Human body composition detection method and system, terminal and medium

Abstract

The invention discloses a human body composition detection method and system, a terminal and a medium, and the method comprises the steps: creating an exclusive data set for a specific crowd, and constructing a model equation based on the exclusive data set, the exclusive data set including but not limited to height, age, weight, gender and measured impedance; the coefficient of the model equation is solved based on a multiple regression analysis method, the overall significance of the model is analyzed and evaluated based on variance, and a human body composition detection model is obtained; and migrating the fusion model of the LSTM and the Transform into the human body composition detection model by using a migration learning technology, and obtaining a human body composition detection result based on the human body composition detection model. According to the invention, information of a specific crowd in multiple dimensions such as height, age, weight, gender, measured impedance and the like can be recorded, and comprehensiveness and accuracy of data are ensured. In addition, deep optimization and adjustment can be carried out on a traditional human body composition detection model, and the applicability and accuracy of the model to specific crowds are improved.

Description

Technical Field

[0001] This invention relates to the field of biomedical technology, and in particular to a method, system, terminal, and medium for detecting human body composition. Background Technology

[0002] Peritoneal dialysis is a common treatment for kidney disease, used to remove waste products and excess water from the body. The assessment of dialysis effectiveness typically relies on blood biochemical markers and body composition measurements. However, most existing body composition detection models are based on universally accepted threshold settings, and in the field of peritoneal dialysis, body composition detection is often limited to threshold settings specific to foreign populations, failing to adequately consider physiological differences between different populations. This can lead to potential biases in the results when assessing different population groups.

[0003] Significant differences exist among individuals in terms of age, gender, and physical condition, which can lead to substantial errors when existing empirical algorithms are applied to specific individuals. Furthermore, empirical algorithms are typically calibrated under specific conditions, such as within a particular age group or body type. When applied to individuals under different conditions, the accuracy of the algorithms may decrease significantly. In addition, short-term physiological changes, such as aging and disease progression, can also affect the measurement results of body composition. Empirical algorithms typically cannot dynamically adjust to adapt to these long-term changes, resulting in decreased accuracy. Empirical algorithms are often based on simplified assumptions, such as the assumption of uniform body composition distribution or linear relationships between certain physiological parameters. These assumptions often do not hold true in reality, leading to discrepancies between the model's predictions and actual conditions. Moreover, foreign body composition detection models face several limitations when applied to Asian populations, primarily due to differences in physiological characteristics, dietary habits, lifestyles, and data samples.

[0004] Therefore, existing technologies still have shortcomings. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method, system, terminal and medium for detecting human body composition, addressing the above-mentioned deficiencies of the prior art. The technical solution adopted by the present invention is as follows:

[0006] In a first aspect, the present invention provides a method for detecting human body composition, wherein the method includes:

[0007] Create a dedicated dataset for a specific population and construct model equations based on the dedicated dataset, wherein the dedicated dataset includes, but is not limited to, height, age, weight, gender, and measured impedance;

[0008] The coefficients of the model equation are solved using multiple regression analysis, and the overall significance of the model is evaluated using analysis of variance, thus obtaining the human body composition detection model.

[0009] The fusion model of LSTM and Transformer is transferred to the human body composition detection model using transfer learning technology, and the human body composition detection results are obtained based on the human body composition detection model.

[0010] In one implementation, the model equation is:

[0011] X = a0 + a1Hi + a2Z i / Hi+a3sex i +a4weight i +a5agei

[0012] Where X is the human body composition value to be predicted, Hi is the height of the i-th sample, and Z is the height of the i-th sample. i It is the system impedance value of the i-th sample, sex i It is the gender of the i-th sample, and the weight i Let be the weight of the i-th sample, agei be the age of the i-th sample, and a0, a1, ..., a5 be the coefficients of the model method to be solved.

[0013] In one implementation, the coefficients of the model equation are solved using a multiple regression analysis method, and the overall significance of the model is evaluated using analysis of variance to obtain a human body composition detection model, including:

[0014] For each physiological parameter, the correlation coefficient of the model is calculated, and the relevant factor terms with high correlation coefficients are added to the multiple regression model.

