Physiological feature signal processing method and system based on fourth-order kinetic model

Abstract

According to the physiological feature signal processing method and system based on the fourth-order kinetic model, the fourth-order kinetic model is constructed for the collected physiological signals of the target individual based on the fourth-order three-point boundary value problem, various boundary conditions can be synchronously met, it is guaranteed that model output conforms to the physiology principle from the mathematical level, and the physiological feature signal processing accuracy is improved. Results which do not conform to actual physiological conditions are avoided; by establishing the mapping relation between the patient physiological turning time and the model boundary point parameters, modeling is performed for the unique physiological rhythm of each target individual, the integrating degree of the fourth-order kinetic model and the actual physiological state of the target individual is improved, and a more accurate basis is provided for personalized medical diagnosis; the physiological consistency of multi-modal signal output is ensured in combination with cone constraint, and accurate synchronization of different modal signals in a time domain is realized.

Description

Technical Field

[0001] This invention belongs to the field of physiological signal processing technology, and in particular relates to a physiological feature signal processing method and system based on a fourth-order dynamic model. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] As people pay more attention to their health, electronic devices with physiological measurement functions are becoming increasingly popular. For example, wearable devices can measure vital signs such as heart rate, blood oxygen, exercise, and sleep by collecting physiological signals from the human body.

[0004] Currently, physiological signals acquired in real time by wearable medical devices are modeled using traditional third-order linear / nonlinear dynamic models. However, these models can only describe the basic trends of physiological indicators and cannot effectively characterize the unique physiological rhythm-specific parameters of individual patients, such as the timing of morning blood glucose peaks in diabetic patients or the duration of morning blood pressure peaks in hypertensive patients. The lack of quantitative modeling capabilities for these individualized physiological characteristics leads to a systematic deviation between the model output and the patient's actual physiological state, making it difficult to support accurate individualized diagnosis and risk assessment. Furthermore, the signal-to-noise ratio (SNR) of physiological signals acquired in real time by wearable medical devices is significantly reduced due to environmental and physiological noise such as motion artifacts, electromyographic interference, and electromagnetic radiation, resulting in deficiencies in model diagnostic robustness. For example, during electrocardiogram (ECG) signal acquisition, motion artifacts generated by patient limb movements can cause QRS complex morphology distortion and ST segment baseline drift. The misdiagnosis rate of cardiovascular diseases caused by this type of noise is as high as 8%–12%, especially leading to missed or misjudged conditions such as myocardial ischemia and arrhythmias.

[0005] Wearable medical devices must simultaneously meet multi-dimensional time-domain boundary constraints throughout the entire physiological signal monitoring cycle. Specific constraints include the initial contact state t=0, key inflection points, etc. Traditional models, particularly those with multivariate constraints terminating at t=1, often suffer from insufficient boundary constraint dimensions, leading to situations where one constraint is satisfied while others are violated, resulting in outputs that do not conform to physiological laws. Furthermore, traditional dynamic models exhibit significant real-time limitations in physiological signal processing due to model complexity and numerical computation efficiency. Specifically, the total delay from receiving raw physiological signals to outputting diagnostic results exceeds 5 seconds, far exceeding the clinical threshold for emergency warnings, missing the optimal intervention window, and failing to meet the clinical application requirements for real-time monitoring and early warning of emergencies. In multimodal physiological signal monitoring scenarios such as ECG+EEG and ECG+PPG combined monitoring, traditional models lack a unified time-domain benchmark and synchronization calibration mechanism, resulting in time-domain asynchrony issues between different modal signals. Specifically, ECG signals reflect cardiac electrical activity with millisecond-level time-domain resolution, while EEG signals reflect brain electrical activity; due to differences in acquisition frequency and signal preprocessing procedures, the time-domain alignment error between EEG and ECG signals can reach hundreds of milliseconds or even seconds. This time-domain asynchrony disrupts the physiological correlation between multimodal signals, leading to feature misalignment during multimodal data fusion and failing to fully realize the collaborative diagnostic value of multimodal signals.

