage features for each analyte anomaly candidate event, wherein the multi-stage features include, but are not limited to, descent stage features, recovery stage features, and behavioral or posture auxiliary features; performing confidence processing on the analyte anomaly candidate events based on the combination relationship between the multi-stage features, excluding candidate events identified as suspected pressure-induced sensor attenuation; and confirming the analyte anomaly candidate events retained after confidence processing as genuine analyte anomaly events.

[0045] It should be noted that the characteristics of the analyte concentration decrease phase include the magnitude, duration, and rate of decrease; the characteristics of the recovery phase include the magnitude, duration, and rate of recovery of the analyte concentration; and the behavioral or posture-assisted characteristics include whether a change in body movement or device orientation that reaches a preset body movement threshold occurs during the event, and the time interval between the moment of the corresponding change in body movement or device orientation and the moment when the analyte concentration recovery begins.

[0046] Furthermore, the credibility can be processed as follows: when an analyte anomaly candidate event meets the following conditions simultaneously, it is identified as a sensor artifact event suspected to be caused by physical pressure or mechanical interference and is excluded: the magnitude or rate of decrease in analyte concentration exceeds a preset threshold; a change in body movement or device orientation that reaches a preset body movement threshold occurs before the analyte concentration recovery begins; the analyte reading begins to recover rapidly within a preset time after the body movement or device orientation change and the recovery rate exceeds a preset recovery rate; the magnitude of analyte concentration recovery reaches a preset recovery magnitude or approaches the baseline level before the event occurs within a preset recovery time.

[0047] To facilitate understanding of the screening process for abnormal analyte events described above, the following explanation is provided in conjunction with some specific examples.

[0048] Taking hypoglycemia as an example, during nighttime sleep, candidate hypoglycemia events are identified based on collected blood glucose levels, the duration of low values, and actual changes in the blood glucose curve. For instance, when a blood glucose level is below a preset hypoglycemia threshold (e.g., 3.9 mmol / L), it is identified as a candidate hypoglycemia event. It should be noted that adjacent low-value segments with an interval shorter than a preset merging time can be merged into the same candidate event. The aforementioned threshold and merging time can be determined based on the evaluation object, equipment settings, or application requirements.

[0049] For each hypoglycemia candidate event, at least the features of the descent phase, the recovery phase, and behavioral or postural auxiliary features should be extracted. Low-value phase features and other auxiliary features can also be extracted, including:

[0050] (1) Characteristics of the decline phase, including the magnitude of blood glucose decline, duration of decline, and rate of decline;

[0051] (2) Low-value phase characteristics, including the lowest blood glucose value and the duration below the preset low blood glucose threshold;

[0052] (3) Characteristics of the recovery phase, including the magnitude of blood glucose recovery, duration of recovery, and rate of recovery;

[0053] (4) Behavioral or postural auxiliary features, including whether a body movement or device orientation change that reaches a preset body movement threshold occurs during the event, and the time interval between the time of the corresponding body movement or device orientation change and the time of blood glucose recovery.

[0054] (5) Other auxiliary features, including sensor status or signal quality markers, second blood glucose signals, other physiological signals or human confirmation information.

[0055] Furthermore, based on the combined relationships between descent phase features, recovery phase features, and behavioral or posture-assisted features, the credibility of hypoglycemia candidate events is processed. Specifically, a hypoglycemia candidate event is identified as a suspected pressure-induced sensor attenuation event when it simultaneously meets the following conditions:

[0056] (1) The magnitude or rate of blood glucose decrease exceeds the corresponding preset threshold;

[0057] (2) During the low-value phase or before the blood glucose recovery begins, a change in body movement or device orientation occurs that reaches the preset body movement threshold;

[0058] (3) The blood glucose reading begins to recover within a preset time after the change in the direction of the body movement or device, and the recovery rate exceeds the preset recovery rate;

[0059] (4) The blood glucose recovery reaches the preset recovery range, or is close to the blood glucose level before the event occurs within the preset recovery time.

[0060] Candidate events identified as potentially pressure-induced sensor attenuation are excluded. Furthermore, when sensor status information, signal quality markers, secondary blood glucose signals, other physiological signals, or manual confirmation information are available, the judgment results can be further verified using this information. Finally, the hypoglycemic candidate events retained after confidence processing are considered as genuine nocturnal hypoglycemic events and proceed to subsequent analysis.

[0061] It should be noted that the features and judgment conditions extracted in the above embodiments can be used to judge abnormal events of other analytes, and can be reasonably adjusted according to the properties of the analytes, without limitation here.

[0062] Step S206: Determine the behavioral response parameters of the target object to the analyte abnormal event, wherein the behavioral response parameters are used to characterize the degree to which the response behavior of the target object deviates from the behavioral baseline of the target object in a non-moving state when the analyte abnormal event occurs.

[0063] In step S206 above, the behavioral response parameter refers to a quantitative value calculated based on the behavioral characteristic data of the target object when the abnormal event of the analyte occurs, relative to its own behavioral baseline established in a non-moving state. It is used to numerically characterize the target object's perception and response intensity to the abnormal event of the analyte. In some embodiments of this application, the behavioral response parameter may include a non-alarm behavioral response parameter (i.e., a second behavioral response parameter) and an alarm-post-alarm behavioral response parameter (i.e., a first behavioral response parameter). The non-alarm behavioral response parameter is used to quantify the degree of arousal or physical movement generated by the patient solely through physiological perception without receiving external cues; the alarm-post-alarm behavioral response parameter is used to quantify the degree of behavioral activation of the patient after receiving an alarm in a specific manner.

