Multi Disease Detection Using Machine Learning

Abstract

Machine learning algorithms are widely used to detect the occurrence of diseases in patients. Many of the machine learning algorithms in use today are only concerned with detecting a single disease. No system exists that can detect multiple diseases in a single application, saving patients’ time. Even though there are some systems that can detect a variety of diseases, the models' accuracy varies, which has a significant impact on patients’ quality of life. Building a multi-disease detection system is preferable to deploying numerous applications for each disease when a health organization wants to use machine learning models. It also saves capital. We are using the Ensembling Voting Classifier (Boosting) in Logistic Regression, SVM, Naïve Bayes, and KNN algorithms, and Random Forest which is high-efficient and gives the best accuracy. We are detecting three diseases namely Thyroid, PCOS, and Liver. When a user submits their medical information, we will be able to detect which of three diseases they are likely to be suffering from.