Detection of Cardiovascular Diseases in ECG Images Using Machine Learning and Deep Learning Methods

Cardiovascular conditions (CVDs) are a major global health challenge and the leading cause of mortality worldwide. Early opinion and effective bracket of these conditions can significantly reduce losses. Electrocardiograms (ECGs), as a readily available and cost-effective individual tool, give critical perceptivity into the electrical exertion of the heart, enabling the identification of CVDs. This exploration focuses on using deep literacy models to prognosticate four distinct cardiac conditions irregular twinkle, myocardial infarction, a history of myocardial infarction, and normal heart function, exercising a intimately accessible ECG image dataset. Originally, transfer literacy was explored using pretrained models similar as SqueezeNet and AlexNet. latterly, a custom- designed convolutional neural network(CNN) was developed to enhance the discovery of cardiac abnormalities. also, these pretrained networks, along with the proposed CNN, were employed as point birth mechanisms for machine literacy algorithms, including support vector machines(SVM), K- nearest neighbors(K- NN), decision trees(DT), arbitrary timbers (RF), and Naïve Bayes(NB). The findings reveal that the proposed CNN armature surpasses being styles, achieving an delicacy of 98.23, a recall rate of 98.22, a perfection score of 98.31, and an F1 score of 98.21. also, when used as a point extractor, the CNN model achieved an emotional delicacy of 99.79 with the Naïve Bayes algorithm. Impact Statement — The integration of artificial intelligence (AI) into healthcare has the implicit to revise complaint discovery, significantly perfecting patient issues. This study introduces a feather light and effective CNN model that achieves 98.23delicacy in classifying cardiovascular conditions using ECG image data. The model operates efficiently on standard computing tackle, making it practical for real- world operations. Likewise, its use as a point birth tool enhances traditional machine literacy ways, delivering a remarkable 99.79 delicacy with the Naïve Bayes classifier. This approach holds pledge for IoT healthcare ecosystems, paving the way for farther exploration in AI- driven cardiovascular diagnostics