Research On Intelligent Measurement Technology Of Muscle Mass Based On Bioelectrical Impedance

With the rapid development of the modern global economy,people have gained access to better living conditions.As they strive to improve their quality of life,people have also become more focused on physical health.Body composition parameters,which reflect an individual’s physical condition,have become highly valued.Muscle mass,in particular,is an important parameter that receives attention from a diverse group of people ranging from ordinary individuals to fitness enthusiasts,teenagers,and patients with muscle diseases.Unfortunately,current muscle measurement methods are limited by factors such as low accuracy or the need for expensive instruments and specialized environments.As a result,it can be challenging to provide individuals with convenient muscle measurement options that meet their needs.Bioelectrical impedance analysis(BIA)muscle mass analysis equipment is popular in the market because of its low overall cost,ease of use,and non-invasive measurement advantages.Against this background,this article proposes an intelligent muscle measurement system,which directly evaluates the whole-body and partial muscle mass of the human body through segmented multi-frequency bioimpedance analysis using eight electrodes to detect five parts of the human body,namely the upper limbs,lower limbs,trunk.The main content of the work is as follows:(1)This article uses eight electrodes to detect the left and right upper limbs,lower limbs,and trunk of the human body.High-frequency current(100 Khz)and low-frequency current(20 Khz)are used to collect voltage information from these five parts,and impedance information is obtained through orthogonal demodulation algorithms.Since there are external factors and measurement process interference during electrode measurement,which cause measurement signals to have errors,a Kalman digital filter is used to preprocess the impedance signal.(2)The bioelectrical impedance values of 20 Khz and 100 Khz obtained using the above methods,as well as other parameters such as waist-hip ratio,upper-arm circumference,upper-arm length,leg circumference,leg length,and other human characteristics provided by users,including gender,age,and height are a total of 28 dimensions.By using the Improved Marine Predator Algorithm(IMPA)for feature reduction,the optimal feature subset is obtained,and 15 highly correlated human parameters are selected as input parameters for the prediction model.(3)By using the Improved Sparrow Search Algorithm to optimize the BP Neural Network(ISSABP),the sample’s optimal feature subset is used to establish a human muscle mass prediction model,which is then compared with traditional methods for validation and completes individual muscle mass predictions.Experimental results are compared with other methods,demonstrating the superiority of the muscle mass prediction model based on ISSA-BP.This verifies that the proposed muscle mass measurement method based on ISSA-BP is feasible and can be applied in real-life situations.(4)This article suggests the utilization of networking to establish a muscle mass measurement system based on an already established muscle mass prediction model.The article mainly focuses on the software aspect of the system,specifically on data communication,algorithm implementation,and client display.The feasibility of the proposed system was tested through an experimental verification involving ten individuals with varying physiological characteristics.By comparing the results of the system with the DEXA method,it showed promising detection performance and practical application feasibility for an intelligent muscle mass detection system.