[0006] In summary, the processing of physiological signals monitored in real time by wearable medical devices suffers from at least the following problems: insufficient ability to model individual differences, sensitivity to noise, difficulty in coordinating and satisfying boundary conditions, poor real-time performance, and poor temporal synchronization of multimodal signals. Summary of the Invention

[0007] To overcome the shortcomings of the prior art, this invention provides a physiological feature signal processing method and system based on a fourth-order dynamic model. The fourth-order dynamic model is constructed based on the fourth-order three-point boundary value problem, which can simultaneously satisfy multiple boundary conditions. The model is designed for the unique physiological rhythms of each target individual, improving the fit between the fourth-order dynamic model and the actual physiological state of the target individual, and providing a more accurate basis for personalized medical diagnosis.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a physiological feature signal processing method based on a fourth-order dynamic model, comprising: A fourth-order dynamic model is constructed based on a fourth-order three-point boundary value problem for the physiological signals of the collected target individuals. By fitting and calibrating the boundary point parameters using the historical physiological data of the target individual, the boundary point parameters correspond to the specific physiological turning points of the target individual, and a mapping relationship is established between the physiological turning points of the target individual and the boundary point parameters of the fourth-order dynamic model. The fourth-order dynamic model is solved based on cone constraints, and the multimodal physiological characteristics of the target individual are analyzed based on the solution results.

[0009] Secondly, the present invention provides a physiological feature signal processing system based on a fourth-order dynamic model, comprising: The modeling module is configured to: construct a fourth-order dynamic model based on a fourth-order three-point boundary value problem for the physiological signals of the collected target individuals; The mapping module is configured to: fit and calibrate the boundary point parameters by using the historical physiological data of the target individual, so that the boundary point parameters correspond to the specific physiological turning point of the target individual, and establish the mapping relationship between the physiological turning point of the target individual and the boundary point parameters of the fourth-order dynamic model. The analysis module is configured to: solve the fourth-order dynamic model based on cone constraints, and analyze the multimodal physiological characteristics of the target individual based on the solution results.

[0010] Thirdly, the present invention provides an electronic device including a memory and a processor, and computer instructions stored in the memory and running on the processor, wherein the computer instructions, when executed by the processor, perform the method described in the first aspect.

[0011] Fourthly, the present invention provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, perform the method described in the first aspect.

[0012] Fifthly, the present invention provides a computer program product, including a computer program that, when executed by a processor, implements the method described in the first aspect.

[0013] The above one or more technical solutions have the following beneficial effects: In this invention, a fourth-order dynamic model is constructed based on a fourth-order three-point boundary value problem for the physiological signals of the acquired target individuals. This model can simultaneously satisfy multiple boundary conditions, ensuring from a mathematical perspective that the model output conforms to physiological principles and avoids results that do not conform to actual physiological conditions. By establishing a mapping relationship between the patient's physiological turning points and the parameters of the model's boundary points, the unique physiological rhythms of each target individual are modeled, improving the fit between the fourth-order dynamic model and the actual physiological state of the target individual, and providing a more accurate basis for personalized medical diagnosis. Combined with cone constraints, the physiological consistency of multimodal signal output is ensured, achieving precise synchronization of different modal signals in the time domain.

[0014] This invention is applicable to various scenarios, such as chronic disease management, postoperative rehabilitation, emergency early warning, and mental illness. In chronic disease management, taking diabetes as an example, this solution reduces blood glucose prediction error and significantly improves prediction accuracy compared to traditional LSTM models. This allows doctors to understand patients' conditions more accurately, develop more effective treatment plans, and achieve precise management of chronic diseases. In multimodal signal monitoring scenarios, it can also effectively cope with noise interference, ensure the reliability of each modality signal, and provide stable and accurate data support for comprehensive diagnosis.

[0015] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0016] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0017] Figure 1 This is a flowchart of a physiological feature signal processing method based on a fourth-order dynamic model in Embodiment 1 of the present invention. Detailed Implementation

[0018] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations of the present invention.

[0020] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0021] Example 1 like Figure 1 As shown, this embodiment discloses a physiological feature signal processing method based on a fourth-order dynamic model, including: A fourth-order dynamic model is constructed based on a fourth-order three-point boundary value problem for the physiological signals of the collected target individuals. By fitting and calibrating the boundary point parameters using the historical physiological data of the target individual, the boundary point parameters correspond to the specific physiological turning points of the target individual, and a mapping relationship is established between the physiological turning points of the target individual and the boundary point parameters of the fourth-order dynamic model. The fourth-order dynamic model is solved based on cone constraints, and the multimodal physiological characteristics of the target individual are analyzed based on the solution results.

