Streamlit-Based AI for Multi-Disease Prediction
Abstract
Health analysis at the right time may prevent and cure diseases. Poor nutrition, less sleep, and less exercise have increased health-related problems. This paper aims to forecast diseases-diabetes, heart disease, breast cancer, and liver diseaseby using machine learning models such as Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, and KNearest Neighbor and deep learning models like Artificial Neural Networks, Convolutional Neural Networks, and Gated Recurrent Units. Logistic Regression ensemble methods, along with Support Vector Machine voting and Random Forest bagging, enhance the predictions. Gradient Boosting is also applied to maximize the accuracy. Explainable AI tools such as LIME and SHAP help understand the contribution of features to understanding clinical decision-making. Integration of this system with Streamlit offers ease of interface toward real-time disease prediction and helps the health professional arrive at decisions with efficiency.