Human defecation prediction method, device and electronic device based on logistic regression

Abstract

The present application relates to a method, device and electronic device for predicting human bowel movements based on logistic regression, and belongs to the field of medical care technology. The method of the present application includes obtaining human pulse rate data to be predicted; inputting the human pulse rate data into a pre-constructed bowel movement prediction algorithm model, and outputting the prediction results of the bowel movement prediction algorithm model; wherein the bowel movement prediction algorithm model is pre-constructed based on logistic regression. The technical solution of the present application, according to the pre-constructed bowel movement prediction algorithm model, predicts bowel movements based on human pulse rate data, which can easily realize the prediction of human bowel movements, thereby facilitating the improvement of nursing efficiency in relevant scenarios and reducing the psychological burden of the person being cared for.

Description

Technical Field

[0001] The present application belongs to the field of medical care technology, and specifically relates to a method, device and electronic equipment for predicting human defecation and urination based on logistic regression. Background Art

[0002] In the related technology, there are intelligent defecation and urination care devices for the needs of defecation and urination care. However, the intelligent defecation and urination care devices currently on the market are only convenient for the elderly to handle their defecation, which will still increase the care burden of relevant medical staff and reduce the care efficiency. If defecation and urination warnings can be achieved and applied to intelligent defecation and urination care devices, it will greatly improve the care efficiency, reduce the care burden of medical staff and other personnel, and at the same time play a role in maintaining the self-esteem of the elderly and reducing the psychological burden of the elderly.

[0003] The above contents are only used to assist in understanding the technical solution of the present invention and do not constitute an admission that the above contents are prior art. Summary of the invention

[0004] In order to overcome the problems existing in the related technology at least to a certain extent, the present application provides a method for predicting human defecation and urination based on logistic regression, which facilitates the prediction of human defecation and urination, thereby helping to improve nursing efficiency in related scenarios and reduce the psychological burden of the cared for personnel.

[0005] To achieve the above objectives, this application adopts the following technical solutions:

[0006] First,

[0007] The present application provides a method for predicting human defecation based on logistic regression, the method comprising:

[0008] Acquiring human pulse rate data to be predicted;

[0009] Inputting the human pulse rate data into a pre-built defecation and urination prediction algorithm model, and outputting the prediction results of the defecation and urination prediction algorithm model;

[0010] Among them, the defecation prediction algorithm model is pre-constructed based on logistic regression.

[0011] Optionally, the pre-construction of the defecation prediction algorithm model based on logistic regression includes:

[0012] Collect human pulse rate parameter data;

[0013] Preprocessing the pulse rate parameter data to obtain a pulse rate parameter data sample set;

[0014] According to the pulse rate parameter data sample set, an initial model of the defecation and urination prediction algorithm is constructed based on logistic regression;

[0015] The initial model of the two-feces prediction algorithm is tuned and tested to obtain the two-feces prediction algorithm model.

[0016] Optionally, the collecting of human pulse rate parameter data includes:

[0017] The pulse rate data of the subjects are collected within a preset period of time in a day, and the time periods when defecation and urination events occur, the time periods when the subjects lie still, and the time periods when sudden events that cause abnormal pulse values ​​occur are marked in the pulse rate data.

[0018] Optionally, pulse rate data is collected through a medical-grade pulse rate watch worn by the subject; the preset time period is at least 9 hours.

[0019] Optionally, the preprocessing of the pulse rate parameter data comprises performing the following processing steps on the pulse rate parameter data of each individual subject:

[0020] Screening the pulse rate parameter data of the subject, and cleaning the data of abnormal pulse rate values ​​caused by emergencies based on the annotation information;

[0021] The screened pulse rate parameter data are analyzed and processed, and the pulse rate values ​​in the defecation and urination and lying still process periods are averaged based on the annotation information, and the difference between the two average values ​​is used as the pulse rate data sample when the human body has defecation and urination behavior, and the data in the lying still process period is divided into two sections, the average values ​​of the two sections are taken, and the difference between the two average values ​​is taken as the pulse rate data sample when the human body has not had defecation and urination behavior.