[0064] A behavioral baseline is an individualized reference model of normal behavior established based on acceleration data collected from the target subject in multiple non-motor states (such as normal sleep segments), after filtering, feature extraction, and statistical processing (such as calculating the median and median absolute deviation). It represents the typical behavioral pattern of the target subject in a resting or micro-movement state (such as normal breathing fluctuations, occasional turning over, etc.).

[0065] In some embodiments of this application, the behavioral response parameters of a target object to an analyte anomaly event can be determined by: determining target data from historical monitoring data, wherein the target data is used to characterize the behavioral characteristics of the target object when it is in a non-moving state and no analyte anomaly event has occurred; determining the behavioral baseline of the target object in a non-moving state based on the target data, wherein the behavioral baseline is used to characterize the normal behavioral pattern of the target object in a non-moving state; determining the response behavioral characteristics of the target object when an analyte anomaly event occurs; and determining the behavioral response parameters based on the behavioral baseline and the response behavioral characteristics.

[0066] It should be noted that target data refers to raw behavioral monitoring data segments selected from historical monitoring data that meet specific filtering conditions. These segments must meet conditions such as "non-motor state" (e.g., sleep or rest), "no abnormal events of the analyte" and "no alarm output" to provide a clean sample set for establishing an individualized behavioral reference model, eliminate interference from physiological abnormalities and external cues, and ensure that the established behavioral baseline can truly reflect the natural behavioral patterns of the target object in a healthy, resting state.

[0067] Specifically, taking hypoglycemia monitoring during sleep as an example, the historical monitoring data is traversed, and normal sleep segments that are neither hypoglycemic events nor have triggered alarms and whose acceleration quality is qualified are selected based on timestamps, blood glucose status markers and alarm markers, and a normal sleep data set (i.e. target data) is constructed.

[0068] Furthermore, the behavioral baseline of the target object in a non-motion state can be determined in the following ways: Based on the target data, determine multiple target behavioral features of the target object, wherein the multiple target behavioral features are used together to characterize the normal behavior of the target object in a non-motion state; determine the baseline parameters corresponding to each target behavioral feature, wherein the baseline parameters are used to characterize the central trend and normal fluctuation range of the target object's behavior in a non-motion state; and use the baseline parameters corresponding to the multiple target behavioral features as the behavioral baseline.

[0069] It should be noted that the multiple target behavioral features include, but are not limited to, a first target behavioral feature reflecting the amplitude of body movement, a second target behavioral feature reflecting the frequency of posture changes, a third target behavioral feature reflecting the duration of violent movements, and a fourth target behavioral feature reflecting the degree of stability in a non-motion state.

[0070] Baseline parameters are quantitative indicators used to define the central value and dispersion of normal behavioral ranges, calculated based on the statistical distribution of target behavioral characteristics. In this application, baseline parameters typically include a central value and a fluctuation scale. Their function is to establish a personalized normal reference standard for each behavioral characteristic. The central value represents the normal level, and the fluctuation scale represents the boundary of normal physiological fluctuations (such as breathing fluctuations, micro-turning, etc.).

[0071] Specifically, taking hypoglycemia monitoring during sleep as an example, four types of behavioral features (activity intensity, proportion of posture changes, proportion of time spent on obvious movements, and proportion of time spent on low activity) are extracted from each segment of the normal sleep data set. The median of these four features in the entire set is calculated as the baseline center value, and the median absolute deviation after scaling is calculated as the baseline fluctuation scale (i.e., baseline parameter), thereby establishing an individual's normal sleep behavior baseline.

[0072] Furthermore, for each real-world analyte anomaly, the corresponding observation window is captured, and the same four types of response behavior features are extracted. The behavior response parameters are determined based on the standardized deviation score (Z-score) relative to the behavior baseline. For example, the standardized score can be divided by a preset normalization scale parameter, and the individual score can be obtained by limiting it to the range of 0-1 through a truncation function. Then, the scores are weighted and summed according to preset weights to generate the final behavior response parameters.

[0073] To facilitate understanding of the process of establishing the above behavioral baseline, the following explanation will be provided in conjunction with some specific examples.

[0074] Taking hypoglycemia during sleep as an example, multiple nighttime sleep periods are divided into behavioral analysis time units of duration T. From these, time units that do not involve hypoglycemia events or their preset adjacent periods, do not actually output alarms, and whose triaxial acceleration data quality meets requirements are selected to form an individual normal sleep data set.

[0075]

[0076] in, Let N represent the j-th normal sleep behavior analysis time unit, and let N represent the number of normal sleep behavior analysis time units.

[0077] (1) Extraction of normal sleep behavior characteristics.

[0078] set up Contains There are 3 triaxial acceleration sampling points, with a sampling interval of . Calculate the resultant vector of triaxial accelerations. The three axes are then low-pass filtered to obtain the low-frequency gravity direction vector. and its modulus :

[0079]

[0080]

[0081] Furthermore, according to and Calculate dynamic acceleration :

[0082]

[0083] Furthermore, for each Extract the following four types of behavioral features (i.e., target behavioral features):

[0084] 1) Activity intensity characteristics:

[0085]

[0086] It should be noted that this feature represents The larger the value of the overall body motion amplitude, the higher the activity intensity within that time unit.

