journal-of-information-and-optimization-sciences Journal of Information and Optimization Sciences 2169-0103 0252-2667 Taru Publications 10.47974/JIOS-1808 A hybrid fine-tuned optimizer for enhancing ECG data security in heart attack detection systems Kiran Ajmeera Muma College of Business, 8350 N. Tamiami Trail Sarasota, University of South Florida, Florida, FL 33620, U.S.A. Department of Computer Science and Engineering, MLR Institute of Technology, Hyderabad, Telangana, 500043, India Unhelkar Bhuvan Muma College of Business, 8350 N. Tamiami Trail Sarasota, University of South Florida, Florida, FL 34243, U.S.A. Shankar S. Siva Department of Computer Science and Engineering, KG Reddy College of Engineering and Technology, Hyderabad, Telangana, 500075, India Chakrabarti Tulika Department of Chemistry, Sir Padampat Singhania University, Udaipur, Rajasthan, 313601, lndia Chakrabarti Prasun Department of Computer Science and Engineering, Sir Padampat Singhania University, Udaipur, Rajasthan, 313601, lndia Sivaneasan B. Specialist Adult Educator Engineering, Electrical Power Engineering Programme, Singapore Institute of Technology, 828608, Singapore Margala Martin School of Computing and Informatics, University of Louisiana at Lafayette, LA 70503, U.S.A. 45 8 2309 2323 18 12 2024

Accurate analysis of ECG data is necessary for effective heart attack detection. The InRes-106 model, a hybrid fine-tuned optimizer that improves ECG data security and detection, is presented in this study. It blends deep learning models like ResNet50 and InceptionV3 with sophisticated image processing techniques like artifact removal and Histogram Equalization (HE). The study assesses five pre-trained models: VGG19, DenseNet201, MobileNetV2, and ResNet50 after pre-processing ECG images to enhance quality and remove artifacts before putting out the innovative InRes-106 model. With a remarkable 98.34% accuracy, this model establishes a new standard for heart attack detection and demonstrates improvements in cardiac care.

 Data security Coronary heart disease (CHD) Electrocardiogram (ECG) Heart attack detection access open retracted no