[0022] Optionally, the constructing an initial model of a defecation prediction algorithm based on logistic regression according to the pulse rate parameter data sample set includes:

[0023] Dividing the pulse rate parameter data sample set based on a preset method to form a training sample set and a test sample set;

[0024] According to the training sample set, the maximum likelihood method is used to solve the model parameters in the constructed model hypothesis function, and the initial model of the defecation prediction algorithm is determined according to the obtained model parameters.

[0025] Optionally, the pulse rate parameter data sample set is divided and processed based on a preset method, specifically:

[0026] In the pulse rate parameter data sample set, the pulse rate data samples of the human body having defecation and urination behavior and the pulse rate data samples of the human body not having defecation and urination behavior are divided into a quantity ratio of 8:2 respectively;

[0027] Then, in the division results, the two samples with a proportion of 8 are combined to form a training sample set, and the two samples with a proportion of 2 are combined to form a test sample set.

[0028] Optionally, the initial model of the two-feces prediction algorithm is tuned and tested to obtain the two-feces prediction algorithm model, including:

[0029] The initial model of the two-feces prediction algorithm is trained according to the training sample set, and the learning rate parameter in the model is adjusted based on adaptive learning to achieve model tuning;

[0030] The tuned model is tested and verified according to the test sample set, and when the prediction accuracy of the model is greater than a preset threshold value, the model is used as the binary prediction algorithm model.

[0031] Second,

[0032] The present application provides a human defecation prediction device based on logistic regression, the device comprising:

[0033] An acquisition module is used to acquire human pulse rate data to be predicted;

[0034] A prediction processing module, used for inputting the human pulse rate data into a pre-built two-feces prediction algorithm model, and outputting the prediction results of the two-feces prediction algorithm model;

[0035] Among them, the defecation prediction algorithm model is pre-constructed based on logistic regression.

[0036] Thirdly,

[0037] The present application provides an electronic device, including:

[0038] A memory having an executable program stored therein;

[0039] A processor is used to execute the executable program in the memory to implement the steps of the above method.

[0040] This application adopts the above technical solution, which has at least the following beneficial effects:

[0041] The technical solution of the present application, according to a pre-constructed defecation and urination prediction algorithm model, performs defecation and urination prediction based on human pulse rate data, which can easily realize the prediction of human defecation and urination, thereby helping to improve nursing efficiency in related scenarios and reduce the psychological burden of the cared for personnel.

[0042] Other advantages, objectives, and features of the present invention will be described in the following description to some extent, and will be apparent to those skilled in the art based on the following examination and study, or can be taught from the practice of the present invention to some extent. The objectives and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the following text and the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings are used to provide a further understanding of the technical solution of the present application or the prior art, and constitute a part of the specification. Among them, the accompanying drawings expressing the embodiments of the present application are used together with the embodiments of the present application to explain the technical solution of the present application, but do not constitute a limitation on the technical solution of the present application.

[0044] Figure 1 This is a flow chart of a method for predicting human defecation and urination based on logistic regression in one embodiment of the present application;

[0045] Figure 2 This is a schematic diagram illustrating the application of a two-step prediction algorithm model in one embodiment of the present application;

[0046] Figure 3 This is a schematic diagram illustrating the data collection process in one embodiment of the present application;

[0047] Figure 4 This is a schematic diagram illustrating preprocessing of pulse rate parameter data in one embodiment of the present application;

[0048] Figure 5 A schematic diagram of the structure of a human defecation prediction device based on logistic regression provided in one embodiment of the present application;

[0049] Figure 6 A schematic diagram of the structure of an electronic device provided for one embodiment of the present application. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be described in detail below. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other implementation methods obtained by ordinary technicians in the field without creative work belong to the scope of protection of the present application.

[0051] As mentioned in the background technology, the smart defecation and urination care equipment currently available on the market only facilitates the handling of the elderly after defecation, which will still increase the nursing burden of relevant medical staff and reduce the nursing efficiency. If the defecation and urination warning can be realized and applied to the smart defecation and urination care equipment, it will greatly improve the nursing efficiency, reduce the nursing burden of medical staff and other personnel, and at the same time play a role in maintaining the self-esteem of the elderly and reducing the psychological burden of the elderly.