[0087] 2) Characteristics of posture change ratio:

[0088] Calculate the angle between the directions at adjacent attitude determination moments based on the aforementioned low-frequency gravity direction vector G(t):

[0089]

[0090]

[0091] It should be noted that this feature represents The percentage of sampling intervals where a significant change in device orientation occurs; the larger the value, the more frequent the change in attitude orientation.

[0092] 3) Obvious characteristics in the proportion of action time:

[0093]

[0094] It should be noted that this feature represents The proportion of time it takes for the internal dynamic acceleration to reach the action threshold; the larger the value, the longer the duration of the obvious body movement.

[0095] 4) Low activity time percentage characteristic:

[0096]

[0097] It should be noted that this feature represents The proportion of time spent in a static or low-activity state; the larger the value, the closer the time unit is to a stable sleep state.

[0098] in, Action threshold, For low activity threshold, The threshold values ​​are for the angle of posture change. Each threshold can be determined based on the device noise level, the distribution of individual normal sleep data, or labeled data.

[0099] In some embodiments of this application, < This allows for a slight range of activity between a low-activity state and a state of significant movement.

[0100] (2) Baseline parameter calculation.

[0101] For the kth behavioral characteristic Where k=1, 2, 3, 4, calculate the baseline center value respectively. absolute deviation of median and baseline fluctuation scale :

[0102]

[0103]

[0104]

[0105] in, To prevent Take a positive number that is zero.

[0106] Furthermore, it consists of four types of characteristics ( , This constitutes a baseline for an individual's normal sleep behavior. It should be noted that when normal sleep data is insufficient, the reliability of the baseline should be reduced or the corresponding event should be marked as insufficient baseline evidence. Alternatively, the baseline can be established using the mean and standard deviation or quantile range.

[0107] In some embodiments of this application, the historical monitoring data also includes alarm records triggered by the continuous analyte monitoring device. Based on this, the behavioral response parameters of the target object to analyte anomalies can be determined as follows: determining whether an analyte anomaly triggers an alarm based on the alarm records, obtaining a judgment result; determining the observation time window corresponding to each analyte anomaly based on the judgment result, wherein a first observation time window is used when an analyte anomaly triggers an alarm, and a second observation time window is used when an analyte anomaly does not trigger an alarm, the first observation time window starting at the time of the first output alarm corresponding to the analyte anomaly, and the second observation time window starting at the time the analyte anomaly is triggered; and determining the behavioral response parameters of the target object to the analyte anomalies within the observation time window.

[0108] It should be noted that alarm records refer to structured data generated by the continuous analyte monitoring device during monitoring when the blood glucose level reaches a preset threshold. This data includes information such as the alarm trigger time, alarm method (e.g., sound, vibration, screen prompt), alarm intensity, and duration. In some embodiments of this application, alarm records are used to determine the starting anchor point for behavioral response assessment: if an alarm is triggered, the patient's response to external prompts is assessed starting from the alarm output time; if no alarm is triggered, the patient's autonomous response ability is assessed starting from the time of the hypoglycemic physiological event.

[0109] Specifically, by reading alarm records from historical monitoring data, the system determines whether the current analyte abnormal event actually outputs a patient-perceptible alarm (such as sound or vibration): if it is determined that an alarm has been triggered, the system extracts the timestamp of the first alarm output of the event as the starting point of the first observation time window, and extracts data for a preset duration (such as 15 minutes) thereafter; if it is determined that no alarm has been triggered (including reasons such as the setting not being enabled or the threshold not being reached), the system extracts the trigger time of the actual hypoglycemic event (the moment when blood glucose is lower than the threshold) as the starting point of the second observation time window, and extracts data for the same preset duration thereafter.

[0110] Taking hypoglycemia as an example, the alarm records corresponding to the identified actual hypoglycemia events are read, and the events are divided into non-alarm events and alarm events according to whether an alarm that can be perceived by the patient is actually output after the event occurs.

[0111] Sound, vibration, screen prompts, or other notifications actually output by devices worn by the patient or placed within the patient's perception range are considered actual alarms. If an alarm function is set but no notification is actually output in the current event, it is considered a non-alarm event. If alarm records are missing, it cannot be confirmed whether an actual alarm was output, or it cannot be confirmed whether the notification was within the patient's perception range, it is marked as an alarm status uncertainty event and will not be included in the calculation of capability parameters for non-alarm or alarm events.

[0112] (1) For non-alarm events, take the time of triggering the hypoglycemia event as the starting point of the window and take the preset observation time T to form a non-alarm behavior response observation window.

[0113] (2) For alarm events, take the moment when CGM first actually outputs an alarm as the starting point of the window, and take the same preset observation duration T to form an observation window for post-alarm behavior response, and record the actual alarm method, intensity, duration and number of repetitions.

[0114] Furthermore, within the selected time window, the acceleration data is filtered, gravity components are estimated, and dynamic acceleration is calculated to extract multi-dimensional features (such as activity intensity, attitude changes, etc.), which are then substituted into a pre-established behavioral baseline to obtain the corresponding behavioral response parameters.

