Diagnosis of Disease Using Wrist Pulse Signal for classification of pre-meal and post-meal samples

摘要

Diagnosis of disease is done by physical examination of patient by physician. For internal observation physician requires help of sonography, MRI, pathological tests reports etc. In Ayurveda Nadi-Pariksha (pulse examination) is used for making the diagnosis. It uses pulse signal sensed at radial artery on wrist below the thumb for diagnosis manually. The pulse signal contains very useful information related to human body. In some cases physician needs patient to tell symptoms of disease. For babies and animals it is not possible to tell the symptoms. To overcome the above mentioned problems a system for `Diagnosis of Disease Using Wrist Pulse Signal' based on Nadi-Pariksha has been investigated. As an example pulse diagnosis has been investigated for detecting whether the person has taken meal or not. Hardware has been developed for recording pulse from wrist. It contains microphone as sensor, amplifier and low pass filter. In proposed method autocorrelation of signal is taken. Using Fast Fourier Transform spectral features are extracted. Support Vector Machine is used for classification of pulse samples. The experimental results show that pulse signal contains important information for detecting whether the person has taken meal or not. With proposed method accuracy of 88.8% for pre-meal signal and of 81.48% for post-meal signal is obtained.