[0015] Use SPSS to perform linear regression and solve for the coefficients in the model equation;

[0016] The coefficients whose significance is less than the significance level are retained to obtain the human body composition detection model.

[0017] In one implementation, the method of solving the coefficients of the model equation based on multiple regression analysis, evaluating the overall significance of the model based on analysis of variance, and obtaining the human body composition detection model further includes:

[0018] The model is calculated using its coefficient of determination, adjusted coefficient of determination, and standard estimation error. A larger coefficient of determination indicates a more complete explanation of the dependent variable by the model; a larger adjusted coefficient of determination indicates a better model fit; and the standard estimation error reflects the accuracy of the statistics obtained through sampling.

[0019] In one implementation, the method further includes:

[0020] The effectiveness of dialysis is evaluated based on the results of human body composition analysis, and early warning results are output.

[0021] In one implementation, the method further includes:

[0022] Treatment plans are developed based on the results of the human body composition analysis, and the parameters in the treatment plans are adjusted accordingly.

[0023] Secondly, an embodiment of the present invention provides a human body composition detection system, wherein the system comprises:

[0024] The dataset creation module is used to create a dedicated dataset for a specific population and to build model equations based on the dedicated dataset. The dedicated dataset includes, but is not limited to, height, age, weight, gender, and measured impedance.

[0025] The multivariate regression analysis module is used to solve the coefficients of the model equation based on the multivariate regression analysis method, and to evaluate the overall significance of the model based on the analysis of variance, so as to obtain the human body composition detection model.

[0026] The transfer learning module is used to transfer the fusion model of LSTM and Transformer to the human body composition detection model using transfer learning technology, and to obtain the human body composition detection results based on the human body composition detection model.

[0027] In one implementation, the system is used to implement the steps of the human body composition detection method described above, and the system further includes:

[0028] The assessment and early warning module is used to assess the effectiveness of dialysis based on the results of human body composition analysis and to output early warning results.

[0029] The treatment plan suggestion and adjustment module is used to formulate a treatment plan based on the human body composition detection results and adjust the parameters in the treatment plan.

[0030] Thirdly, embodiments of the present invention also provide a terminal, wherein the terminal includes a memory, a processor, and a human composition detection program stored in the memory and executable on the processor. When the processor executes the human composition detection program, it implements the steps of the human composition detection method of any of the above solutions.

[0031] Fourthly, embodiments of the present invention also provide a computer-readable storage medium, wherein a human composition detection program is stored on the computer-readable storage medium, and the human composition detection program implements the steps of the human composition detection method according to any one of the embodiments on the computer-readable storage medium.

[0032] Beneficial Effects: Compared with existing technologies, this invention provides a method for body composition detection. First, this invention creates a dedicated dataset for a specific population and constructs a model equation based on this dataset. The dedicated dataset includes, but is not limited to, height, age, weight, gender, and measured impedance. Then, the coefficients of the model equation are solved using multiple regression analysis, and the overall significance of the model is evaluated using analysis of variance, resulting in a body composition detection model. Next, transfer learning technology is used to transfer a fusion model of LSTM and Transformer to the body composition detection model, and body composition detection results are obtained based on the model. This invention can record information on a specific population across multiple dimensions, such as height, age, weight, gender, and measured impedance, ensuring the comprehensiveness and accuracy of the data. Furthermore, this invention can deeply optimize and adjust traditional body composition detection models, improving the model's applicability and accuracy for specific populations. Attached Figure Description

[0033] Figure 1 A flowchart of a preferred embodiment of the human body composition detection method provided in this invention.

[0034] Figure 2 This is a schematic diagram of the human body composition detection system according to an embodiment of the present invention.

[0035] Figure 3 A schematic diagram of a terminal provided in an embodiment of the present invention. Detailed Implementation

[0036] To make the objectives, technical solutions, and effects of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0037] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content, operations, or steps, nor does it require execution in the described order. For example, some operations or steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0038] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0039] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. For example, "first control information" and "second control information" are only used to distinguish different control information and do not limit their order.

