k-NN binary classification of heart failures using myocardial current density distribution maps

Abstract

Magnetocardiography is an advanced technique of measuring weak magnetic fields generated during heart functioning for diagnostics of huge number of different cardiovascular diseases. In this paper, k-nearest neighbor algorithm is applied for binary classification of myocardium current density distribution maps (CDDM). CDDMs from patients with negative T-peak, male and female patients with microvessels (diffuse) abnormalities and sportsmen are compared with normal subjects. Number of neighbors selection for k-NN classifier was performed to obtain highest classification characteristics. Specificity, accuracy, precision and sensitivity of classification as functions of number of neighbors in k-NN are obtained. Depending on group of heart state, accuracy in a range of 80-88%, 70-95% sensitivity, 78-95% specificity and 77-93% precision were achieved. Obtained results are acceptable for further patient's state evaluation.