[0115] In some embodiments of this application, behavioral response parameters of a target object to an abnormal event of the analyte can be determined within an observation time window in the following manner: Multiple behavioral response characteristics of the target object to the abnormal event of the analyte are determined within the observation time window, wherein the multiple behavioral response characteristics are collectively used to characterize the target object's physiological or behavioral response level in multiple dimensions; multiple behavioral response scores are determined based on the deviation of the multiple behavioral response characteristics from the corresponding behavioral baseline, wherein the multiple behavioral response scores are collectively used to characterize the target object's perception level of the abnormal event of the analyte; and behavioral response parameters are determined based on the multiple behavioral response scores.

[0116] It should be noted that the multiple response behavior features include, but are not limited to, a first response behavior feature reflecting the amplitude of body movement, a second response behavior feature reflecting the frequency of posture changes, a third response behavior feature reflecting the duration of violent movements, and a fourth response behavior feature reflecting the degree of stability in a non-motion state.

[0117] Behavioral response scores are standardized numerical values ​​calculated by comparing response behavioral characteristics to an individual's baseline normal sleep behavior, reflecting the degree of deviation of the response characteristics from the norm. For example, an increased activity score reflects an abnormal increase in activity intensity, while a decreased rest score reflects an abnormal disruption of rest.

[0118] In some embodiments of this application, multiple behavioral response scores can be determined by: determining the degree of deviation between multiple response behavioral features and the corresponding target behavioral features in the behavioral baseline of the target object; obtaining the scale parameters corresponding to the multiple response behavioral features, wherein the scale parameters are determined by the preset high quantile of the standardized deviation degree corresponding to the normal behavioral data of the target object in a non-motion state; and determining multiple behavioral response scores based on the degree of deviation values ​​and scale parameters corresponding to the multiple response behavioral features.

[0119] It should be noted that the deviation value refers to the statistical difference between the response behavioral characteristic and the corresponding target behavioral characteristic in the baseline of the target object's behavior. For example, it can be characterized by a standardized score (Z-score) obtained by calculating the difference between the response characteristic value and the baseline center value (such as the median), and then dividing it by the baseline fluctuation scale (such as the median absolute deviation transformed value). This can be used to quantify the extent to which the current behavioral state deviates from the normal physiological state, eliminate the influence of differences in baseline levels between different individuals, and make the behavioral data of different patients comparable. For example, a patient with extremely low baseline activity intensity may show a significant deviation after standardization if their activity intensity increases due to a slight turning over, while another patient with a high activity baseline will show a smaller deviation with the same turning over.

[0120] Specifically, taking sleep state as an example, the standardized deviation score of each response behavior feature relative to the individual's normal sleep behavior baseline can be calculated; the standardized deviation score of all segments in the historical normal sleep data can be reviewed, and its preset high percentile (e.g., the 95th percentile) can be calculated. This percentile value can be used as the scale parameter of the feature; the standardized deviation score of the current event can be divided by the scale parameter to obtain a preliminary score; a clipping function can be applied to limit the calculation result to the range of 0 to 1: if the result is less than 0, it is taken as 0, and if it is greater than 1, it is taken as 1, thus obtaining the final behavioral response score.

[0121] The above steps normalize the degree of deviation by introducing a scale parameter, thereby achieving standardization and robustness of behavioral response assessment. An adaptive threshold mechanism is established to transform absolute behavioral differences into relative probability intensity. By using the high quantile of normal data to determine the scale parameter, normal physiological micro-movements and abnormal arousal responses can be effectively distinguished, avoiding misjudgments caused by individual baseline differences or short-term sleep state fluctuations.

[0122] In some embodiments of this application, the analyte anomaly event includes a first analyte anomaly event that has triggered an alarm and a second analyte anomaly event that has not triggered an alarm. Based on this, the behavioral response parameters can be determined as follows: a first behavioral response score set corresponding to at least one first analyte anomaly event is obtained, and a first behavioral response parameter corresponding to the alarm type of the first analyte anomaly event is determined based on the first behavioral response score set; a second behavioral response score set corresponding to at least one second analyte anomaly event is obtained, and a second behavioral response parameter corresponding to the second analyte anomaly event is determined based on the second behavioral response score set.

[0123] It should be noted that the first analyte anomaly event refers to an event in which, during monitoring, the concentration of the analyte in the target object (such as blood glucose) exceeds the normal range, and the continuous analyte monitoring device actually outputs an alarm signal (such as sound, vibration, screen prompt, etc.) that can be perceived by the patient according to preset rules. The second analyte anomaly event refers to an event in which, during monitoring, the concentration of the analyte in the target object exceeds the normal range, but the continuous analyte monitoring device does not actually output an alarm signal (e.g., due to alarm function being turned off, improper threshold setting, or system malfunction).

[0124] Specifically, historical monitoring data can be traversed, and identified hypoglycemic events can be divided into two categories based on alarm records: one category is the set of events where the device actually output an alarm signal (the first analyte abnormal event set), and the other category is the set of events where the device did not output an alarm signal (the second analyte abnormal event set). For each set, the system extracts multiple behavioral response scores (such as body movement enhancement score, posture change score, etc.) corresponding to each event, and performs a weighted summation based on preset weights (such as equal weight 0.25) to obtain the single-event behavioral response parameters for each event.