[0040] Those skilled in the art will understand that the words "first" and "second" do not limit the quantity or the order of execution, and that the words "first" and "second" do not necessarily imply that they are different.

[0041] It should also be understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0042] To address the insufficient accuracy of existing empirical formulas and body composition detection models in evaluating the treatment efficacy of peritoneal dialysis patients, this invention proposes a body composition detection method for a specific population. In practical application, this embodiment first creates a dedicated dataset for the specific population and constructs a model equation based on this dataset. The dedicated dataset includes, but is not limited to, height, age, weight, gender, and measured impedance. Then, the coefficients of the model equation are solved using multiple regression analysis, and the overall significance of the model is evaluated using analysis of variance, resulting in a body composition detection model. Finally, a fusion model of LSTM and Transformer is transferred to the body composition detection model using transfer learning technology, and the body composition detection results are obtained based on the model. This embodiment can record information on multiple dimensions of the specific population, such as height, age, weight, gender, and measured impedance, ensuring the comprehensiveness and accuracy of the data. Furthermore, this invention can deeply optimize and adjust traditional body composition detection models, improving the model's applicability and accuracy for specific populations.

[0043] The human body composition detection method of this embodiment can be applied to a terminal, which can be a smart product terminal such as a computer, mobile phone, or smart TV. Figure 1 As shown in the figure, the human body composition detection method of this embodiment includes the following steps:

[0044] Step S100: Create a dedicated dataset for a specific population and construct a model equation based on the dedicated dataset, wherein the dedicated dataset includes, but is not limited to, height, age, weight, gender, and measured impedance.

[0045] To address the shortcomings of existing technologies, such as insufficient consideration of population specificity, limited model construction and training methods, inadequate functional modules and performance, and a lack of in-depth research on the subsequent application of detection results, this invention first constructs a dedicated dataset for a specific population by collecting sample data from long-term dialysis patients, particularly those in the Asian population. This dedicated dataset reflects multi-dimensional information about the population, including height, age, weight, gender, and measured impedance, ensuring the comprehensiveness and accuracy of the data. Based on this dedicated dataset, this embodiment utilizes machine learning techniques to deeply optimize and adjust the traditional body composition detection model. Specifically, in setting the water load threshold, it no longer relies on general standards but is dynamically customized based on population-specific data. By collecting physiological indicators, clinical data, and environmental factors from different populations across multiple centers, a multivariate predictive model, such as a machine learning or statistical model, is established to dynamically adjust the water load threshold. Furthermore, this embodiment can also optimize personalized water load threshold adjustments based on real-time monitoring of relevant physiological parameters such as body composition, weight, and cardiovascular status using portable wearable devices. This embodiment utilizes the aforementioned dedicated dataset to deeply optimize and adjust the traditional body composition detection model. In particular, the threshold setting no longer relies on general standards, but is dynamically customized based on Chinese-specific data to improve the model's applicability and accuracy to the Chinese population.

[0046] Next, this embodiment constructs a model equation based on the aforementioned dedicated dataset. The model equation is:

[0047] X = a0 + a1Hi + a2Z i / Hi+a3sex i +a4weight i +a5agei

[0048] Where X is the human body composition value to be predicted, Hi is the height of the i-th sample, and Z is the height of the i-th sample. i It is the system impedance value of the i-th sample, sex i It is the gender of the i-th sample, and the weight i Let be the weight of the i-th sample, agei be the age of the i-th sample, and a0, a1, ..., a5 be the coefficients of the model method to be solved.

[0049] It should be noted that, in addition to basic parameters such as height, age, weight, gender, and system impedance, competitors can introduce more physiological parameters, such as heart rate, blood pressure, and blood oxygen saturation, to further enrich the data dimensions and improve the accuracy of the model.

[0050] Step S200: Solve the coefficients of the model equation based on the multiple regression analysis method, and evaluate the overall significance of the model based on the analysis of variance to obtain the human body composition detection model.

