2 - Heart rate variability

Abstract

Analysis of heart rate variability (HRV) aids in understanding the factors affecting the heart rhythm. HRV parameters deal with linear as well as nonlinear parameters while analyzing sympathetic and parasympathetic nervous systems. Information about cardiac diseases is derived from the heart rate (HR) variation, an important indicator. As the abnormality progresses, the frequency of HRV increases for long-duration electrocardiogram (ECG) recordings. Manual interpretation and classification of the abnormalities from long-duration ECG datasets is a laborious task. Therefore, analysis of HRV over time has turned out to be one of the most potent noninvasive indicators for quantifying the autonomic nervous system's activity. Hence, the HRV signal parameters are extracted and evaluated by numerous computing techniques helpful in predicting and diagnosing cardiac disorders. This chapter assesses HRV features in time and frequency-domain, its techniques, and applications for identifying, classifying, and predicting various cardiac abnormalities. This chapter also presents the variation of HRV parameters in sinus rhythm (SR), sinus tachycardia (ST) volunteers, and during different menstrual cycle phases. A machine learning classifier, extra trees (ET), has been employed to classify SR, ST ECG signals based on HRV parameters. Effect of hormonal fluctuations, during the menstrual cycle, on HRV time and frequency-domain parameters on a large population will provide a more comprehensive approach.