[0125] Furthermore, the arithmetic mean of all single-event behavioral response parameters in the first set of abnormal events of the analyte is taken to generate the first behavioral response parameter; the arithmetic mean of all single-event behavioral response parameters in the second set of abnormal events of the analyte is taken to generate the second behavioral response parameter.

[0126] The above steps achieve a refined separation of assessment dimensions by distinguishing between two distinct physiological-behavioral interaction scenarios: "alarm triggered" and "no alarm triggered," and calculating the corresponding behavioral response parameters for each. On one hand, the second behavioral response parameter objectively quantifies the patient's ability to autonomously perceive hypoglycemia, identifying the risk of impaired hypoglycemia perception. On the other hand, the first behavioral response parameter objectively evaluates the effectiveness of existing alarm strategies, avoiding misjudging alarm-induced behaviors as autonomous perception.

[0127] To facilitate understanding of the process of determining the above behavioral response parameters, the following explanation is provided in conjunction with some specific embodiments.

[0128] Taking the monitoring of hypoglycemic events during sleep as an example, the data processing method in the above-described example of establishing the behavioral baseline can be used to perform the same filtering, gravity component estimation, and dynamic acceleration calculation on the entire observation window corresponding to the i-th real hypoglycemic event, and the following four types of event behavioral features (i.e., response behavioral features) can be extracted using the same calculation method as when establishing the baseline:

[0129] (1) Activity intensity characteristics The root mean square of the dynamic acceleration within the observation window is used as the representation.

[0130] (2) Characteristics of posture change ratio The percentage of sampling intervals in which the low-frequency gravity direction angle within the observation window reaches the preset attitude change angle is used to represent the angle.

[0131] (3) Obvious characteristics of the proportion of action time The percentage of time within the observation window during which the dynamic acceleration reaches the action threshold is used to represent the action threshold.

[0132] (4) Low activity time ratio The percentage of time during which dynamic acceleration is in a low activity range within the observation window is used to represent this.

[0133] in, , , , Compared with the above embodiments respectively , , , One-to-one correspondence. The standardized deviation of the k-th event behavior characteristic from the individual's normal sleep behavior baseline is: .

[0134] Furthermore, the motion enhancement score is calculated based on the degree of deviation of the four types of features. Posture change score Continuous activity score Static points decrease (i.e., multiple behavioral response scores):

[0135]

[0136]

[0137]

[0138]

[0139] in, to For normalized scaling parameters greater than 0, clip[x,0,1] represents restricting x to the range of 0 to 1. In one implementation, for normal sleep behavior characteristics, according to... Calculate its standardized deviation from the baseline of normal sleep behavior, and then | The preset high quantile is determined as the corresponding The preset high quantile can be the 90th to 99th percentiles. It should be noted that when normal sleep data is insufficient or the preset high quantile is too small, it can be determined using a preset positive number, historical individual data, or group reference data. .

[0140] Furthermore, for the i-th non-alarm event mentioned above (i.e., the second analyte abnormal event), based on the motion enhancement score... Posture change score Continuous activity score And static reduces score Generate single-event behavior response parameters without alarms:

[0141] For the i-th alarmed event in step four (i.e., the first analyte anomaly event), the same four-category feature definitions, observation duration, normalization range, and scoring direction are used to generate the alarm single-event behavior response parameters corresponding to the m-th alarm method actually used in this case:

[0142] Where m represents the actual alarm method used in this case. to The weights are non-negative and their sum is 1.

[0143] In some embodiments of this application, to All values ​​are set to 0.25. In other implementations, the weights can also be determined based on historical annotation data, verification results, device type, or wearing position. It should be noted that when a certain type of feature is unavailable, its weight is set to 0, and the weights of the remaining available features are normalized to sum to 1.

[0144] The above steps employ a consistent scoring rule for non-alarm events and alarm events with different alarm methods to maintain the comparability of single-event behavior response parameters. When all four scores are 0, the corresponding event can be marked as no response, and the corresponding single-event behavior response parameter is 0. When the observation window or triaxial acceleration data does not meet the preset validity requirements, the corresponding event is marked as unevaluable, and no single-event behavior response parameter is generated.

[0145] Furthermore, the single-event behavior response parameters are categorized into the non-alarm event set according to their alarm status. Or the set of alarmed events corresponding to the m-th actual alarm method It should be noted that events that are excluded or marked as unevaluable are not included in the above set.

[0146] right The arithmetic mean of the single-event behavior response parameters is used to obtain the no-alarm behavior response capability parameter (i.e., the second behavior response parameter). : .

[0147] right The arithmetic mean of the single-event behavior response parameters is used to obtain the historical behavior response parameters (i.e., the first behavior response parameters) corresponding to the m-th alarm method. : .

[0148] Furthermore, the response parameters of a single event can be time-weighted and averaged based on the time interval between the occurrence and evaluation of each event, so that events that occurred more recently have a greater weight.

[0149] Step S208: Determine the behavioral response capability of the target object based on the behavioral response parameters.

[0150] In step S208 above, behavioral response capability refers to the comprehensive physiological and behavioral level of the target object in a non-motor state (such as a nighttime sleep environment) in response to abnormal events of the analyte (such as hypoglycemia), including perception, arousal, and action. It is used to determine whether the patient is at risk of impaired perception and whether the current alarm configuration is sufficient to ensure their safety.