[0051] After establishing the model equations, the next step is to estimate the values ​​of the coefficients a0, a1, ..., a5 in the model using multiple regression analysis. Specifically, in this embodiment, the correlation coefficients of the model are first calculated for each physiological parameter, and the relevant factor terms with high correlation coefficients are added to the multiple regression model. Then, linear regression is performed using SPSS to solve for the coefficients in the model equations. Finally, coefficients with significance levels below the significance level are retained, resulting in the human body composition detection model.

[0052] Taking non-fat content as an example to explain the multiple regression processing method, firstly, a linear model of non-fat mass and weight is established using multiple sets of sample data for a physiological parameter such as weight. Secondly, the correlation coefficient between non-fat mass and weight is calculated. If the correlation coefficient is high, it is necessary to add the correlation factor. Thirdly, the above process is repeated to calculate the correlation parameters between other parameters and non-fat tissue, and the correlation factor with high correlation coefficients is added to the multiple regression model. Next, linear regression is performed using SPSS to obtain the regression parameters and the model equation. These coefficients represent the degree of influence of each variable on the predicted value. This embodiment calculates the correlation coefficient (R) and the coefficient of determination (R²) of the model. 2 Adjusted R-value 2 The correlation coefficient (R) of the model is obtained using the CORREL function. Taking the relationship between non-fat mass and body weight as an example, the closer the absolute value of the correlation coefficient obtained by the CORREL function is to 1, the higher the correlation level. The coefficient of determination (R²) 2 The formula for calculating R is: R = 1 - SSE / SST, where SSE (Sum of Squares due to Error) is the residual sum of squares, representing the difference between the observed values ​​and the predicted values ​​of the regression model, i.e., the degree of unexplained variation by the model. SST (Total Sum of Squares) is the total sum of squares, which is the total variation of the observed values ​​of the dependent variable relative to its mean. Therefore, by calculating the residual sum of squares SSE between fat and body weight and the total sum of squares SST, the coefficient of determination can be calculated. The larger the R value, the more fully the model explains the dependent variable. Adjusted coefficient of determination:

[0053] AdjustedR 2 = [SSE / (nk-1)] / [SST / (n-1)], by increasing the degrees of freedom k, the influence of the number of independent variables n on the coefficient of determination is avoided. Generally, the larger the adjusted coefficient of determination, the better the fit. The standard error of the estimate. Where s is the standard deviation The accuracy of the statistics obtained through sampling calculations is described. Finally, coefficients with significance levels below the significance level are retained, resulting in the final body composition detection model. This embodiment uses analysis of variance (ANOVA) to evaluate the overall significance of the model. SPSS is used to calculate and analyze the significance level of the multiple linear regression model. If the model equation has a high goodness of fit and the significance level of the parameter to be solved is below the significance level, it indicates that the parameter is highly representative. In the non-fat mass calculation, the significance level is obtained for each parameter in the calculation model (height, weight, age, gender, impedance), and parameters that do not meet the significance level are then removed, resulting in the final body composition detection model. This invention uses multiple regression analysis to estimate model coefficients, and by calculating relevant statistics (such as R, R², etc.), the model performance can be evaluated more accurately.

[0054] Step S300: Use transfer learning technology to transfer the fusion model of LSTM and Transformer to the human body composition detection model, and obtain the human body composition detection result based on the human body composition detection model.

[0055] This embodiment utilizes transfer learning technology to transfer a fusion model of LSTM (Long Short-Term Memory) + Transformer to a body composition detection model. Since processing time-series data is essential for body composition detection, the LSTM model is chosen to handle the transform data during the detection process. Simultaneously, because the Transformer employs a self-attention mechanism, it allows the model to focus on information at different locations when processing sequential data. This mechanism allows for parallel computation, giving the Transformer model extremely high computational efficiency and parallel processing capabilities, effectively reducing computational cost. Furthermore, the Transformer's self-attention mechanism enables the body composition detection model to better adapt to long-series data, a significant advantage in processing long-term sequence data collected by adult body composition analyzers. After transferring the model, basic fine-tuning parameters such as learning rate, batch size, optimizer, and number of training epochs, as well as efficient fine-tuning parameters such as rank and the Drupout regularization technique, are fitted and fine-tuned based on population data to find the most suitable parameters for this task. Mature model architectures and parameters from other related fields are also transferred to the body composition detection model and fitted and fine-tuned based on population data.

