A bioimpedance-based intelligent optimization system and method for human body composition analysis

Abstract

This invention provides a bioimpedance-based intelligent optimization system and method for human body composition analysis, comprising: an impedance measurement module for acquiring multi-frequency electrical signals applied to human body parts and collecting multi-frequency complex impedance data; a data preprocessing module for preprocessing the multi-frequency complex impedance data; an algorithm fitting module for globally optimizing and fitting a Cole-Cole model using the multi-frequency complex impedance data as input to obtain model parameters for each measurement loop; and a human body composition calculation module for calculating human body composition parameters for multiple limb segments based on the model parameters of each measurement loop. The advantages of this invention are that by employing multi-frequency complex impedance measurement technology, it simultaneously acquires information on both the real and imaginary parts of impedance, accurately distinguishing between intracellular and extracellular fluids, and significantly improving the accuracy of human body composition analysis; and by using an improved flower pollination algorithm instead of the traditional least squares method, it effectively avoids the problem of local optima and improves the fitting accuracy and robustness of the model parameters.

Description

Technical Field

[0001] This invention belongs to the field of human body composition analysis technology, and in particular relates to a bioimpedance-based intelligent optimization human body composition analysis system and method. Background Technology

[0002] With an aging population and rising incidence of chronic diseases, the demand for accurate and convenient body composition analysis has surged. Bioelectrical impedance analysis (BIA) is currently a commonly used method for body composition analysis. It estimates body composition by measuring the body's electrical impedance after applying a weak alternating current signal. However, traditional body composition measurement methods, such as dual-energy X-ray ablation (DXA), are expensive and radioactive, while existing BIA equipment is limited by algorithms and hardware design, making it difficult to meet clinical-grade accuracy requirements. Existing BIA analysis algorithms also have some limitations. When fitting human impedance data, they often use simple linear models or models with fixed parameters, which cannot well adapt to the physiological differences between individuals. Static models based on anthropometric indicators such as height, weight, and waist circumference are simple to operate but rely on static anthropometric indicators (such as BMI) and cannot deeply analyze the components of intracellular and extracellular fluids. Single-frequency BIA cannot distinguish the frequency response characteristics of different tissues, leading to errors in the estimation of body fluid distribution and failing to correlate with disease risk. Multi-frequency BIA is prone to getting trapped in local optima, has slow convergence speed, and insufficient parameter extraction accuracy. Therefore, there is an urgent need for a body composition analysis system and method that solves the above problems. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention provides a bioimpedance intelligent optimization human body composition analysis system and method, which is particularly suitable for high-precision analysis of human body composition based on bioimpedance measurement and intelligent optimization algorithms.

[0004] The technical solution adopted in this invention is as follows: Firstly, a bioimpedance-based intelligent optimized human body composition analysis system is provided, comprising:

[0005] The impedance measurement module includes a multi-electrode measurement unit, an analog switch array, and a multi-frequency complex impedance detection unit. It is used to acquire multi-frequency electrical signals applied to human body parts and to collect multi-frequency complex impedance data. The multi-electrode measurement unit is electrically connected to the human body, with at least one excitation electrode and at least one detection electrode set at multiple measurement sites on the body. The analog switch array, all excitation electrodes, all detection electrodes, and the multi-frequency complex impedance detection unit are electrically connected to time-division multiplex different electrode combinations, forming multiple independent measurement loops. The multi-frequency complex impedance detection unit injects a constant alternating current into the selected measurement loop and detects the voltage signal across the measurement loop, calculating the multi-frequency complex impedance data for each measurement loop.

[0006] The data preprocessing module is used to perform noise reduction and standardization on the multi-frequency complex impedance data of each measurement loop;

[0007] The algorithm fitting module is used to perform global optimization fitting of the Cole-Cole model with the multi-frequency complex impedance data of each measurement loop as input, and obtain the Cole-Cole model parameters of each measurement loop.

[0008] The body composition calculation module is used to calculate the body composition parameters of multiple limb segments based on the Cole-Cole model parameters of each measurement loop.

[0009] The results output module is used to output the calculated human body composition parameters.