[0022] This embodiment constructs a fourth-order dynamic model based on a fourth-order three-point boundary value problem for the collected physiological signals of the target individual. This model can simultaneously satisfy multiple boundary conditions, ensuring from a mathematical perspective that the model output conforms to physiological principles and avoids results that do not conform to actual physiological conditions. By establishing a mapping relationship between the patient's physiological turning points and the model's boundary point parameters, the unique physiological rhythm of each target individual is modeled, improving the fit between the fourth-order dynamic model and the actual physiological state of the target individual, and providing a more accurate basis for personalized medical diagnosis. Combined with cone constraints, the physiological consistency of multimodal signal output is ensured, achieving precise synchronization of different modal signals in the time domain.

[0023] In this embodiment, a fourth-order dynamic model is constructed based on a fourth-order three-point boundary value problem for the collected physiological signals of the target individual. The constructed fourth-order dynamic model is as follows: (1) in, continuous, It is a constant. The function to be solved... u ( t () represents dynamic physiological indicators, such as blood glucose concentration; nonlinear source terms f ( t,u () represents the pathological risk diagnostic model; normalized time. t t=0 represents the device startup time, such as the time when the patient falls asleep. t=1 indicates the end time of monitoring; boundary point parameters Indicates a patient's physiological turning point, such as the morning peak blood pressure; zero point of the first derivative. u' (0)=0 indicates that the initial contact signal of the device is stable and the patch jitter has been eliminated; the second derivative is zero. u'' (0)=0 indicates that the initial acceleration of the signal is zero, suppressing motion artifacts; zero point of the third derivative. u''' ( )=0 indicates a critical turning point in the state transition, such as the transition from sleep to wakefulness; the function's final value is zero. u (1)=0 indicates that the monitoring period returns to the physiological baseline, such as resting blood oxygen.

[0024] Example of mapping in existing pathological risk models:

[0025] Pathological risk diagnosis model f (t,u) is a mathematical abstract framework of pathological risk mechanism, which needs to be dynamically constructed according to specific diseases.

[0026] For example, 1. Myocardial infarction risk model:

[0027] Where u represents the ST segment depression amplitude (mm).

[0028] Equipment motion artifact noise 2. Diabetes morning surge model:

[0029] t=0.7: Normalized time point at 6:00 AM, where constants 0.08 and 100 are patient-specific parameters.

[0030] The above example formulas illustrate the modeling methodology; specific parameters require clinical calibration.

[0031] Specific monitoring point constraints. u (1) = 0 At the end of the monitoring cycle, the signal returns to the baseline, such as blood oxygen returning to the resting level.

[0032] physiological turning point .

[0033] In this embodiment, the cone constraint is defined as follows:

[0034] in, Non-negative constraints on physiological indicators, such as blood oxygen saturation. [0,1]; Terminal decay characteristics (in line with physiological recovery period patterns) meet AHA recovery period blood pressure decay standards (Circulation 2023).

[0035] This embodiment uses historical physiological data of the target individual to fit and calibrate boundary point parameters, making the boundary point parameters correspond to the specific physiological turning points of the target individual, and establishing a mapping relationship between the physiological turning points of the target individual and the boundary point parameters of the fourth-order kinetic model. Specifically, the collected 24-hour physiological indicator monitoring sequence of the target individual is substituted into the fourth-order kinetic model, and the specific values ​​of the boundary point parameters are calculated through a data fitting algorithm. The calculated specific values ​​of the boundary point parameters are then substituted into the boundary conditions of the fourth-order kinetic model to verify whether the normalized time corresponding to the boundary point parameters is consistent with the actual physiological turning points of the target individual, thus realizing the mapping between the physiological turning points of the target individual and the boundary point parameters of the fourth-order kinetic model.

[0036] As a specific implementation method, the 24-hour physiological indicator monitoring sequence of the target individual, collected at a sampling frequency of ≥1 time / hour, is substituted into the fourth-order kinetic core model. The indicator types include, but are not limited to, blood pressure, blood glucose, electrocardiogram (ECG), and electroencephalogram (EEG) signals. The specific value of the boundary point parameter η of the model is calculated through a data fitting algorithm. The specific value of the calculated boundary point parameter η is substituted into the specific boundary condition of the fourth-order kinetic model, namely the zero point of the third derivative, to verify whether the normalized time corresponding to the boundary point parameter η is consistent with the actual physiological turning point of the target individual. Finally, the accurate mapping between the physiological turning point of the target individual and the boundary point parameter η of the fourth-order kinetic model is achieved.

