Usage of Model Driven Environment for the Classification of ECG features: A Systematic Review

Abstract

Electrocardiography (ECG) constitutes a perfect and primary diagnostic tool for measuring the different morbidity conditions of the heart in the context of different heart diseases and arrhythmia. Various studies have proposed different techniques of classification of ECG features and defined the parametric structure of different features of ECG. This paper is primarily designed to provide a more accurate classification by inducting the concept of a model-driven environment (MDE). Such induction works on the basis of reusable factors that are fitted to the state-of-the-art parametric structure of the ECG features. Some issues and challenges related to the embedding process are highlighted in the form of research questions. The aim of this paper is to provide the solutions to these research questions. The literature review is completed in two phases. In the first phase, those articles are collected that have been published in IEEE Xplore, ACM Library, Science Direct, and Springer, from 2008 to 2017.The second phase as a part of the execution stage is completed at the three different levels of the rectification process by adhering to the Kitchen ham guideline. At the first level, articles are filtered according to title and abstract. At the second level, articles are filtered according to specific eligibility criteria, and at the last level, articles are selected based on the skills of different domain experts (authors) by checking the quality assessment parameters. The significance of MDE in the classification of different ECG features is reflected in their compatibility with the research questions. Furthermore, future directions are proposed that depict the significance of dependencies involvement in classification analysis of ECG. These future directions are identified based on the planning and execution of our operational investigation and our critical observation of the existing gaps between dependencies of features classification that is the major cause of cardiovascular diseases.

Keywords

• Software metrics
• classification algorithms
• electrocardiography
• model driven environment