Entropy-based reliable non-invasive detection of coronary microvascular dysfunction using machine learning algorithm

Abstract

<abstract> <sec><title>Purpose</title><p>Coronary microvascular dysfunction (CMD) is emerging as an important cause of myocardial ischemia, but there is a lack of a non-invasive method for reliable early detection of CMD.</p> </sec> <sec><title>Aim</title><p>To develop an electrocardiogram (ECG)-based machine learning algorithm for CMD detection that will lay the groundwork for patient-specific non-invasive early detection of CMD.</p> </sec> <sec><title>Methods</title><p>Vectorcardiography (VCG) was calculated from each 10-second ECG of CMD patients and healthy controls. Sample entropy (<italic>SampEn</italic>), approximate entropy (<italic>ApEn</italic>), and complexity index (<italic>CI</italic>) derived from multiscale entropy were extracted from ST-T segments of each lead in ECGs and VCGs. The most effective entropy subset was determined using the sequential backward selection algorithm under the intra-patient and inter-patient schemes, separately. Then, the corresponding optimal model was selected from eight machine learning models for each entropy feature based on five-fold cross-validations. Finally, the classification performance of <italic>SampEn</italic>-based, <italic>ApEn</italic>-based, and <italic>CI</italic>-based models was comprehensively evaluated and tested on a testing dataset to investigate the best one under each scheme.</p> </sec> <sec><title>Results</title><p><italic>ApEn-</italic>based SVM model was validated as the optimal one under the intra-patient scheme, with all testing evaluation metrics over 0.8. Similarly, <italic>ApEn</italic>-based SVM model was selected as the best one under the intra-patient scheme, with major evaluation metrics over 0.8.</p> </sec> <sec><title>Conclusions</title><p>Entropies derived from ECGs and VCGs can effectively detect CMD under both intra-patient and inter-patient schemes. Our proposed models may provide the possibility of an ECG-based tool for non-invasive detection of CMD.</p> </sec> </abstract>