[0151] In some embodiments of this application, the behavioral response capability of a target object can be determined by: determining the confidence level of the behavioral response parameters, wherein the confidence level is used to quantify the data quality of generating the behavioral response parameters; when the confidence level is greater than or equal to a preset confidence level, comparing the behavioral response parameters with a preset threshold to obtain a comparison result; and determining the behavioral response capability of the target object based on the comparison result.

[0152] It should be noted that confidence level refers to a numerical indicator used to quantify the reliability of the assessment results, calculated based on the amount of valid data (e.g., the number of valid events and the number of valid nights) upon which the generated behavioral response parameters depend. This prevents statistical bias caused by insufficient sample size (e.g., only one instance of hypoglycemia or only one night of monitoring). Specifically, confidence level can be determined as follows: determine the first number of analyte abnormal events in historical monitoring data, and the second number of different independent monitoring periods in which the analyte abnormal events are distributed; determine the confidence level based on the first number and its corresponding first preset number, and the second number and its corresponding second preset number.

[0153] For example, the confidence level (i.e., the number of valid events) is generated based on the number of valid events and the number of valid nights. Let the number of valid events for the g-th type of event be... The number of effective nights is The preset number of reference events is The preset reference number of nights is The reliability parameter of this type of evaluation result It can be represented as:

[0154]

[0155] Where g=0 represents a non-alarm event, and g=m represents an alarm event corresponding to the m-th alarm method; The value ranges from 0 to 1; a larger value indicates more sufficient valid events and valid nights for generating the corresponding long-term parameters. It should be noted that... and It can be determined based on application requirements, historical data, or verification results.

[0156] Furthermore, let the preset confidence threshold be... When the corresponding confidence parameter is lower than When the evidence is insufficient, no corresponding evaluation report is generated; only a "insufficient evidence" flag is retained internally. When the corresponding confidence parameter reaches or exceeds [a certain threshold], [the system will not generate a corresponding evaluation report]. When, output or corresponding .

[0157] In some embodiments of this application, the following steps may also be performed: when the analyte anomaly event is a first analyte anomaly event that has triggered an alarm and the comparison result indicates that the behavior response parameter is less than a first preset threshold, the alarm type corresponding to the first analyte anomaly event is enhanced; when the analyte anomaly event is a second analyte anomaly event that has not triggered an alarm and the comparison result indicates that the behavior response parameter is less than a second preset threshold, alarm type recommendation information corresponding to the target object is generated.

[0158] To facilitate understanding of the process of determining the above-mentioned behavioral response capabilities, the following explanation is provided in conjunction with some specific embodiments.

[0159] (1) For analyte anomalies that do not trigger alarms, when At that time, according to Compared with the preset level threshold and Relationship generation for behavioral response ability levels related to nocturnal hypoglycemia events:

[0160] when < If the alarm is determined to be at a low response level, further preliminary recommendations for enhanced alarms can be generated, including higher alarm intensity, a combination of sound and vibration, a shorter repetition interval, or enabling guardian sharing.

[0161] when ≤ < If the response level is determined to be medium, further preliminary recommendations can be generated using a single sound or vibration at standard repetition intervals.

[0162] when ≥ If the response level is determined to be high, a preliminary recommendation for using a preset standard alarm method can be generated.

[0163] (2) For analyte anomalies that have triggered alarms, each This is used to represent the patient's historical behavioral responses to different alarm methods. When the current alarm method is... and At that time, With preset alarm response threshold Compare and generate alarm recommendation information:

[0164] when < In such cases, it is recommended to enhance the current alarm method or switch to another alarm method that meets the reliability requirements and has higher historical behavior response parameters. It should be noted that the enhancement may include increasing the alarm intensity, adding a combination of sound and vibration, shortening the repetition interval, or enabling guardian sharing.

[0165] when ≥ If the current alarm method has sufficient historical behavioral responses, it is recommended to maintain the current alarm method.

[0166] It should be noted that the above-mentioned level thresholds and It can be determined based on historical data, verification results, or application requirements.

[0167] Through steps S202 to S208, historical monitoring data of the target object in a non-motor state is acquired. Abnormal events of the analyte are determined based on the analyte data and acceleration data in the monitoring data. The behavioral response parameters of the target object to the abnormal events of the analyte are determined to characterize the degree to which the response behavior deviates from the behavioral baseline in the non-motor state. Then, the behavioral response ability of the target object is determined based on the behavioral response parameters. This achieves the goal of objectively quantifying the behavioral response level of the target object to the abnormal events of the analyte in a natural state without artificial induction. This achieves the technical effects of reducing testing costs and risks, improving long-term assessment efficiency, and providing a basis for personalized intervention. It also solves the technical problem of related technologies that use clamp tests and other methods to observe the patient's response to the abnormal events of the analyte under controlled conditions, which is complex and inefficient.

[0168] The embodiments of this application can be applied, for example, to the assessment of behavioral response capabilities related to nocturnal hypoglycemia events in natural home sleep environments. By distinguishing and evaluating behavioral responses under conditions without actual alarms and behavioral responses after alarms under different alarm methods, corresponding response capability parameters and parameter confidence levels are generated. Response capability parameters can be used to assist in evaluating patients' ability to perceive nocturnal hypoglycemia and provide objective data for subsequent hypoglycemia risk management and personalized alarm method recommendations.

