Towards Continuous Electrocardiogram Monitoring Based on Rules and Ontologies

Abstract

Based on rules and ontologies, this paper proposes a framework for predicting types of arrhythmia from electro-cardiogram (ECG) signals acquired using a BSN node. Using terms in an ECG signal ontology, ECG signals are annotated by locating the positions of elementary waves, including their onset, offset, and peak positions. Rules are used for extracting features, e.g., heart rate, PR intervals, RR intervals, and QRS intervals, from annotated signals. An arrhythmia indicator ontology is constructed in order to define concepts representing different characteristics of ECG waveforms, which are then used for defining necessary and sufficient conditions for arrhythmia classification of signal portions. Using standard semantic web ontology and rule languages, i.e., OWL and SWRL, for rule and ontology representation, knowledge content in this framework can be integrated with other existing knowledge sources for retrieval of related information, e.g., recommended treatment.