[0037] In this embodiment, the fourth-order dynamic model of the fourth-order three-point boundary value problem is transformed into an integral equation through the Green's function, thereby constructing a quantitative relationship between pathological stimuli and physiological responses.

[0038] Construct the Green's function for any Consider the boundary value problem: (2) Then its solution is equivalent to: (3) Green's function G ( t,s ) constructed pathological stimuli ( s Time) and physiological response ( t The quantization relationship at time (time). The function characteristics are shown in Table 1.

[0039] Table 1:

[0040] Proof: The expression for the Green's function in the boundary value problem formula (1) can be obtained through calculation. : (1) t t>s hour, ; (2) t=s hour, ; (3) t When =0, Therefore, B=C=0. ; (4) t When =1, Therefore, E+F+G+H=0. again ,so ; (5) ​ hour, , ,so .

[0041] We can obtain the following in sequence .but:

[0042] when hour, (4) when hour, (5) The solution to the problem is equivalent to the integral equation:

[0043] when hour, ;when hour, .make , but .

[0044] Fixed-point theorem for conical tension and compression: E It is the Banach space. P yes E The cone in the middle, It is a bounded open set in E. For a fully continuous mapping operator, The radius threshold for cone stretching.

[0045] If one of the following conditions is true: (1) ; (2) .

[0046] Then T is There must exist a fixed point in the equation, where T is a continuous and compact nonlinear operator on the Banach space. The fully continuous operator constructed by the Green's function G and the cone constraint P is the core technology for solving the medical boundary optimization problem.

[0047] Conical tension and conical compression conditions:

[0048] Superlinear adaptation: Acute illness (the risk of minor abnormalities in myocardial infarction increases gradually, and explodes exponentially in severe cases).

[0049] Sublinear fit: Chronic diseases (sensitive in the early stages of diabetes, with a slowdown in the rate of increase of high values ​​in the long term).

[0050] Lemma 1 Suppose and , and It is the only solution, in addition satisfy ,here . ( critical period Physiological indicators (such as during critical periods at night) should not be lower than the overall amplitude. This forcefully maintains the lower limit of ECG / EEG signal synchronization. This is the critical monitoring period; To ensure the fidelity of physiological indicators, such as Ensure that the minimum blood oxygen level is ≥ 30% of the peak level.

[0051] Proof: As can be seen from the preceding text, when hour, ,thereby exist If the top is convex, then:

[0052] have to:

[0053] So Similarly, when When the time comes, the conclusion holds true.

[0054] Theorem 1 Assume It is continuous and satisfies the following conditions: For each , mapping It is decreasing; For each , mapping It is increasing. Therefore, we can conclude that: (1) Existence of positive solution: The system has a nontrivial solution under the superlinear / sublinear condition. ; (2) Individualized calibration: through Implement key parameter mapping: Cardiovascular patients experience peak risk at 6:00 AM. (3) Robustness guarantee: Cone constraint ensures (Physiological feasibility).

[0055] Proof: Let ,like If is a fixed point of the operator equation, then It must be a solution to problem (1). Let Let P be a cone on E, and let the function on P be positive. It is completely continuous. Therefore, A is a completely continuous operator.

[0056] For the superlinear case: by It can be concluded that: it exists. , hour, .

[0057] in, This is a linearly increasing control parameter.

[0058] set up ,promise ,make ,but: , Then there is .

[0059] Secondly by It can be concluded that: it exists. ,when hour, .

[0060] make ,set up ,promise ,make ,but In the interval superior Combining f According to nonnegative continuity, for any have: .

[0061] when hour,

[0062] but , Satisfying the theorem, A is in There is a fixed point on the surface, such that... .

[0063] For the sublinear case: by It can be concluded that: it exists. ,when hour, , .set up ,like ,make ,but:

[0064] so .

[0065] Secondly by It can be concluded that: it exists. ,when , .

[0066] like Bounded, that is , have In this case, take ,when ,have:

[0067] so .like Unbounded, fixed variables ,but It is about A continuous function in one variable. Take... ,have hour, ,when ,have:

[0068] In conclusion, the boundary value problem has a correct solution.

