A denoising strategy for the improvements of PPGi’s signal-to-noise-ratio

Abstract

Abstract Photoplethysmographic imaging (PPGi) is a burgeoning technology to monitor physiological parameters. Under the background of information technologies’ development, both the PPGi signal’s collecting set-ups and its data processing methods have been widely studied. However, the unsatisfactory signal-to-noise-ratio is still an obstacle to its productization. For this problem, an appropriate denoising strategy would be a feasible solution. Therefore, through the analysis and screening of Butterworth Low Pass filter, Butterworth High Pass filter, Median filter, Wavelet Transform, Hilbert-Huang Transform and Independent Component Correlation Algorithm, three of them are combined to form 3 tactics, their effects in improving signal-to-noise-ratio of PPGi signals are evaluated. It is found, the integration of Butterworth Low Pass filter, Median filter and Wavelet Transform (BMW) can acquire the best signal-to-noise-ratio (31.42dB) in the designed strategies. Furthermore, these denoised PPGi signals by the three tactics are used to calculate the blood pressure (BP) and heart rate (HR). The comparisons with the actually BP and HR data which are measured by a commercial sphygmomanometer also indicate that the close-to-actual BP and HR are obtained from the BMW denoised PPGi signals.