[0052] Based on relevant research findings and the applicant's analysis of relevant collected data (see below for details), a person's defecation and urination behavior is related to a person's pulse rate characteristics. Based on this, this application proposes a method for predicting human defecation and urination based on logistic regression to meet the actual needs for defecation and urination prediction.

[0053] like Figure 1 As shown, in one embodiment, the human defecation prediction method based on logistic regression proposed in this application includes:

[0054] Step S110, obtaining human pulse rate data to be predicted;

[0055] For example, in practice, an elderly person being cared for in a hospital wears a pulse rate monitoring device, and human pulse rate data is obtained from the output port of the device as the human pulse rate data to be predicted.

[0056] Step S120, inputting the human pulse rate data obtained in step S110 into a pre-constructed defecation and urination prediction algorithm model, and outputting the prediction results of the defecation and urination prediction algorithm model, wherein the defecation and urination prediction algorithm model is pre-constructed based on logistic regression.

[0057] For example, the prediction results can be output and sent to the smart terminal held by medical staff, so that the medical staff can perform relevant processing based on the prediction results, such as performing defecation processing operations in advance.

[0058] In this embodiment, the defecation prediction algorithm model is pre-built based on logistic regression, such as Figure 2 As shown, when the probability of two defecations output by the model is greater than 0.5, the prediction result is that two defecations will occur.

[0059] The technical solution of the present application, according to a pre-constructed defecation and urination prediction algorithm model, performs defecation and urination prediction based on human pulse rate data, which can easily realize the prediction of human defecation and urination, thereby helping to improve nursing efficiency in related scenarios and reduce the psychological burden of the cared for personnel.

[0060] In order to understand the technical solution of the present application, the construction process of the defecation prediction algorithm model in the technical solution of the present application is explained below.

[0061] The construction process of the defecation prediction algorithm model in this application includes:

[0062] Step 1, collecting human pulse rate parameter data;

[0063] Step 2: preprocessing the collected human pulse rate parameter data to obtain a pulse rate parameter data sample set;

[0064] Step 3: Based on the pulse rate parameter data sample set, an initial model of the defecation and urination prediction algorithm is constructed based on logistic regression;

[0065] Step 4: Optimize and test the initial model of the two-feces prediction algorithm to obtain the two-feces prediction algorithm model.

[0066] This will be described below in conjunction with the embodiments.

[0067] In one embodiment, during the model building process, step 1, the process of collecting human pulse rate parameter data, specifically includes:

[0068] The pulse rate data of the subjects are collected within a preset period of time in a day, and the time periods when defecation and urination events occur, the time periods when the subjects lie still, and the time periods when sudden events that cause abnormal pulse rate values ​​occur are marked in the pulse rate data. It is easy to understand that the sudden events here include exercise, equipment failure, etc., and marking of sudden events is conducive to the subsequent data preprocessing.

[0069] In actual implementation, the experimenter (subject) can be asked to wear a medical-grade pulse rate watch for a long time, and pulse rate data can be collected through the medical-grade pulse rate watch worn by the subject. The pulse rate data of no less than 9 hours a day should be collected, and the wearer or other relevant personnel (such as medical staff) should make relevant annotations.

[0070] like Figure 3 As shown, it is a schematic diagram for explaining the data collection for a certain subject. It is easy to understand that in practice, relevant data of multiple subjects for multiple days need to be collected separately, and it should be noted that the applicant conducted a preliminary analysis of these data and indeed found that the human body's defecation and urination behavior is related to the human pulse rate characteristics, which to a certain extent verifies the validity of the relevant research conclusions mentioned above.

[0071] Afterwards, step 2 is performed to pre-process the obtained pulse rate parameter data to obtain a pulse rate parameter data sample set; in this embodiment, the process includes:

[0072] The pulse rate parameter data of each individual subject are processed as follows:

[0073] Screening the pulse rate parameter data of the subject, and cleaning the data of abnormal pulse rate values ​​caused by emergencies based on the annotation information;

[0074] The screened pulse rate parameter data are analyzed and processed, and the pulse rate values ​​in the defecation and urination and lying down processes are averaged based on the annotation information, and the difference between the two average values ​​is used as the pulse rate data sample when the human body has defecation and urination behavior, and the data in the lying down process period is divided into two sections, and the average values ​​of the two sections are taken, and the difference between the two average values ​​is taken as the pulse rate data sample when the human body has not had defecation and urination behavior. This process is also called labeling.