[0010] Furthermore, the multi-electrode measurement unit is equipped with an excitation electrode and a detection electrode on the left upper limb, right upper limb, left lower limb, and right lower limb of the human body, forming six measurement circuits including the left upper limb-right upper limb circuit, the left upper limb-left lower limb circuit, the left upper limb-right lower limb circuit, the right upper limb-left lower limb circuit, the right upper limb-right lower limb circuit, and the left lower limb-right lower limb circuit.

[0011] Furthermore, it also includes:

[0012] The local edema monitoring module is used to calculate the local tissue edema index based on the phase characteristics of the multi-frequency complex impedance data of each measurement circuit, identify the risk of edema, and classify it.

[0013] The metabolic rate calculation module is used to acquire the user's physiological data and estimate the user's basal metabolic rate based on the Cole-Cole model parameters and physiological data.

[0014] The health risk warning module is used to fuse complex impedance data, human body composition parameters and physiological data at various frequencies, input them into a pre-trained neural network model, and output health risk classification results and personalized intervention suggestions.

[0015] The multimodal timing and topology modeling module is used to process complex impedance timing data using the Transformer architecture and to construct a human topology model using a graph neural network to improve the ability to detect local component anomalies. The complex impedance timing data consists of multi-frequency complex impedance data of each measurement loop acquired multiple times at different time points.

[0016] Furthermore, the algorithm fitting module employs an improved flower pollination algorithm, which includes a dual-mode pollination mechanism: performing a global pollination operation based on the Levy flight strategy with a first probability, and performing a local pollination operation based on Gaussian perturbation with a second probability; the Cole-Cole model parameters include low-frequency impedance, high-frequency impedance, characteristic frequency, and dispersion coefficient.

[0017] Furthermore, the human body composition calculation module uses the Zhu-Moissl fusion equation to calculate the human body composition parameters of each limb segment. The limb segments are calculated using the Zhu equation, and the trunk segments are calculated using the difference between total body water and limb water. The human body composition parameters include extracellular fluid, intracellular fluid, and total water.

[0018] Secondly, a bioimpedance-based intelligent optimization method for human body composition analysis is provided, comprising the following steps:

[0019] Establish measurement circuits for multiple parts of the human body and collect multi-frequency complex impedance data;

[0020] Denoising and standardization processing is performed on multi-frequency complex impedance data;

[0021] Using multi-frequency complex impedance data as input, the Cole-Cole model is globally optimized and fitted, and the corresponding Cole-Cole model parameters are obtained by solving the problem.

[0022] The human body composition parameters of each limb segment were derived based on the parameters of the Cole-Cole model.

[0023] Furthermore, deriving the human body composition parameters for each limb segment based on the Cole-Cole model parameters includes the following steps:

[0024] Establish a set of impedance relationship equations between limb segments and measurement circuits, and extract low-frequency and high-frequency impedances from the Cole-Cole model parameters;

[0025] Human body composition parameters for each limb segment were calculated based on sex-specific segmental resistivity and in combination with low-frequency and high-frequency impedance. These parameters included extracellular fluid volume, intracellular fluid volume, and total water content.

[0026] Thirdly, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the bioimpedance intelligent optimization human composition analysis method provided herein.

[0027] Fourthly, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause a computer to execute the bioimpedance intelligent optimization human body composition analysis method provided in this disclosure.

[0028] Fifthly, a computer program product is provided, including a computer program / instructions, which are executed by a processor using the bioimpedance intelligent optimization human composition analysis method disclosed herein.

[0029] The advantages and positive effects of this invention are as follows: by adopting the above technical solution and using multi-frequency complex impedance measurement technology to simultaneously obtain the real and imaginary parts of impedance, it is possible to accurately distinguish between intracellular and extracellular fluids, which greatly improves the accuracy of human body composition analysis; by using an improved flower pollination algorithm to replace the traditional least squares method, the problem of local optima is effectively avoided, and the fitting accuracy and robustness of the Cole-Cole model parameters are improved. Attached Figure Description

[0030] Figure 1 This is a schematic flowchart of a human body composition analysis method according to an embodiment of the present invention. Detailed Implementation

[0031] The present disclosure will now be described more fully with reference to the accompanying drawings, which illustrate exemplary embodiments of the present disclosure. The technical solutions of the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present disclosure, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present disclosure without creative effort are within the scope of protection of the present disclosure.