[0069] Taking blood pressure as an example, consider the pathokinetic model corresponding to a fourth-order three-point boundary value problem:

[0070] Among them, dynamic blood pressure values It can represent the change in systolic blood pressure over time; normal range: 90–140 mmHg; sympathetic nerve excitability coefficient. k> 0, the higher the value, the higher the risk of a sudden rise in blood pressure, such as k =1.2 corresponds to hypertensive patients; diurnal rhythm amplitude c> 0, Morning ( t =0.7) and nighttime ( t =0.3) Blood pressure fluctuation range (typical clinical value: c =15); zero point of third derivative Individualized turning point: corresponding to the peak blood pressure time at 6:00 AM; calibration basis: patient's historical data; cyclical rhythm. 24-hour blood pressure diurnal rhythm (in line with medical consensus).

[0071] Boundary condition medical mapping: Wearable devices during the morning peak hours ( The system detects that the acceleration of blood pressure change has reached zero, triggering an early warning signal. At the end of the monitoring period, blood pressure returns to the resting baseline (e.g., sleep diastolic blood pressure ≈ 70 mmHg).

[0072] Personalized parameter calibration demonstration: Patient data fitting: Input to the fourth-order kinetic model: a 24-hour blood pressure monitoring sequence sampled once per hour; output of the fourth-order kinetic model: →Corresponds to the blood pressure turning point at 6:12 AM.

[0073] Verification of computational and medical rationale: Construction of the above solution (clinical safety threshold): Let... =140 (the upper limit of clinical systolic blood pressure), then verification is required:

[0074] When k=1.2 and c=15, the right end 0.07>0 → Satisfies the above solution conditions.

[0075] Lower solution construction (physiological minimum value): Let (If blood pressure fluctuates within a healthy range), then:

[0076]

[0077] At t=0.25 (nighttime trough), the left end →The following solution is valid.

[0078] This embodiment can be achieved through... Calibrate patient-specific risk moments for personalized diagnosis; superlinear conditions (increased k) correspond to acute hypertensive crises, while sublinear conditions (decreased k) correspond to chronic regulation, enabling dynamic risk quantification; the Green's function G(t,s) can be embedded in medical chips for real-time calculation. The error is <0.1 mmHg.

[0079] The fourth-order dynamic model constructed in this embodiment has noise resistance and improves the ability to suppress high-frequency disturbances; it also has boundary adaptation, which supports the simultaneous satisfaction of multiple constraints of medical devices.

[0080] Example 2 The purpose of this embodiment is to provide a physiological feature signal processing system based on a fourth-order dynamic model, including: The modeling module is configured to: construct a fourth-order dynamic model based on a fourth-order three-point boundary value problem for the physiological signals of the collected target individuals; The mapping module is configured to: fit and calibrate the boundary point parameters by using the historical physiological data of the target individual, so that the boundary point parameters correspond to the specific physiological turning point of the target individual, and establish the mapping relationship between the physiological turning point of the target individual and the boundary point parameters of the fourth-order dynamic model. The analysis module is configured to: solve the fourth-order dynamic model based on cone constraints, and analyze the multimodal physiological characteristics of the target individual based on the solution results.

[0081] In further embodiments, the following is also provided: An electronic device includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor. When executed by the processor, the computer instructions perform the method described in Embodiment 1. For brevity, further details are omitted here.

[0082] It should be understood that in this embodiment, the processor can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0083] Memory may include read-only memory and random access memory, and provides instructions and data to the processor. A portion of memory may also include non-volatile random access memory. For example, memory may also store information about the device type.

[0084] A computer-readable storage medium for storing computer instructions, which, when executed by a processor, perform the method described in Embodiment 1.

[0085] The method in Embodiment 1 can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor. The software modules can reside in readily available storage media in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory; the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, a detailed description is not provided here.

[0086] A computer program product includes a computer program that, when executed by a processor, implements the method described in Embodiment 1.

[0087] The present invention also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions included in program modules, which execute in a device on a target real or virtual processor to perform the processes / methods described above. Typically, program modules include routines, programs, libraries, objects, classes, components, data structures, etc., that perform specific tasks or implement specific abstract data types. In various embodiments, the functionality of program modules can be combined or divided among program modules as needed. The machine-executable instructions for the program modules can execute within a local or distributed device. In a distributed device, the program modules can reside in both local and remote storage media.

[0088] The computer program code used to implement the methods of the present invention may be written in one or more programming languages. This computer program code may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the computer or other programmable data processing device, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a computer, partially on a computer, as a stand-alone software package, partially on a computer and partially on a remote computer, or entirely on a remote computer or server.