[0056] This invention introduces transfer learning technology, transferring mature model architectures and parameters from other related fields to a body composition detection model, and fine-tuning the fit based on data from a specific population to further improve the model's predictive ability. While this invention uses multiple regression analysis and transfer learning, competitors can explore other advanced machine learning methods, such as deep neural networks, support vector machines (SVM), and random forests, to improve the model's predictive performance.

[0057] In application, this embodiment can also use body composition analysis results for dialysis effectiveness evaluation and early warning systems, and it is closely integrated with the formulation and adjustment of treatment plans. Furthermore, artificial intelligence technology can be used to develop more advanced algorithms to automatically provide doctors with personalized treatment plans based on body composition analysis results, such as adjusting dialysis parameters and optimizing drug dosages. This invention can also collaborate with experts in other fields, such as nutrition and sports science, to jointly research how to better utilize body composition analysis results to develop comprehensive health management programs.

[0058] Based on the above embodiments, the present invention also provides a body composition detection system, which is used to implement the steps of the above-described body composition detection method embodiments. This body composition detection system is integrated with a cloud computing platform and mobile application to achieve remote monitoring and management, improving user experience and service scope. It includes, but is not limited to, functional modules such as dialysis effect evaluation, early warning system, treatment plan formulation and adjustment, etc. Furthermore, it can continuously collect user data and provide real-time feedback results, offering more dynamic and personalized health management services.

[0059] In practical applications, such as Figure 2 As shown, the system in this embodiment includes: a dataset creation module 10, a multiple regression analysis module 20, and a transfer learning module 30. The dataset creation module 10 is used to create a specific dataset for a particular population and construct a model equation based on the specific dataset. The specific dataset includes, but is not limited to, height, age, weight, gender, and measured impedance. The multiple regression analysis module 20 is used to solve the coefficients of the model equation using multiple regression analysis methods and evaluate the overall significance of the model based on analysis of variance to obtain a body composition detection model. The transfer learning module 30 is used to transfer a fusion model of LSTM and Transformer to the body composition detection model using transfer learning techniques and obtain body composition detection results based on the body composition detection model.

[0060] In one implementation, the system further includes:

[0061] The assessment and early warning module is used to assess the effectiveness of dialysis based on the results of human body composition analysis and to output early warning results.

[0062] The treatment plan suggestion and adjustment module is used to formulate a treatment plan based on the human body composition detection results and adjust the parameters in the treatment plan.

[0063] The principles of each step in the human body composition detection system of this embodiment are the same as those of the method embodiments described above, and will not be repeated here.

[0064] Each module of the aforementioned drug concentration control device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of the terminal in hardware form or independent of it, or stored in the memory of the terminal in software form, so that the processor can call and execute the corresponding operations of each module.

[0065] Based on the above embodiments, the present invention also provides a terminal, the principle block diagram of which can be as follows: Figure 3 As shown. The terminal may include one or more processors 100 ( Figure 3 (Only one is shown in the image), a memory 101, and a computer program 102 stored in the memory 101 and executable on one or more processors 100. For example, a body composition detection program. When one or more processors 100 execute the computer program 102, they can implement the various steps in the body composition detection method embodiments. Alternatively, when one or more processors 100 execute the computer program 102, they can implement the functions of various modules / units in the body composition detection method embodiments, without limitation here.

[0066] In one embodiment, the processor 100 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0067] In one embodiment, memory 101 may be an internal storage unit of an electronic device, such as a hard drive or RAM. Memory 101 may also be an external storage device of the electronic device, such as a plug-in hard drive, smart media card (SMC), secure digital (SD) card, flash card, etc. Furthermore, memory 101 may include both internal and external storage units. Memory 101 is used to store computer programs and other programs and data required by the terminal. Memory 101 can also be used to temporarily store data that has been output or will be output.