[0032] This invention provides a bioimpedance-based intelligent optimized human body composition analysis system, comprising:

[0033] The impedance measurement module includes a multi-electrode measurement unit, an analog switch array, and a multi-frequency complex impedance detection unit. It is used to acquire multi-frequency electrical signals applied to human body parts and to collect multi-frequency complex impedance data. The multi-electrode measurement unit forms an electrical connection with the human body, with at least one excitation electrode and at least one detection electrode set at multiple measurement sites on the body. The analog switch array, all excitation electrodes, all detection electrodes, and the multi-frequency complex impedance detection unit are electrically connected to time-division multiplex different electrode combinations, forming multiple independent measurement loops. The multi-frequency complex impedance detection unit injects a constant alternating current into the selected measurement loop and detects the voltage signal across the measurement loop, calculating the multi-frequency complex impedance data for each measurement loop.

[0034] Specifically, the impedance measurement module is the core of the system's data acquisition. It is responsible for injecting a safe, weak alternating current into the human body, detecting the corresponding voltage signal, and calculating multi-frequency complex impedance data. Complex impedance data specifically refers to complex data including both real (resistance) and imaginary (reactance) components, while also containing amplitude and phase angle information. The multi-electrode measurement unit consists of an electrode handle and an electrode pedal, and is equipped with several sets of excitation and detection electrodes. The input of the analog switch array is connected to the electrodes (excitation and detection electrodes) of the multi-electrode measurement unit, and the output is connected to the multi-frequency complex impedance detection unit. Under the control of the main control unit, the analog switch array can quickly switch the electrode connection mode, time-divisionally selecting different electrode combinations to form multiple independent measurement loops. The multi-frequency complex impedance detection unit can generate multi-frequency sinusoidal excitation signals and perform a Fast Fourier Transform (FFT) on the response signal to extract the real and imaginary parts of the impedance, calculating the amplitude and phase of the complex impedance.

[0035] The data preprocessing module is used to perform noise reduction and standardization on the multi-frequency complex impedance data of each measurement loop;

[0036] Specifically, the data preprocessing module is responsible for performing wavelet transform noise reduction, calibration, outlier removal, data smoothing, and data standardization on the original complex impedance data, providing high-quality data input for subsequent model fitting.

[0037] The algorithm fitting module is used to perform global optimization fitting of the Cole-Cole model with the multi-frequency complex impedance data of each measurement loop as input, and obtain the Cole-Cole model parameters of each measurement loop.

[0038] The body composition calculation module is used to calculate the body composition parameters of multiple limb segments based on the Cole-Cole model parameters of each measurement loop.

[0039] The results output module is used to output the calculated human body composition parameters.

[0040] Specifically, the results output module is responsible for presenting the calculated body composition parameters to the user in an intuitive way. The system is equipped with an LCD display screen that can show the measurement progress and body composition parameters in real time. Simultaneously, data can be synchronized to a mobile app or cloud server via Bluetooth 5.0 or Wi-Fi to generate detailed health reports. This system also includes necessary modules to ensure the normal operation of the system, such as a power supply module and a main control module, all of which are existing technologies.

[0041] By employing the above setup and multi-frequency complex impedance measurement technology, the real and imaginary parts of impedance can be acquired simultaneously, providing the necessary data foundation for distinguishing intracellular and extracellular fluids and significantly improving the accuracy of human body composition analysis. By using an analog switch array to time-division select multiple independent measurement loops, multiple site measurements can be achieved without the need for manual electrode switching by the user, significantly improving measurement efficiency and user experience. By using a global optimization fitting algorithm instead of the traditional least squares method, the problem of local optima is effectively avoided, improving the fitting accuracy and stability of the Cole-Cole model parameters.

[0042] To address the issue that general multi-electrode configurations cannot accurately calculate trunk impedance, this embodiment provides an implementation method.

[0043] In one embodiment, the multi-electrode measurement unit is provided with an excitation electrode and a detection electrode on the left upper limb, right upper limb, left lower limb, and right lower limb of the human body, forming six measurement circuits including the left upper limb-right upper limb circuit, the left upper limb-left lower limb circuit, the left upper limb-right lower limb circuit, the right upper limb-left lower limb circuit, the right upper limb-right lower limb circuit, and the left lower limb-right lower limb circuit.

[0044] Using the above setup and a standard eight-electrode, six-circuit configuration, the impedance values ​​of each independent segment can be accurately calculated by solving a system of simultaneous equations, enabling precise analysis of the components of all parts of the body.