[0075] like Figure 4 As shown, after processing the pulse rate parameter data of each subject, a pulse rate parameter data sample set (corresponding to Figure 4 (in the data sample set).

[0076] Then, proceed to step 3, build the initial model of the defecation prediction algorithm based on logistic regression according to the pulse parameter data sample set, which includes:

[0077] Step 3a, dividing the pulse rate parameter data sample set based on a preset method to form a training sample set and a test sample set;

[0078] In this embodiment, step 3a is specifically to divide the pulse rate parameter data sample set into the pulse rate data samples of the human body having defecation and urination behavior and the pulse rate data samples of the human body not having defecation and urination behavior in the ratio of A:B; then, in the division results, the two samples with a proportion of A are combined to form a training sample set, and the two samples with a proportion of B are combined to form a test sample set.

[0079] This method is helpful to make the formed sample set more general. For example, as a preferred embodiment, the A:B ratio is 8:2, which is conducive to the smooth progress of the overall construction process.

[0080] In addition, it should be noted that in some scenarios, there may be an imbalance between the data samples where defecation and urination occurred and the samples where defecation and urination did not occur (the difference in quantity is large, such as 200 on one side and 80 on the other side). In this case, the samples with a smaller number need to be expanded (such as re-performing steps 1 and 2) to achieve a relative balance in the number of data of the two types of samples in this step.

[0081] Afterwards, step 3b is performed, and the model parameters in the constructed model hypothesis function are solved using the maximum likelihood method based on the training sample set, and the initial model of the defecation prediction algorithm is determined based on the obtained model parameters.

[0082] The model hypothesis function and maximum likelihood method involved in step 3b belong to the prior art in logistic regression analysis. This application only briefly describes the construction process here, and the relevant undisclosed parts can be found in the existing public technical materials. Step 3b includes:

[0083] Establish a hypothesis function and set judgment rules; establish a loss function based on the hypothesis function; calculate the partial derivative of the loss function to obtain the gradient and construct a gradient descent function.

[0084] For example, the judgment rule is to set the threshold value to 0.5, that is, if the value of the function is greater than 0.5, it is predicted that the defecation and urination behavior occurs; otherwise, it is predicted that the defecation and urination behavior does not occur.

[0085] In the process of setting up the hypothesis function, the logistic regression algorithm is used to build the model, and when the output result y is 0, it is called the negative class, and when y is 1, it is called the positive class, where 0 represents no defecation and urination behavior, and 1 represents defecation and urination behavior.

[0086] The hypothesis function for logistic regression is:

[0087] (1)

[0088] In expression (1), is the pulse rate data sample input, is the model output, is the model parameter to be solved. Let 0.5 be the critical value. When hour, is 1, defecation occurs; when When hour, It is 0, and no defecation or urination behavior occurs.

[0089] Use the "maximum likelihood method" to estimate the value of θ. The purpose of maximum likelihood estimation is to use known sample results to infer the parameter value with the highest probability (most likely) leading to such a result.

[0090] Logistic regression uses the log-likelihood loss function:

[0091] (2)

[0092] Expression (2), is the pulse rate data sample input, is the value of the category of the input known pulse rate data sample.

[0093] Because the sample output is 0 and 1, then

[0094] (3)

[0095] (4)

[0096] Right now:

[0097] (5)

[0098] In expressions (3)-(5), Enter a value for the pulse rate, is the value of the known pulse rate data sample category, which is 0 and 1. are the model parameter values ​​to be solved. Output values ​​for the model.

[0099] The loss function is:

[0100]

[0101] (6)

[0102] Written in vector form:

[0103] (7)

[0104] In expression (7) is a vector of all 1s, is the value of the input known pulse rate data sample category, is the model output.

[0105] The iteration formula is:

[0106] (8)

[0107] In expression (8) are the old model parameter values, is the learning rate, is the gradient of the loss function.

