Diagnosing Multiple Chronic Diseases based on Machine Learning Techniques: Review, Challenges and Futuristic Approach

The usage of various machine learning methods which is a subset of artificial intelligence has revolutionized healthcare organizations by extracting the relevant patterns and determining correlations and associations among multiple features in large databases. It employs the process of performing mathematical operations to find a value to convert empirical data into usable models, reducing the human toll in disease detection to a large extent. In medical practice, machine learning predictive models can show enhanced rules in the decision-making related to individualpatient care. In this paper we have surveyed latest articles which includes open-source dataset like medical images, clinical health records, laboratory test results and many more of various diseases such as heart attack, asthma, diabetes, dengue, allergy, AIDS, vertigo, acne, fungal infection etc. from different researches published on Excerpta Medical Database Google Scholar, WOS, SSRN, PubMed and Scopus up to January 2022. Hence, the results are analyzed on the basis of certain quality parameters and to successfully apply ML techniques in disease diagnosis, a number of issues and challenges must be addressed in future.