[0045] To address the issue that traditional body composition analysis systems are limited in function, only able to output basic body composition parameters and unable to meet users' needs for health risk assessment, dynamic monitoring, and personalized intervention, this embodiment provides an implementation method.

[0046] In one embodiment, the human body composition analysis system further includes:

[0047] The local edema monitoring module is used to calculate the local tissue edema index based on the phase characteristics of the multi-frequency complex impedance data of each measurement circuit, identify the risk of edema, and classify it.

[0048] Specifically, the local edema monitoring module analyzes the phase characteristics of multi-frequency complex impedance data, comparing the 50kHz phase angle of the test subject with the average 50kHz phase angle of healthy individuals to achieve early detection and grading of edema. It can also utilize wavelet transform to separate the phase response characteristics (50kHz phase change) of different tissues, distinguishing between subcutaneous edema and muscle inflammation. By performing multi-scale decomposition of the phase signal and extracting features from different frequency bands, it is found that subcutaneous edema mainly affects the low-frequency phase, while muscle inflammation mainly affects the high-frequency phase.

[0049] The metabolic rate calculation module is used to acquire the user's physiological data and estimate the user's basal metabolic rate based on the Cole-Cole model parameters and physiological data.

[0050] Specifically, the metabolic rate calculation module is used to acquire the user's physiological data, including BMI, gender, and age. Combined with the low-frequency impedance and high-frequency impedance parameters in the Cole-Cole model, the user's basal metabolic rate is calculated.

[0051] The health risk warning module is used to fuse complex impedance data, human body composition parameters and physiological data at various frequencies, input them into a pre-trained neural network model, and output health risk classification results and personalized intervention suggestions. The neural network model, health risk classification results and personalized intervention suggestions in the health risk warning module are all obtained using existing technologies.

[0052] The multimodal timing and topology modeling module is used to process complex impedance timing data using the Transformer architecture and to construct a human topology model using a graph neural network to improve the ability to detect local component anomalies. The complex impedance timing data consists of multi-frequency complex impedance data of each measurement loop acquired multiple times at different time points.

[0053] Specifically, a Transformer-Encoder architecture is used to process multi-frequency impedance time-series data from 1kHz to 200kHz. A multi-head attention mechanism is employed to capture cross-frequency band correlation features, including: Input layer: converting impedance amplitude and phase data from 200 frequency points into 512-dimensional embedding vectors; Attention head configuration: an 8-head attention mechanism is used, with 4 heads focusing on low-frequency band (1kHz-50kHz) feature extraction and 4 heads focusing on high-frequency band (50kHz-200kHz) feature extraction; Position encoding optimization: a physically heuristic position encoding based on biological tissue relaxation time is designed, incorporating frequency band physical characteristics into the model's prior knowledge. The position encoding formula is: ,in The relaxation time constant of the measured biological tissue is... For characteristic frequencies, For low-frequency impedance, This is the high-frequency impedance.

[0054] A human topology model is constructed based on anatomical knowledge, and a dynamic graph convolutional network is used to mine the physiological connections between segments. This includes: Human topology graph construction: Based on anatomical knowledge, a graph structure containing 5 segments (left upper limb, right upper limb, left lower limb, right lower limb, and trunk) is constructed. The node features include the extracellular fluid ECW value and intracellular fluid ICW value of each segment, and the edge weights reflect the physiological connection strength between segments; Dynamic graph convolutional network: The GraphSAGE algorithm is used to iteratively update the node representation; Gating mechanism: The gating mechanism dynamically adjusts the information transmission strength between segments, focusing on strengthening the information flow between segments related to local abnormalities.

[0055] The temporal features output by the Transformer-Encoder are concatenated and fused with the topological features output by the graph neural network. Based on the fused features, local component anomaly detection is performed, focusing on identifying pathological states such as abnormal body fluid distribution and abnormal muscle mass in each segment. Finally, the component anomaly probability value and anomaly type label of each segment are output.

[0056] With the above settings, the local edema monitoring module can identify the risk of edema through the phase characteristics of multi-frequency complex impedance, realizing the early detection and grading of edema, and providing strong support for the auxiliary diagnosis of diseases such as chronic kidney disease.