[0108] Deriving the gradient of the loss function :

[0109] (9)

[0110] (10)

[0111] (11)

[0112] in is a vector of all 1s, is the pulse rate data sample input, are the model parameter values ​​to be solved, is the model output.

[0113] Substituting the above equations (10) and (11) into the logistic regression loss function formula (7), we get:

[0114] (12)

[0115] Calculate differentials based on matrix derivatives have to:

[0116] (13)

[0117] in is a vector of all 1s, is the pulse rate data sample input, are the model parameter values ​​to be solved, is the value of the category of the input known pulse rate data sample.

[0118] according to ,but:

[0119] (14)

[0120] Depend on , the gradient is:

[0121] (15)

[0122] in, is the pulse rate data sample input, is the model output, is the value of the input known pulse rate data sample category.

[0123] Substituting formula (15) into formula (8), the gradient descent method solves the parameter The iteration formula is:

[0124] (16)

[0125] In expression (16), are the old model parameter values, To predict the probability of category 1, is the learning rate, is the value of the input known pulse rate data sample category, is the value of the known pulse rate data sample feature input, and the condition for iteration cutoff is when the loss function value is less than the agreed value. In this way, the parameters in the initial model of the logistic regression prediction algorithm can be solved based on formula (16): worth it.

[0126] Then, based on the obtained parameters The initial model of the defecation prediction algorithm has been determined.

[0127] Finally, step 4 is performed to optimize and test the initial model of the two-feces prediction algorithm to obtain the two-feces prediction algorithm model. Specifically, this step includes:

[0128] The initial model of the feces prediction algorithm is trained according to the training sample set, and the learning rate parameters (learning rate) in the model are adjusted based on adaptive learning. ), to achieve model tuning;

[0129] The tuned model is tested and verified based on the test sample set. When the prediction accuracy of the model is greater than the preset threshold value (such as 80%), the model is used as the defecation prediction algorithm model.

[0130] It should be noted here that if the verification result is less than or equal to the threshold value, it means that the defecation prediction algorithm model cannot meet the basic application requirements, and the model construction process proposed in the above application needs to be repeated.

[0131] The technical solution of this application has the following advantages:

[0132] (a) In actual implementation and construction, medical-grade wristwatches can be used to collect pulse rate parameter data, which is easy to carry, simple to operate, and has high data accuracy, while reducing the burden on medical staff.

[0133] (b) The collected data is first preprocessed to clean up data mutations caused by emergencies, ensuring the authenticity of the data and providing accurate data for subsequent models.

[0134] (c) The collected data were reasonably labeled to improve the robustness of the model.

[0135] (d) After testing, the model has a high prediction rate and can realize the two-defecation prediction function. Applying it to intelligent nursing equipment can solve the problem of long-term care for disabled elderly people and save social costs. It has significant practical value and social benefits.

[0136] Figure 5 A schematic diagram of the structure of a human defecation prediction device based on logistic regression provided in one embodiment of the present application is shown in FIG. Figure 5 As shown, the human defecation prediction device 300 based on logistic regression includes:

[0137] An acquisition module 301 is used to acquire human pulse rate data to be predicted;

[0138] The prediction processing module 302 is used to input the human pulse rate data into a pre-constructed defecation and urination prediction algorithm model, and output the prediction results of the defecation and urination prediction algorithm model, wherein the pre-construction of the defecation and urination prediction algorithm model is based on logistic regression.

[0139] Regarding the human defecation prediction device 300 based on logistic regression in the above-mentioned related embodiments, the specific manner in which each module performs operations has been described in detail in the embodiments of the method, and will not be elaborated here.

[0140] Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present application is shown as follows: Figure 4 As shown, the electronic device 400 includes:

[0141] Memory 401, on which executable programs are stored;

[0142] The processor 402 is used to execute the executable program in the memory 401 to implement the steps of the above method.

[0143] Regarding the electronic device 400 in the above embodiment, the specific manner in which its processor 402 executes the program in the memory 401 has been described in detail in the embodiment of the method, and will not be elaborated here.

[0144] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0145] It should be noted that, in the description of this application, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of this application, unless otherwise specified, the meaning of "plurality" refers to at least two.

