Heart Disease Detection Using Neural Network
Authors: Astitwa Srivastava, Dr. Devesh Katiyar
Abstract: Heart-related illnesses continue to be a significant public health concern and a leading cause of premature death worldwide. Prompt and accurate diagnosis plays a vital role in minimizing risk and improving treatment outcomes. This study explores the use of machine learning models, with a focus on a custom-built neural network, to predict heart disease. Using a structured dataset with over 2,500 patient records and 13 clinical features, we trained several classification algorithms, including Logistic Regression, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Decision Tree, and Random Forest. Among these, the proposed neural network achieved the highest accuracy of 92%. The model is deployed using Flask to support real-time prediction, highlighting the real-world utility of such AI-based tools in clinical decision-making systems.
