Development And Experimental Research Of Low-Cost Body Composition Analyzer

Human body composition refers to the percentage of fat,protein,minerals,water and other contents in body weight.The body composition analyzer currently on the market is expensive and bulky,which is not conducive to the popularity of home users.Aiming at this problem,this thesis designed and developed a low-cost body composition analyzer based on Bioelectrical Impedance Analysis(BIA);using the body impedance measurement analog front-end chip CS1258 provided by the CHIPSEA company.The main research contents are as follows:1.Systematic review of the body composition measurement method,body composition measurement technology based on the principle of bioelectrical impedance,the research status of the body composition analyzer,the development history of the body composition modeling theory and the development status at home and abroad.2.Designed a human body composition analyzer based on the chip CS1258.The instrument adopts the cight-elcetrode mcasurement method,and uses STM32F407 as the control core to control CS1258 to complete the measurement of human impedance at six frequencies of 5 k,10 k,25 k,50 k,100 k and 250 kHz.At the same time,the instrument is equipped with chip HX711 and four weight measurement modules of 50 kg steel weighing sensors to realize weight measurement.The lower computer in the software design co’mpletes the controt function of the hardware measurement platform through the modular programming of C language.The upper computer uses G language to complete the modeling of the human body component data,and the intuitive display and storage of the results.3.The instrument designed in this thesis is used for experimental research.Firstly,the impedance measurement is performed on the constructed three-element model to complete the system calibration.The relative error of the three-element amplitude is less than 1.2%,and the absolute error of the phase angle is less than 1.6°.Secondly,the instrument designed in this thesis,the standard body composition analyzer Imp SFB7 and the weighing-machine were used to conduct impedance and weight measurement experiments on 40 subjects.The parison results show that the relative error of the instrument’s impedance amplitude is less than 3.9%,and the absolute error of the phase angle is less than 2.2°,The relative error of body weight is within 3.1%,and the repeatability test results are good.4.Completed the establishment of body composition prediction model.The body impedance and body composition data measured by the standard instrument(SFB7)are used as the sample library of the model,Two body composition prediction models based on the BP neural network algorithm are established by the LabVIEW.Completed the function of predicting five body composition parameters of total body water(TBW),extracellular fluid(ECW),intracellular fluid(ICW),adipose tissue(FM)and non-adipose tissue(FFM).Using the instrument designed in this thesis,the average value of the relative error of the prediction model realized by the MATLAB Script node measured by the body composition parameters is 5.11%;the average error of the prediction model realized by the MLT toolkit is 4.6%and the mean absolute error and mean square error being small overall.