[0146] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0147] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0148] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0149] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0150] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0151] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0152] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in the field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for predicting human defecation and urination based on logistic regression, characterized in that: include: Acquiring human pulse rate data to be predicted; Inputting the human pulse rate data into a pre-built defecation and urination prediction algorithm model, and outputting the prediction results of the defecation and urination prediction algorithm model; Wherein, the defecation prediction algorithm model is pre-constructed based on logistic regression; Collecting human pulse rate parameter data includes: collecting pulse rate data of the subject within a preset time period in a day, and marking the time periods when defecation and urination events occur, the time periods when the subject lies still, and the time periods when sudden events that cause abnormal pulse values ​​occur in the pulse rate data.

2. The method according to claim 1, characterized in that The pre-construction of the defecation prediction algorithm model based on logistic regression includes: Collect human pulse rate parameter data; Preprocessing the pulse rate parameter data to obtain a pulse rate parameter data sample set; According to the pulse rate parameter data sample set, an initial model of the defecation and urination prediction algorithm is constructed based on logistic regression; The initial model of the two-feces prediction algorithm is tuned and tested to obtain the two-feces prediction algorithm model.

3. The method according to claim 2, characterized in that The pulse rate data is collected by a medical-grade pulse rate watch worn by the subject; the preset time period is at least 9 hours.

4. The method according to claim 2, characterized in that: The preprocessing of the pulse rate parameter data includes the following processing steps for the pulse rate parameter data of each individual subject: Screening the pulse rate parameter data of the subject, and cleaning the data of abnormal pulse rate values ​​caused by emergencies based on the annotation information; The screened pulse rate parameter data are analyzed and processed, and the pulse rate values ​​in the defecation and urination and lying still process periods are averaged based on the annotation information, and the difference between the two average values ​​is used as the pulse rate data sample when the human body has defecation and urination behavior, and the data in the lying still process period is divided into two sections, the average values ​​of the two sections are taken, and the difference between the two average values ​​is taken as the pulse rate data sample when the human body has not had defecation and urination behavior.

5. The method according to claim 4, characterized in that The method of constructing an initial model of a defecation prediction algorithm based on logistic regression according to the pulse rate parameter data sample set includes: Dividing the pulse rate parameter data sample set based on a preset method to form a training sample set and a test sample set; According to the training sample set, the maximum likelihood method is used to solve the model parameters in the constructed model hypothesis function, and the initial model of the defecation prediction algorithm is determined according to the obtained model parameters.

6. The method according to claim 5, characterized in that The pulse rate parameter data sample set is divided and processed based on a preset method, specifically: In the pulse rate parameter data sample set, the pulse rate data samples of the human body having defecation and urination behavior and the pulse rate data samples of the human body not having defecation and urination behavior are divided into a quantity ratio of 8:2 respectively; Then, in the division results, the two samples with a proportion of 8 are combined to form a training sample set, and the two samples with a proportion of 2 are combined to form a test sample set.

7. The method according to claim 5, characterized in that The initial model of the two-feces prediction algorithm is tuned and tested to obtain the two-feces prediction algorithm model, including: The initial model of the two-feces prediction algorithm is trained according to the training sample set, and the learning rate parameter in the model is adjusted based on adaptive learning to achieve model tuning; The tuned model is tested and verified according to the test sample set, and when the prediction accuracy of the model is greater than a preset threshold value, the model is used as the binary prediction algorithm model.

8. A human defecation prediction device based on logistic regression, characterized in that: include: An acquisition module is used to acquire human pulse rate data to be predicted; A prediction processing module, used for inputting the human pulse rate data into a pre-built two-feces prediction algorithm model, and outputting the prediction results of the two-feces prediction algorithm model; Wherein, the defecation prediction algorithm model is pre-constructed based on logistic regression; Collecting human pulse rate parameter data includes: collecting pulse rate data of the subject within a preset time period in a day, and marking the time periods when defecation and urination events occur, the time periods when the subject lies still, and the time periods when sudden events that cause abnormal pulse values ​​occur in the pulse rate data.

9. An electronic device, characterized in that: include: a memory having an executable program stored therein; A processor, configured to execute the executable program in the memory to implement the steps of the method according to any one of claims 1 to 7.

Images