ABSTRACT
Heart disease is one of the leading causes of mortality worldwide. Early detection of heart disease plays a vital role in reducing fatalities and improving patient outcomes. Traditional diagnostic methods rely on manual interpretation of medical data, which can be time-consuming and prone to human error. This project proposes an automated heart disease detection system using Convolutional Neural Networks (CNN), a deep learning model capable of learning complex patterns from medical images. The system utilizes heart image datasets categorized as “affected” and “normal” to train a CNN model capable of classifying unseen images accurately. The model architecture includes multiple convolutional layers, activation functions, pooling layers, and dense layers, optimized to achieve high prediction accuracy. Data preprocessing involves resizing, normalization, and augmentation to enhance model performance and generalization. Users can interact with the system via a web-based interface developed using Flask or a desktop GUI built with Tkinter. The application allows users to upload heart images and receive predictions with confidence scores in real time. The system incorporates a secure login and registration mechanism to maintain user data privacy. During training, the CNN model achieved high validation accuracy, demonstrating its capability to effectively differentiate between affected and normal heart images. Visualization of training progress through accuracy and loss plots confirms model convergence and minimal overfitting. This automated approach provides a reliable, user-friendly solution for heart disease detection, offering potential benefits to healthcare providers and patients. The integration of deep learning techniques into medical diagnostics reduces dependency on manual analysis, improves efficiency, and enables early intervention. Furthermore, the system can be extended with larger datasets, advanced architectures, or integration with IoT devices for real-time monitoring. Overall, this project showcases the potential of combining artificial intelligence and medical imaging to enhance diagnostic accuracy, reduce workload on healthcare professionals, and improve patient care.


OBJECTIVES
1. To develop an automated system for detecting heart disease using medical images.
2. To implement a Convolutional Neural Network (CNN) capable of accurately classifying heart images as “affected” or “normal.”
3. To preprocess and normalize heart images to improve model performance and generalization.
4. To design a CNN architecture with multiple convolutional, pooling, and dense layers optimized for image classification.
5. To train the model using a labeled dataset and evaluate its performance on validation data.
6. To minimize overfitting and improve model reliability using techniques such as dropout, early stopping, and model checkpointing.