Methodological identification of anomalies episodes in ECG streams: a systematic mapping study

Area of Science:

• Cardiology
• Biomedical Engineering
• Medical Informatics

Background:

• Electrocardiograms (ECG) are crucial for diagnosing heart conditions like arrhythmia and myocardial infarction (MI).
• T wave abnormalities in ECG are key indicators for assessing the severity of MI.
• Existing research focuses on T wave classification, but gaps remain in comprehensive analysis.

Purpose of the Study:

• To systematically review and synthesize state-of-the-art methods for T wave classification in ECG for MI detection.
• To identify research questions and provide solutions based on the latest advancements in the field.
• To highlight gaps and propose future research directions for enhanced ECG analysis.

Main Methods:

• A systematic literature review (SLR) following Kitchenham guidance was conducted.
• Articles were collected from IEEE Xplore, Scopus, Science Direct, and Springer (2008-2023).
• A multi-level filtering process, including keyword search, eligibility criteria, and expert quality assessment, was employed.

Main Results:

• The review identified and discussed various state-of-the-art methods for T wave classification.
• These methods provide solutions to pre-defined research questions concerning MI detection via ECG.
• The analysis highlighted the effectiveness of current approaches in ECG interpretation.

Conclusions:

• Critical observations revealed gaps in current T wave classification techniques.
• Future research should focus on advanced feature engineering for ECG.
• Dimensionality reduction and understanding ECG feature dependencies are crucial for improved analysis.