基于手机指尖视频的心率提取算法研究OACSTPCD

Objective To propose a cell phone fingertip video-based heart rate extraction algorithm to relieve the influences of environmental factors.Methods Firstly,the fingertip video was captured and extracted at a frame rate of 30 frames/s,and the images were separated into three channels:R,G,and B.The brightness changes of the three channels were compared,the G channel,which was the most sensitive to the fingertip blood flow signal,was selected as the signal source,and the brightness change information was extracted to generate a volumetric pulse waveform with a duration of 10 s.Secondly,the waveform underwent signal processing such as digital filtering,removal of baseline drift and Fourier transform,then the heart rate was predicted based on the position information of the maximum spectral peak.Finally,the heart rate prediction efficacy by the proposed algorithm was validated on Brno University of Technology Smartphone PPG Database(BUT PPG dataset).Results The predicted heart rate by the algorithm and the actual value had the mean square deviation,root mean square deviation and mean absolute error being 3.71,1.92 and 1.2 beats/min,respectively.Condusion The algorithm proposed has high accuracy for heart rate prediction,and can be invovled in cell phones for daily heart rate monitoring.[Chinese Medical Equipment Journal,2024,45(8):16-20]