• Prajul Jain
• Syed Rameez Ali

Cardiovascular diseases (CVDs) remain a leading cause of morbidity and mortality globally. Timely and accurate diagnosis of cardiac diseases is crucial for effective intervention and management. In recent years, deep learning techniques, particularly Convolutional Neural Networks (CNNs), have demonstrated remarkable success in medical image classification tasks. This study focuses on enhancing the accuracy and robustness of cardiac disease classification through the implementation of an ensemble technique in CNN models. Our proposed ensemble model integrates multiple CNN architectures, leveraging the diversity of individual models to improve overall performance. The ensemble consists of distinct CNNs, each trained on diverse subsets of the dataset or with different hyperparameters, enhancing the model's ability to capture various aspects of cardiac abnormalities. The fusion of these diverse models is achieved through a carefully designed aggregation strategy, combining their predictions to yield a more reliable and accurate diagnosis. We employed a comprehensive dataset comprising diverse cardiac images, including angiograms, echocardiograms, and magnetic resonance images. The training process involved optimizing individual CNNs using transfer learning and fine-tuning, considering the unique characteristics of cardiac images. The ensemble model demonstrated superior performance compared to individual CNNs, achieving higher accuracy, sensitivity, and specificity in cardiac disease classification. Furthermore, the ensemble model exhibited robustness against variations in image quality, noise, and different acquisition modalities. The study explores the interpretability of the ensemble model, providing insights into its decision-making process through visualization techniques, aiding clinicians in understanding and trusting the model's predictions. Our findings highlight the potential of ensemble techniques in CNN models for cardiac disease image classification, paving the way for more accurate and reliable diagnostic tools in clinical settings. The proposed approach contributes to the ongoing efforts to leverage artificial intelligence in healthcare, promoting timely and precise diagnosis of cardiac diseases for improved patient outcomes.

Heart disease classification; neural network; ensemble-learning model; under-sampling; features selection; deep learning

"Cardiac Disease Image Classification using Ensemble Technique in CNN Model", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 3, page no.b717-b726, March-2024, Available :http://www.jetir.org/papers/JETIR2403184.pdf

"Cardiac Disease Image Classification using Ensemble Technique in CNN Model", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 3, page no. ppb717-b726, March-2024, Available at : http://www.jetir.org/papers/JETIR2403184.pdf