[0057] The metabolic rate calculation module estimates the basal metabolic rate based on Cole-Cole model parameters and user physiological data, obtaining accurate metabolic information without additional equipment, providing a basis for the development of personalized nutrition and exercise programs; the health risk warning module, through multimodal data fusion and artificial intelligence technology, can automatically identify various health risks and provide personalized intervention suggestions, enhancing the system's health management value; the multimodal temporal and topology modeling module combines the advantages of Transformer and graph neural networks, effectively extracting cross-time and cross-site correlation features, significantly improving the detection capability of local component abnormalities.

[0058] To address the issues of traditional least squares fitting of the Cole-Cole model being sensitive to initial values, prone to getting trapped in local optima, and having large fitting errors, this embodiment provides an implementation method.

[0059] In one embodiment, the algorithm fitting module employs an improved flower pollination algorithm, which includes a dual-mode pollination mechanism: performing a global pollination operation based on the Levy flight strategy with a first probability, and performing a local pollination operation based on Gaussian perturbation with a second probability; the Cole-Cole model parameters include low-frequency impedance, high-frequency impedance, characteristic frequency, and dispersion coefficient.

[0060] Specifically, traditional least squares methods for fitting the Cole-Cole model are highly sensitive to initial values ​​and easily get trapped in local optima. An improved flower pollination algorithm is used for global optimization fitting. This algorithm simulates the pollination process of flowers in nature and has advantages such as strong global search capability, fast convergence speed, and good robustness. The iterative optimization process of the improved flower pollination algorithm includes: setting the probability of global pollination (first probability) and the probability of local pollination (second probability); during global pollination, pollen individuals update their positions according to the Lévy flight strategy to expand the search range and escape local optima; local pollination is used to perform a fine search (Gaussian perturbation strategy) within the neighborhood of the current optimal solution, calculating the objective function value of the updated pollen individuals; if the fitness of the updated pollen individuals is better than the current global optimum, then the global optimum is updated; after the iteration terminates, the globally optimal parameter vector is output, which is the optimal parameter of the Cole-Cole model for this measurement loop.

[0061] With the above settings, the improved flower pollination algorithm combines the global search capability of Levy flight with the local fine-grained search capability of Gaussian perturbation, which can quickly converge to the global optimum and greatly improve the fitting accuracy. The dual-mode pollination mechanism can effectively balance the relationship between global exploration and local development, avoid premature convergence of the algorithm, and improve the adaptability to different individual data.

[0062] To address the problem of significant errors in calculating human body composition using torso impedance directly, this embodiment provides an implementation method.

[0063] In one embodiment, the human body composition calculation module uses the Zhu-Moissl fusion equation to calculate the human body composition parameters of each limb segment. The limb segments are calculated using the Zhu equation, and the trunk segments are calculated using the difference between total body water and limb water. The human body composition parameters include extracellular fluid, intracellular fluid, and total water.

[0064] Using the above settings and the Zhu-Moissl fusion equation, which combines the high accuracy of the Zhu equation in limb calculations with the stability of the Moissl equation in whole-body calculations, accurate calculation of the components of each segment of the whole body is achieved. The trunk segment is calculated using the difference between the total body water and the limb water, avoiding the huge error caused by directly using trunk impedance, and significantly improving the accuracy of trunk component parameters.

[0065] like Figure 1 As shown, in order to better utilize the bioimpedance intelligent optimization human composition analysis system, this application also provides a bioimpedance intelligent optimization human composition analysis method, including the following steps:

[0066] S10. Establish measurement circuits for multiple parts of the human body and collect multi-frequency complex impedance data;

[0067] S20. Perform noise reduction and standardization processing on multi-frequency complex impedance data;

[0068] Specifically, the measurement system is calibrated using a reference RC circuit with three known parameters to establish a quadratic nonlinear mapping relationship between the original measured values ​​(multi-frequency complex impedance data) and standard values, generating calibrated impedance data after system error compensation. The original binary data is converted into a standardized CSV file containing fields such as frequency, amplitude, phase, measurement loop number, and timestamp. Data from different measurement loops and at different frequency points undergo unified unit conversion and numerical standardization to generate standardized impedance spectrum data. A key focus is on noise reduction of the phase data in the 50kHz band, as phase abrupt changes in this band are a sensitive indicator of edema. Wavelet transform technology is used to separate the phase response characteristics of different tissues, remove noise interference, and calculate the 50kHz phase change as a core characteristic parameter for local edema monitoring.

