Early Disease Diagnosis Using Multivariate Linear Regression

Abstract

The world population is rapidly increasing; people are prone to more diseases due to their food habits and lifestyle changes. The proper diagnosis of the disease at the initial stages will save the lives of millions. So this kind of disease prediction system will help in predicting the disease in the initial stages using their symptoms. We study the historical dataset of the patients with the symptoms and their diseases. We apply data analytics skills to extract hidden patterns from the data. We calculate the number of symptoms contributing to the disease and the weightage factor for every contributing symptom. The weightage indicates the score value contributing to the disease. The ultimate goal of this research is to develop a high-end prediction prototype for disease diagnosis with improved accuracy and efficiency. We implemented a Multivariate Linear Regression algorithm using python to predict the diseases. The model is evaluated using metrics like R2, MSE, and RMSE, and the Multivariate Linear Regression algorithm is found to be the best fit with an accuracy of 95%.