[0169] This application uses independent monitoring cycles (such as natural sleep nights) under multiple non-motor states of the same object as the observation cycle, mapping continuous analyte monitoring data, triaxial acceleration data, and alarm records to a unified time axis; it performs credibility processing on candidate analyte abnormal events, excluding suspected pressure-induced sensor attenuation events, and selecting credible analyte abnormal events; it establishes an individual normal behavior baseline (such as an individual sleep behavior baseline) using target data (such as data corresponding to normal sleep segments); based on whether each credible analyte abnormal event actually outputs an alarm, it determines the observation window for no-alarm behavior response and the observation window for behavior response after an alarm, and calculates the behavior response parameters for no-alarm single events and the behavior response parameters for alarm single events corresponding to the alarm method; it generates no-alarm behavior response capability parameters, historical behavior response parameters corresponding to different alarm methods, and their credibility based on multiple valid events, thereby assisting in evaluating the patient's behavior response capability related to analyte abnormal events and providing objective data basis for personalized alarm method recommendations.

[0170] The beneficial effects of the embodiments of this application are as follows:

[0171] (1) Provide objective basis for personalized prompting strategy recommendation: The non-alarm behavior response capability parameter characterizes the patient's behavioral response level to abnormal events of the analyte in a non-motor state under the condition of no actual alarm, and evaluates the patient's response to each alarm mode by using the historical behavioral response parameters corresponding to different alarm modes, thereby providing objective data basis for personalized alarm configuration and the maintenance, enhancement or replacement of the current alarm mode.

[0172] (2) Achieve objective long-term evaluation in a natural home environment: Generate behavioral response parameters by using actual abnormal events and corresponding behavioral data of the analyte that occur during multiple non-motor states (such as sleep) without actively inducing abnormalities in the analyte, thus reducing reliance on the subjective memory and medical history of the target subject.

[0173] (3) Reduce the interference of abnormally low values ​​and normal non-motor state actions: eliminate sensor signal attenuation events suspected to be induced by external pressure through event credibility processing, and reduce the impact of natural non-motor state body movements on behavioral response evaluation by using the individual's normal non-motor state behavior baseline (such as the individual's normal sleep behavior baseline).

[0174] (4) Avoid prompt-induced behavior leading to incorrect evaluation: Set up separate observation windows for non-alarm behavior response and post-alarm behavior response to avoid mistaking behavior changes induced by sound, vibration or screen prompts as autonomous behavior responses under non-alarm conditions.

[0175] (5) Provide credible behavioral response capability parameters: Generate long-term behavioral response parameters based on multiple valid events, and generate credibility parameters based on the number of valid events and the number of valid independent monitoring periods (such as the number of valid events and the number of valid nights); do not generate evaluation reports when credibility is insufficient, so as to avoid forming low-reliability evaluation results due to insufficient observation data.

[0176] Figure 3 This is a structural diagram of a behavioral response capability assessment device according to an embodiment of this application, as shown below. Figure 3 As shown, the device includes:

[0177] The acquisition module 302 is used to acquire historical monitoring data of the target object in a non-motion state, wherein the historical monitoring data includes analytical data and acceleration data collected by the continuous analytical monitoring device;

[0178] The screening module 304 is used to determine abnormal events of the analyte based on the analyte data and acceleration data.

[0179] The determination module 306 is used to determine the behavioral response parameters of the target object to the analyte abnormal event, wherein the behavioral response parameters are used to characterize the degree to which the response behavior of the target object deviates from the behavioral baseline of the target object in a non-moving state when the analyte abnormal event occurs;

[0180] Evaluation module 308 is used to determine the behavioral response capability of the target object based on behavioral response parameters.

[0181] It should be noted that, Figure 3 The behavioral response capability assessment device shown is used to perform Figure 2 The method for assessing behavioral response capabilities shown is therefore Figure 2 The relevant explanations in the assessment methods for behavioral response capabilities also apply to... Figure 3 The device for assessing behavioral response capabilities shown is not described in detail here.

[0182] This application also provides an electronic device, which includes a memory and a processor, wherein the memory is used to store program instructions; the processor is connected to the memory and is used to execute the steps of the evaluation method for behavioral response capability implemented in the various embodiments of this application.

[0183] This application also provides a non-volatile storage medium including a stored computer program, wherein the device containing the non-volatile storage medium executes the steps of the behavior response capability evaluation method in various embodiments of this application by running the computer program.

[0184] This application also provides a computer program product, including computer instructions that, when executed by a processor, implement the steps of the behavior response capability evaluation method in various embodiments of this application.

[0185] This application also provides a computer program that, when executed by a processor, implements the steps of the behavior response capability evaluation method in various embodiments of this application.

[0186] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0187] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0188] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0189] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0190] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0191] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0192] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for assessing behavioral response capability, characterized in that, include: Acquire historical monitoring data of the target object in a non-motion state, wherein the historical monitoring data includes analyte data and acceleration data collected by the continuous analyte monitoring device; Based on the analyte data and the acceleration data, anomaly events related to the analyte are determined; Determine the behavioral response parameters of the target object to the analyte anomalous event, wherein the behavioral response parameters are used to characterize the degree to which the response behavior of the target object deviates from the behavioral baseline of the target object in the non-motion state when the analyte anomalous event occurs; The behavioral response capability of the target object is determined based on the behavioral response parameters.

