Multiple Disease Prediction Using Machine Learning

Abstract

Abstract: In this comprehensive health analysis project, we delve into the evaluation of Diabetes, heart disease, and Parkinson's disease. Basic health parameters like Pulse Rate, Cholesterol, Blood Pressure, and Heart Rate are scrutinized, enabling the identification of associated risk factors through a prediction model known for its accuracy and precision. The implementation involves leveraging machine learning algorithms, employing Streamlit for interactive interfaces, and employing Python pickling to store model behaviour effectively. Future expansions may encompass diverse health domains such as chronic diseases, skin conditions, and more. The methodology adopts a sophisticated approach, concurrently predicting multiple diseases by synergizing the strengths of XG Boost, K-Nearest Neighbours (KNN), and naïve Bayes (NB) algorithms within a unified framework. This integration aims to capitalize on the complementary attributes of these algorithms, augmenting prediction accuracy and robustness across varied healthcare datasets