基于指端脉搏波视频信号的心率稳定检测算法

目的:研究一种基于脉搏波的稳定检测心率的算法。 方法:提出一种基于快速独立成分分析（FastICA）算法处理指端脉搏波视频信号。首先通过手机摄像头采集手指视频，在每帧图片中提取感兴趣区域（ROI），根据每个区域中像素灰度值的变化得到血液容积变化的时序曲线；然后通过对ROI进行RGB通道分离和FastICA后，分别选取红、绿色分量与盲源分离后的估计信号进行相关性分析，筛选出相关性最大的作为后续提取心率的信号，并与波峰法测得的心率进行对比，得到一种稳定的心率检测算法，并利用SPSS软件做相关性分析。 结果:选取R、G通道信号的一致性在95%以上，基于FastICA的算法与统计波峰法获取心率的一致性在95%以上。 结论:FastICA算法能够有效地提高心率测量的稳定性，实验结果验证了该方法的可行性和有效性，对于基于脉搏波的人体生理参数获取具有重要意义。

An algorithm based on video signals of pulse waves on the finger tip for stable detection of heart rates

Abstract: Objective To propose a pulse wave-based algorithm that can detect the heart rates stably. Methods A fast independent component analysis (FastICA) algorithm is proposed to process the video signals of pulse waves on the finger tip. Firstly, the video of the finger was obtained by cell phone camera; and the region of interest was extracted in each frame of the picture; and the curve of the blood volume changing with time was obtained by investigating the variation of pixel gray values of each frame image. Then, RGB channel segregation and FastICA were conducted on regions of interest. The correlative analysis was performed on R, G channel signals and the estimated signals after blind source separation. The most relevant signals were adopted to obtain heart rates which were then compared with the heart rates measured by wave peak detection. Finally, an algorithm for stably detecting heart rates was obtained, and SPSS Software was utilized for correlation analysis. Results The consistency of the R and G channel signals was higher than 95%. Moreover, the agreement between the algorithm and the wave peak detection was higher than 95%. Conclusion FastICA algorithm can effectively improve the stability of heart rate measurement. The experimental results verify the feasibility and effectiveness of the proposed method which is of great significance for the acquisition of human physiological parameters based on the pulse waves.