[0069] S30. Using multi-frequency complex impedance data as input, perform global optimization fitting on the Cole-Cole model and solve for the corresponding Cole-Cole model parameters.

[0070] Specifically, the Cole-Cole model is a classic mathematical model describing the complex impedance frequency characteristics of biological tissues, and its expression is: ,in, For frequency The complex impedance below, For characteristic frequencies, The dispersion coefficients are used. An improved flower pollination algorithm is employed for global optimization fitting to output the optimal Cole-Cole model parameters.

[0071] S40. Derive the human body composition parameters of each limb segment based on the Cole-Cole model parameters.

[0072] In one embodiment, deriving the human body composition parameters of each limb segment based on the Cole-Cole model parameters includes the following steps:

[0073] Establish a set of impedance relationship equations between limb segments and measurement circuits, and extract low-frequency and high-frequency impedances from the Cole-Cole model parameters;

[0074] Human body composition parameters for each limb segment were calculated based on sex-specific segmental resistivity and in combination with low-frequency and high-frequency impedance. These parameters included extracellular fluid volume, intracellular fluid volume, and total water content.

[0075] In a preferred embodiment, on the left upper limb of the human body Right upper limb , left lower limb Right lower limb An excitation electrode and a detection electrode are respectively set to form a circuit including the left upper limb and the right upper limb. Left upper limb-left lower limb circuit Left upper limb-right lower limb circuit Right upper limb-left lower limb circuit Right upper limb-right lower limb circuit Left lower limb-right lower limb circuit The six measurement loops were established. A set of equations relating the impedance of the limb segments to the measurement loops was constructed: ; ; ; ; ; ,in, Let be the impedance value of the trunk stage (and the impedance values ​​of each segment or loop in the rest). By solving the above system of equations simultaneously, we can obtain the independent impedance value of each segment (including the low-frequency impedance of the j-th segment). and high frequency impedance Based on the impedance-frequency characteristics of biological tissues: ; extracellular fluid resistance was obtained. With intracellular fluid resistance For limb segments, the Zhu equation is used for calculation: ; ; ,in, The extracellular fluid volume of segment j is... Let be the intracellular fluid volume of segment j, TBW be the total water volume, and L be the conductive path length of each segment. For the resistivity of each segment, anatomical standard values ​​based on sex and segment were used. To address the issue of low impedance sensitivity in the trunk, the Zhu-Moissl fusion equation was used to calculate the trunk fluid volume. ; Where ECW is the total extracellular fluid volume, ICW is the total intracellular fluid volume, H is height, and W is weight. and As a coefficient related to BMI, the extracellular fluid volume of the limbs was finally obtained by subtracting the extracellular fluid volume of the limbs from the total extracellular fluid volume. Similarly, the intracellular fluid volume of the somatic stem cells was calculated.

[0076] Based on embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0077] An electronic device includes at least one processor and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the bioimpedance intelligent optimization human composition analysis method provided in this disclosure.

[0078] Electronic devices are intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0079] A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the bioimpedance intelligent optimization human composition analysis method provided in this disclosure.

[0080] The various embodiments of this disclosure can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0081] A computer program product includes a computer program / instructions that are executed by a processor using the bioimpedance intelligent optimization method for human body composition analysis disclosed herein.

[0082] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, 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 machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0083] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0084] The embodiments of the present invention have been described in detail above, but the content described is only a preferred embodiment of the present invention and should not be considered as limiting the scope of the present invention. All equivalent changes and improvements made in accordance with the scope of the present invention should still fall within the patent coverage of the present invention.

Claims

1. A bioimpedance-based intelligent optimized human body composition analysis system, characterized in that, include: The impedance measurement module includes a multi-electrode measurement unit, an analog switch array, and a multi-frequency complex impedance detection unit, which is used to acquire multi-frequency electrical signals applied to human body parts and acquire multi-frequency complex impedance data. The multi-electrode measurement unit is used to form an electrical connection with the human body, and at least one excitation electrode and at least one detection electrode are respectively set at multiple measurement sites on the human body; the analog switch array, all excitation electrodes, all detection electrodes, and the multi-frequency complex impedance detection unit are electrically connected to select different electrode combinations in a time-division manner to form multiple independent measurement circuits; the multi-frequency complex impedance detection unit is used to inject a constant alternating current into the selected measurement circuit and detect the voltage signal across the measurement circuit to calculate the multi-frequency complex impedance data of each measurement circuit; The data preprocessing module is used to perform noise reduction and standardization on the multi-frequency complex impedance data of each measurement loop; The algorithm fitting module is used to perform global optimization fitting of the Cole-Cole model with the multi-frequency complex impedance data of each measurement loop as input, and obtain the Cole-Cole model parameters of each measurement loop. The body composition calculation module is used to calculate the body composition parameters of multiple limb segments based on the Cole-Cole model parameters of each measurement loop. The results output module is used to output the calculated human body composition parameters.

