Detection of supraventricular tachycardia using decision tree model. (17th August 2021)

• Record Type: Journal Article
• Authors: Mohanty, Monalisa; Subudhi, Asit; Mohanty, Mihir Narayan
• Is Part Of: International journal of computer applications technology. Volume 65:Number 4(2021)
• Page Start: 378
• Page End: 388
• Publication Date: 2021-08-17
• Subjects: tachycardia -- SVT -- feature extraction -- MLP -- LMT; Technology -- Data processing -- Periodicals; 620.00285
• Languages: English
• ISSNs: 0952-8091

Abstract

Supra Ventricular Tachycardia (SVT) refers to an abnormally fast heartbeat that arises because of the improper electrical activity in the upper chamber of the heart. In this paper, authors have attempted to detect the SVT of human subjects. The ECG recordings have been collected from the MIT-BIH supraventricular arrhythmia database (SVDB) of the Physionet repository. Using Gain Ratio Attribute Evaluation method features are extracted. The evaluated features are then ranked according to their weightage value using the Ranker Search algorithm. The set of features are extracted for ST, N and VF. Machine learning-based classifiers such as Multi-Layer Perceptron (MLP) and Logistic Model Tree (LMT) are utilised to classify the ECG signals from the feature set. It is found that the proposed LMT model outperforms the MLP model and provides 99.23% accuracy. Also, the performance measures are done with sensitivity, specificity and precision as exhibited in the result section.