2. The method according to claim 1, characterized in that, Determining the behavioral response parameters of the target object to the anomalous events of the analyte includes: Target data is determined from the historical monitoring data, wherein the target data is used to characterize the behavioral features of the target object when it is in the non-motion state and no abnormal event of the analyte has occurred; Based on the target data, a behavioral baseline of the target object in the non-motion state is determined, wherein the behavioral baseline is used to characterize the normal behavioral pattern of the target object in the non-motion state; Determine the response behavior characteristics of the target object when the abnormal event of the analyte occurs; The behavioral response parameters are determined based on the behavioral baseline and the behavioral characteristics.

3. The method according to claim 2, characterized in that, Determining the behavioral baseline of the target object in the non-motion state based on the target data includes: Based on the target data, multiple target behavioral features of the target object are determined, wherein the multiple target behavioral features are used together to characterize the normal behavior of the target object in the non-motion state; Determine the baseline parameters corresponding to each of the target behavioral features, wherein the baseline parameters are used to characterize the central trend and normal fluctuation range of the target object's behavior in the non-motion state; The baseline parameters corresponding to the multiple target behavioral features are collectively used as the behavioral baseline.

4. The method according to claim 1, characterized in that, The historical monitoring data also includes alarm records triggered by the continuous analyte monitoring equipment; Determining the behavioral response parameters of the target object to the anomalous events of the analyte includes: Based on the alarm records, determine whether the abnormal event of the analyte triggered an alarm, and obtain the judgment result; Based on the judgment result, the observation time window corresponding to the analyte abnormal event is determined respectively. The first observation time window is used when the analyte abnormal event triggers an alarm, and the second observation time window is used when the analyte abnormal event does not trigger an alarm. The first observation time window starts from the time when the first alarm is output corresponding to the analyte abnormal event, and the second observation time window starts from the time when the analyte abnormal event is triggered. Within the observation time window, determine the behavioral response parameters of the target object to the abnormal events of the analyte.

5. The method according to claim 4, characterized in that, Within the observation time window, determine the behavioral response parameters of the target object to the anomalous events of the analyte, including: Within the observation time window, multiple response behavior characteristics of the target object to the abnormal event of the analyte are determined, wherein the multiple response behavior characteristics are used together to characterize the degree of physiological or behavioral response of the target object in multiple dimensions; Multiple behavioral response scores are determined based on the degree of deviation between the multiple response behavioral characteristics and the corresponding behavioral baselines, wherein the multiple behavioral response scores are used together to characterize the degree of perception of the target object of the abnormal event of the analyte; The behavioral response parameters are determined based on the multiple behavioral response scores.

6. The method according to claim 5, characterized in that, The analyte anomaly events include a first analyte anomaly event that has triggered an alarm and a second analyte anomaly event that has not triggered an alarm. The behavioral response parameters are determined based on the multiple behavioral response scores, including: Obtain at least one first behavioral response score set corresponding to the first analyte abnormal event, and determine the first behavioral response parameter of the alarm type corresponding to the first analyte abnormal event based on the first behavioral response score set; Obtain at least one second behavioral response score set corresponding to an abnormal event of the second analyte, and determine the second behavioral response parameter corresponding to the abnormal event of the second analyte based on the second behavioral response score set.

7. The method according to claim 5, characterized in that, Multiple behavioral response scores are determined based on the degree of deviation between the aforementioned multiple behavioral characteristics and the corresponding behavioral baseline, including: Determine the degree of deviation between each of the plurality of response behavior features and the corresponding target behavior feature in the behavior baseline of the target object; Obtain the scale parameters corresponding to the multiple response behavior features respectively, wherein the scale parameters are determined by the preset high quantile of the standardized deviation of the normal behavior data of the target object in the non-motion state; The scores of the multiple behavioral responses are determined based on the deviation values ​​and scale parameters corresponding to the multiple behavioral characteristics.

8. The method according to claim 1, characterized in that, Determining the behavioral response capability of the target object based on the behavioral response parameters includes: Determine the confidence level of the behavioral response parameters, wherein the confidence level is used to quantify the data quality that generates the behavioral response parameters; If the confidence level is greater than or equal to the preset confidence level, the behavioral response parameter is compared with the preset threshold to obtain the comparison result; The behavioral response capability of the target object is determined based on the comparison results.

9. The method according to claim 8, characterized in that, The method further includes: If the analyte anomaly event is a first analyte anomaly event that has triggered an alarm and the comparison result indicates that the behavior response parameter is less than a first preset threshold, the alarm type corresponding to the first analyte anomaly event is enhanced. If the analyte anomaly event is a second analyte anomaly event that does not trigger an alarm and the comparison result indicates that the behavior response parameter is less than a second preset threshold, alarm type recommendation information corresponding to the target object is generated.

10. An electronic device, characterized in that, include: A memory and a processor, the memory being used to store program instructions; the processor being connected to the memory and used to execute an evaluation method for implementing the behavioral responsiveness of any one of claims 1 to 9.