A 4.88 mu W ECG Delineator Using Wavelet Transform for Mobile Healthcare Application

Abstract

This work presents a low power ECG delineator based on the dyadic wavelet transform. The delineator detects the five most significant ECG fiducial points (P, QRS(on), R, QRS(end), and T waves) for the advanced signal analysis. The detection result is evaluated on two annotated databases including MIT-BIH arrhythmia database and QT database. The obtained sensitivity and positive predictivity are over 99.4% and 96.1% for the five ECG features, respectively. Implemented using UMC 90nm technology with 0.5V supply voltage scaling, duty-cycled operation and power gating technique, the average power dissipation of the delineator is 4.88 mu W for real-time ECG monitoring.