Heart disease is one of the reasons to the human death. Detection of heart diseases at early stages is essential for timely intervention and treatment. Deep learning plays an important role in the field of medical science. Recently, there has been great interest into the use of deep learning algorithms, such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and their hybrids, for the purpose of heart disease detection due to the ability of deep learning to process medical data including ECGs, medical imaging, and health records. The ability of CNNs to process images of echocardiograms, CT scans, and MRI scans into real life application comes from its incredible ability to extract features from images. This literature review seeks to evaluate the available literature from the development of deep learning models, specifically with regards to CNNs, to detect heart disease, their architecture, dataset used, barriers in the field, and identified gaps. Lastly, this literature review provides suggestions for further work on improving the scope and efficiency of deep learning algorithms in the cardiovascular sector to upgrade functionality and usability.

CNN, RNN, Hybrid CNN, AlexNet, ResNet-50, Inception-V3, Deep Learning, AI

In conclusion, advancements in AI, CNNs, and DL models have significantly improved the heart disease detection and classification accuracy. Techniques such as hybrid model feature selection, and domain-specific optimization have played key roles in achieving high accuracy across diverse datasets. There is a need for a better CNN approach balances accuracy and efficiency in identifying heart diseases. Developing a new approach is essential to help doctors quickly detect and classify diseases in heart. This new approach will focus on fine-tuning model parameters and using data augmentation to address existing challenges. The review highlights that current algorithms often show low accuracy, emphasizing the importance of continuous improvement. By solving these issues, a new CNN model could greatly enhance the early detection of heart diseases. This gap in research motivates us to create a deep learning model for detecting heart diseases

-