Enhancing Myocardial Infarction Diagnosis: Insights from ECG Image Analysis and Machine Learning

• Published in: SN computer science Vol. 5; no. 5; p. 448
• Main Authors: Raghukumar, B. S., Naveen, B.
• Format: Journal Article
• Language: English
• Published: Singapore Springer Nature Singapore 01.06.2024; Springer Nature B.V
• Subjects: Advances in Computational Approaches for Image Processing; Algorithms; Artificial intelligence; Automation; Cardiology; Classification; Classifiers; Cloud Applications and Network Security; Computer Imaging; Computer Science; Computer Systems Organization and Communication Networks; Data Structures and Information Theory; Diagnosis; Electrocardiography; Feature selection; Heart; Heart attacks; Image analysis; Image enhancement; Information Systems and Communication Service; Machine learning; Medical imaging; Medical research; Myocardial infarction; Original Research; Pattern Recognition and Graphics; Scanners; Software; Software Engineering/Programming and Operating Systems; Vision; Wireless Networks; Myocardial infarction; ECG; Feature extraction; Gradient boosting classifier
• ISSN: 2661-8907; 2662-995X; 2661-8907
• DOI: 10.1007/s42979-024-02827-z

This study is dedicated to extracting distinctive features from ECG graph images, which are vital for detecting myocardial infarction or heart attacks due to the variations observed in ECG signal images within ECG report images. These features serve as significant indicators for distinguishing between various cardiac conditions. The utilization of diverse machine learning techniques simplifies and expedites the diagnostic process considerably. These methods leverage the unique features extracted from ECG signal images to make the diagnosis more straightforward and quicker, reducing the time needed for evaluation. The authors identify and employ a set of 10 distinct features extracted from the ECG signal images. These features are subsequently applied to various classification algorithms to evaluate their effectiveness in diagnosing heart conditions. Among the classifiers tested, the Gradient Boosting Classifier (GBC) stands out as the most efficient, achieving an impressive test accuracy rate of 88.79%. This underscores the potential of machine learning in accurately identifying various heart conditions, including myocardial infarction. The findings of this study hold significant clinical importance. The application of the gradient-boosting classifier, along with the extracted features from ECG signal images, demonstrates a promising avenue for enhancing the accuracy and efficiency of heart attack diagnosis. This can be instrumental for medical professionals in making timely and precise diagnoses, potentially leading to improved patient outcomes.