QRS detection using a fuzzy neural network

Abstract

We developed a QRS detection algorithm which uses a fuzzy neural network (FNN) to process lead II recordings of the ECG. We trained and tested our algorithm using the MIT/BIH arrhythmia database, and compared our results to existing algorithms. For tapes 100, 105 and 108, our FNN reduced the total number of combined false-positive and false-negative detections from 174 to 44.