Pitta Detection Using Second Derivative Features of Finger Photo-Plethysmogram

Abstract

The Human heart beats almost rhythmically, thereby producing pulse wave in different parts of the body. Such a pulse is sensed from finger tips using Photoplethysmography (PPG) technique and is analyzed for its diagnostic ability. The aim of this study is to analyze the features extracted from the second derivative of the pulse waveform and find such parameters that may directly link to the increased level of pitta. Pitta is an Ayurvedic dosha that may be the cause of a disease in the human body. Further it has been observed that pitta level increases after having mid-day meal. In this work the second derivative of the finger PPG waveform has been studied and some prominent features have been extracted from it. The data has been recorded using BIOPAC MP System and AcqKnowledge software. The features thus extracted have been used to find a relation between increased pitta and the Second Derivative of Finger Photo Plethysmogram (SDPTG) wave. For this study data was acquired from 12 subjects. Finger PPG’s of three fingers namely index, middle and ring of both left and right hands were acquired before and after lunch. Eight parameters extracted from each finger were first analysed for six subjects. From a total of 48 parameters 19 such parameters were found that were changing consistently for 5 or 6 subjects. Further analysis was done on so found 19 parameters. From the data analysis on all the 12 subjects for these 19 parameters, 5 parameters were found that consistently held for 10 out of 12 subjects. On these 5 parameters the test of significance, t test for checking the statistical significance of results obtained was applied. Further an Artificial Neural Network (ANN) classifier has been designed that detects the enhanced level of Pitta with an accuracy of 91.67%.