Multiple Disease Prediction Using Machine Learning

Abstract

There are several techniques in machine learning that can perform predictive analytics on large amounts of data across industries. Predictive analytics in healthcare is a difficult task, but it can ultimately help practitioners in making timely decisions regarding the health and treatment of patients based on massive data. Diseases such as breast cancer, diabetes and heart disease outbreaks cause many deaths worldwide, but most of these deaths are due to a lack of early disease control. The above problem occurs due to inadequate medical infrastructure and low ratio of doctors to population. Statistics clearly show the same, WHO advises, doctor to patient ratio is 1:1000, while doctor to population ratio in India is 1:1456, this shows shortage of doctors. Diseases related to heart, cancer and diabetes can pose a potential threat to humanity if not detected early. Therefore, early recognition and diagnosis of these diseases can save many lives. This thesis is all about predicting diseases that are harmful using machine learning classification algorithms. Breast cancer, heart and diabetes are included in this work. To make this work seamless and usable by the general public, our team has created a medical test web application that predicts various diseases using the concept of machine learning. In this work, we aim to develop a machine learning-based prediction concept for various diseases such as breast cancer, diabetes, and heart disease.