Evaluating cardiac health through semantic soft computing techniques

Heart Rate Variability (HRV) represents a physiological phenomenon which consists in the oscillation in the interval between consecutive heartbeats. Based on the HRV analysis, cardiology experts can make a assessment for both the cardiac health and the condition of the autonomic nervous system that is responsible for controlling heart activity and, consequently, they try to prevent cardiovascular mortality. In this scenario, one of the most widely accepted and low-cost diagnostic procedures useful for deriving and evaluating the HRV is surely the electrocardiogram (ECG), i.e., a transthoracic interpretation of the electrical activity of the heart over a period of time. With the advent of modern signal processing techniques, the diagnostic power of the ECG is increased exponentially due to the huge number of features that are typically extracted from the ECG signal. Even though this expanded set of features could allow medical staffs to diagnose various pathologies in an accurate way, it is too complex to manage in a manual way and, for this reason, methods for feature representation and evaluation are necessary for supporting medical diagnosis. Starting from this consideration, this paper proposes an enhanced ECG-based decision making system exploiting a collection of ontological models representing the ECG and HRV feature sets and a fuzzy inference engine based on Type-2 Fuzzy Markup Language capable of evaluating the ECG and HRV properties related to a given person and infer detailed information about his health quality level. As will be shown in the experimental section, where the proposed approach has been tested on a set of under exams students, our diagnostic framework yields good performances both in terms of precision and recall.