[0089] In the context of this invention, computer program code or related data may be carried by any suitable carrier to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carriers include signals, computer-readable media, and the like. Examples of signals may include electrical, optical, radio, sound, or other forms of propagation signals, such as carrier waves, infrared signals, etc.

[0090] Those skilled in the art will recognize that the units and algorithm steps described in conjunction with the embodiments herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0091] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A physiological feature signal processing method based on a fourth-order dynamic model, characterized in that, include: A fourth-order dynamic model is constructed based on a fourth-order three-point boundary value problem for the physiological signals of the collected target individuals. By fitting and calibrating the boundary point parameters using the historical physiological data of the target individual, the boundary point parameters correspond to the specific physiological turning points of the target individual, and a mapping relationship is established between the physiological turning points of the target individual and the boundary point parameters of the fourth-order dynamic model. The fourth-order dynamic model is solved based on cone constraints, and the multimodal physiological characteristics of the target individual are analyzed based on the solution results.

2. The physiological feature signal processing method based on a fourth-order dynamic model as described in claim 1, characterized in that, Based on the collected physiological signals of the target individuals, a fourth-order dynamic model is constructed using a fourth-order three-point boundary value problem. The fourth-order dynamic model is specifically as follows: ； in, As dynamic physiological indicators, f ( t,u This is a pathological risk diagnosis model. t t=0 represents the start time of the monitoring equipment, and t=1 represents the end time of the monitoring. For boundary point parameters, u' (0)=0 indicates that the initial contact signal of the device is stable. u'' (0)=0 indicates that the initial acceleration of the signal is zero. u''' ( )=0 indicates a key turning point in the state transition. u (1)=0 indicates that the monitoring period returns to the physiological baseline.

3. The physiological feature signal processing method based on a fourth-order dynamic model as described in claim 1, characterized in that, Also includes: By using Green's function, the fourth-order dynamic model of the fourth-order three-point boundary value problem is transformed into an integral equation, thus constructing a quantitative relationship between pathological stimuli and physiological responses.

4. The physiological feature signal processing method based on a fourth-order dynamic model as described in claim 1, characterized in that, By fitting and calibrating boundary point parameters using historical physiological data of the target individual, the boundary point parameters are made to correspond to the specific physiological turning points of the target individual. A mapping relationship is established between the physiological turning points of the target individual and the boundary point parameters of the fourth-order kinetic model, specifically as follows: The 24-hour physiological index monitoring sequence of the target individual was substituted into the fourth-order dynamic model, and the specific values ​​of the boundary point parameters were calculated through the data fitting algorithm. The calculated boundary point parameters are substituted into the boundary conditions of the fourth-order dynamic model to verify whether the normalized time corresponding to the boundary point parameters is consistent with the actual physiological turning point of the target individual, thus realizing the mapping between the physiological turning point of the target individual and the boundary point parameters of the fourth-order dynamic model.

5. The physiological feature signal processing method based on a fourth-order dynamic model as described in claim 1, characterized in that, The cone constraint is specifically as follows: in, This indicates a non-negativity constraint on physiological indicators; Indicates terminal decay characteristics. E It is the Banach space.

6. The physiological feature signal processing method based on a fourth-order dynamic model as described in claim 1, characterized in that, Based on the solution results, the multimodal physiological characteristics of the target individual are analyzed. Specifically, based on the solution results of the fourth-order dynamic model, the multimodal physiological characteristics of the target individual are analyzed in combination with superlinear and sublinear analysis, and a diagnosis and early warning are given when the physiological indicators exceed the physiological safety threshold.

7. A physiological feature signal processing system based on a fourth-order dynamic model, characterized in that, include: The modeling module is configured to: construct a fourth-order dynamic model based on a fourth-order three-point boundary value problem for the physiological signals of the collected target individuals; The mapping module is configured to: fit and calibrate the boundary point parameters by using the historical physiological data of the target individual, so that the boundary point parameters correspond to the specific physiological turning point of the target individual, and establish the mapping relationship between the physiological turning point of the target individual and the boundary point parameters of the fourth-order dynamic model. The analysis module is configured to: solve the fourth-order dynamic model based on cone constraints, and analyze the multimodal physiological characteristics of the target individual based on the solution results.

8. An electronic device, characterized in that, It includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, which, when executed by the processor, perform the method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, Used to store computer instructions, which, when executed by a processor, perform the method described in any one of claims 1-6.

10. A computer program product, characterized in that, Includes a computer program, which, when executed by a processor, implements the method described in any one of claims 1-6.

Images