Myocardial Infarction Diagnosis: Pattern Analysis of ECG Report Images Using Machine Learning Techniques

The ECG machine data is utilized to diagnose cardiac conditions, specifically focusing on identifying myocardial infarction rates by analyzing pattern variations within ECG report images. Variations in the output of electrodes 2 and 3 are noted as indicative of a heart attack. The authors employ various image processing techniques like thresholding, contrast enhancement and machine learning methods like SVM, GBC, k-neighbors to process these patterns, aiming to enhance accuracy. After extracting four features, the most effective classifiers are employed, with the Gradient Boosting Classifier (GBC) with set of features exhibiting the highest accuracy at 76.60%. This paper emphasizes effective preprocessing methods crucial for obtaining structured and refined data, facilitating better feature selection and extraction from ECG graph images. It underscores the extraction of distinctive features to aid in myocardial infarction rate prediction. The paper evaluates several machines learning classifiers, highlighting their efficiency in simplifying and expediting the diagnosis process. Furthermore, the research suggests that incorporating additional features could potentially improve accuracy.