A PERCEPTIVE SURVEY ON THE PREDICTION OF DISEASE DIAGNOSIS USING MACHINE LEARNING, DEEP LEARNING AND ARTIFICIAL INTELLIGENCE

Abstract

Massive volumes of data are routinely handled in the medical industry. Results can be impacted when large amounts of data are handled using traditional approaches. In medical research, machine learning algorithms are particularly useful for obtaining information related to disease prediction. Early prediction of disease is essential to improve the lifestyle of the patients and prevent them from further worsening their condition. For the examination of patient medications and specialists, early disease detection is essential. A variety of diseases are identified using machine learning algorithms, such as ensemble classifier techniques, clustering models, and classification models. Using predictive machine learning techniques can result in high-accuracy and quick illness prediction. This study examines the many sorts of diseases and how they might be predicted using machine learning techniques. This study highlighted the existing research works that mostly explored the prediction of lung problems, brain-related diseases, and heart disease. While deploying conventional model of machine learning and deep learning the problem of overfitting, hyperparameter initialization and unbalanced classes are the major issues which affect the performance of the prediction models.