Development of Handheld Cardiac Event Monitoring System

This paper contributes the development, prototyping and analysis of proposed methodology on ARM (Advanced RISC Machine) in laboratory for automatic detection of arrhythmia beat in real-time for diagnosis of cardiovascular diseases. The methodology involves the integration of R peak detection algorithm, Principal Component Analysis for feature extraction and feedforward neural network architecture to classify generic heartbeats into six classes. The proposed methodology is implemented on ARM-based SoC (System-on-Chip) platform for diagnosis of six heartbeats. This developed system is validated by generating realtime ECG beats using MIT-BIH database and the output of the proposed system is monitored in the displaying device. The performance metrics of the developed system yields an overall accuracy of 92.81% with average sensitivity, specificity and positive predictivity of 92.68%, 98.51% and 92.42% respectively. Moreover, the developed system can be fabricated into a handheld device for automatic ECG beat monitoring.