Early classification of pathological heartbeats on wireless body sensor nodes

Abstract

Smart Wireless Body Sensor Nodes (WBSNs) are a novel class of unobtrusive, battery-powered devices allowing the continuous monitoring and real-time interpretation ofa subject’s bio-signals, such as the electrocardiogram (ECG). These low-power platforms,while able to perform advanced signal processing to extract information on heart conditions,are usually constrained in terms of computational power and transmission bandwidth. It istherefore essential to identify in the early stages which parts of an ECG are critical for thediagnosis and, only in these cases, activate on demand more detailed and computationallyintensive analysis algorithms. In this work, we present a comprehensive framework forreal-time automatic classification of normal and abnormal heartbeats, targeting embeddedand resource-constrained WBSNs. In particular, we provide a comparative analysis ofdifferent strategies to reduce the heartbeat representation dimensionality, and thereforethe required computational effort. We then combine these techniques with a neuro-fuzzy classification strategy, which effectively discerns normal and pathological heartbeats witha minimal run time and memory overhead. We prove that, by performing a detailedanalysis only on the heartbeats that our classifier identifies as abnormal, a WBSN systemcan drastically reduce its overall energy consumption. Finally, we assess the choice ofneuro-fuzzy classification by comparing its performance and workload with respect to otherstate-of-the-art strategies. Experimental results using the MIT-BIH Arrhythmia databaseshow energy savings of as much as 60% in the signal processing stage, and 63% in thesubsequent wireless transmission, when a neuro-fuzzy classification structure is employed, coupled with a dimensionality reduction technique based on random projections.