[0068] Those skilled in the art will understand that Figure 3 The block diagram shown is merely a partial structural diagram related to the present invention and does not constitute a limitation on the terminal to which the present invention is applied. A specific terminal may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0069] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the methods described above. Any references to memory, storage, operational databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual operating data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAM bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0070] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting human body composition, characterized in that, The method includes: Create a dedicated dataset for a specific population and construct model equations based on the dedicated dataset, wherein the dedicated dataset includes, but is not limited to, height, age, weight, gender, and measured impedance; The coefficients of the model equation are solved using multiple regression analysis, and the overall significance of the model is evaluated using analysis of variance, thus obtaining the human body composition detection model. The fusion model of LSTM and Transformer is transferred to the human body composition detection model using transfer learning technology, and the human body composition detection results are obtained based on the human body composition detection model.

2. The method for detecting human body composition according to claim 1, characterized in that, The model equation is as follows: X=a0+a1Hi+a2Z i / Hi+a3sex i +a4weight i +a5agei Where X is the human body composition value to be predicted, Hi is the height of the i-th sample, and Z is the height of the i-th sample. i It is the system impedance value of the i-th sample, sex i It is the gender of the i-th sample, and the weight i Let be the weight of the i-th sample, agei be the age of the i-th sample, and a0, a1, ..., a5 be the coefficients of the model method to be solved.

3. The method for detecting human body composition according to claim 2, characterized in that, The coefficients of the model equation are solved using multiple regression analysis, and the overall significance of the model is evaluated using analysis of variance to obtain the human body composition detection model, which includes: For each physiological parameter, the correlation coefficient of the model is calculated, and the relevant factor terms with high correlation coefficients are added to the multiple regression model. Use SPSS to perform linear regression and solve for the coefficients in the model equation; The coefficients whose significance is less than the significance level are retained to obtain the human body composition detection model.

4. The method for detecting human body composition according to claim 3, characterized in that, The method of solving the model equations using multiple regression analysis, evaluating the overall significance of the model using analysis of variance, and obtaining the human body composition detection model further includes: The model is calculated using its coefficient of determination, adjusted coefficient of determination, and standard estimation error. A larger coefficient of determination indicates a more complete explanation of the dependent variable by the model; a larger adjusted coefficient of determination indicates a better model fit; and the standard estimation error reflects the accuracy of the statistics obtained through sampling.

5. The method for detecting human body composition according to claim 1, characterized in that, The method further includes: The dialysis effect is evaluated based on the results of human body composition analysis, and early warning results are output.

6. The method for detecting human body composition according to claim 1, characterized in that, The method further includes: Treatment plans are developed based on the results of the human body composition analysis, and the parameters in the treatment plans are adjusted accordingly.

7. A human body composition detection system, characterized in that, The system is used to implement the steps of the human body composition detection method according to any one of claims 1-6, and the system comprises: The dataset creation module is used to create a dedicated dataset for a specific population and to build model equations based on the dedicated dataset. The dedicated dataset includes, but is not limited to, height, age, weight, gender, and measured impedance. The multivariate regression analysis module is used to solve the coefficients of the model equation based on the multivariate regression analysis method, and to evaluate the overall significance of the model based on the analysis of variance, so as to obtain the human body composition detection model. The transfer learning module is used to transfer the fusion model of LSTM and Transformer to the human body composition detection model using transfer learning technology, and to obtain the human body composition detection results based on the human body composition detection model.

8. The human body composition detection system according to claim 7, characterized in that, The system also includes: The assessment and early warning module is used to assess the effectiveness of dialysis based on the results of human body composition analysis and to output early warning results. The treatment plan suggestion and adjustment module is used to formulate a treatment plan based on the human body composition detection results and adjust the parameters in the treatment plan.

9. A terminal, characterized in that, The terminal includes a memory, a processor, and a human composition detection program stored in the memory and executable on the processor. When the processor executes the human composition detection program, it implements the steps of the human composition detection method as described in any one of claims 1-6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a human composition detection program, which implements the steps of the human composition detection method as described in any one of claims 1-6 on the computer-readable storage medium.

Images