2. The human body composition analysis system according to claim 1, characterized in that: The multi-electrode measurement unit has an excitation electrode and a detection electrode set in the left upper limb, right upper limb, left lower limb, and right lower limb of the human body, forming six measurement circuits including the left upper limb-right upper limb circuit, the left upper limb-left lower limb circuit, the left upper limb-right lower limb circuit, the right upper limb-left lower limb circuit, the right upper limb-right lower limb circuit, and the left lower limb-right lower limb circuit.

3. The human body composition analysis system according to claim 1 or 2, characterized in that, Also includes: The local edema monitoring module is used to calculate the local tissue edema index based on the phase characteristics of the multi-frequency complex impedance data of each measurement circuit, identify the risk of edema, and classify it. The metabolic rate calculation module is used to acquire the user's physiological data and estimate the user's basal metabolic rate based on the Cole-Cole model parameters and physiological data. The health risk warning module is used to fuse complex impedance data, human body composition parameters and physiological data at various frequencies, input them into a pre-trained neural network model, and output health risk classification results and personalized intervention suggestions. The multimodal timing and topology modeling module is used to process complex impedance timing data using the Transformer architecture and to construct a human topology model using a graph neural network to improve the ability to detect local component anomalies. The complex impedance timing data consists of multi-frequency complex impedance data of each measurement loop acquired multiple times at different time points.

4. The human body composition analysis system according to claim 1, characterized in that: The algorithm fitting module adopts an improved flower pollination algorithm, which includes a dual-mode pollination mechanism: a global pollination operation based on the Levy flight strategy is executed with a first probability, and a local pollination operation based on Gaussian perturbation is executed with a second probability; the Cole-Cole model parameters include low-frequency impedance, high-frequency impedance, characteristic frequency, and dispersion coefficient.

5. The human body composition analysis system according to claim 1, characterized in that: The human body composition calculation module uses the Zhu-Moissl fusion equation to calculate the human body composition parameters of each limb segment. The limb segments are calculated using the Zhu equation, while the trunk segments are calculated using the difference between total body water and limb water. The human body composition parameters include extracellular fluid, intracellular fluid, and total water.

6. A bioimpedance-based intelligent optimization method for human body composition analysis, characterized in that, Includes the following steps: Establish measurement circuits for multiple parts of the human body and collect multi-frequency complex impedance data; Denoising and standardization processing is performed on multi-frequency complex impedance data; Using multi-frequency complex impedance data as input, the Cole-Cole model is globally optimized and fitted, and the corresponding Cole-Cole model parameters are obtained by solving the problem. The human body composition parameters of each limb segment were derived based on the parameters of the Cole-Cole model.

7. The method for human body composition analysis according to claim 6, characterized in that, The derivation of human body composition parameters for each limb segment based on the Cole-Cole model parameters includes the following steps: Establish a set of impedance relationship equations between limb segments and measurement circuits, and extract low-frequency and high-frequency impedances from the Cole-Cole model parameters; Human body composition parameters for each limb segment were calculated based on sex-specific segmental resistivity and in combination with low-frequency and high-frequency impedance. These parameters included extracellular fluid volume, intracellular fluid volume, and total water content.

8. An electronic device, comprising: At least one processor; as well as A memory that is communicatively connected to at least one processor; wherein, The memory stores instructions that can be executed by at least one processor to enable the at least one processor to perform the method of any one of claims 6 to 7.

9. A non-transitory computer-readable storage medium storing computer instructions, wherein, Computer instructions are used to cause the computer to perform the method according to any one of claims 6 to 7.

10. A computer program product comprising a computer program / instructions that, when executed by a processor, implement the method according to any one of claims 6 to 7.

Images