Multi-disease Prediction using Machine Learning

Authors

G. Sangeetha, K. Meenatchi

Abstract

Data mining for healthcare is an interdisciplinary field of study that has its origins in database statistics and may be used to analyse the success of medical treatments. Many present machine learning models for health care analysis focus on a particular condition at a time. For example, one analysis may be for diabetes, another for thyroid problems, and still another for cancer illnesses. There is no general method for forecasting many illnesses using a single analysis. This project provides a system that forecasts various illnesses using Python Flask API. This inquiry made use of diabetes analysis, thyroid analysis, and breast cancer analysis. To implement various sickness analysis, machine learning algorithms, tensor flow, and the Flask API were employed. Pickling in Python is used to store model behavior, whereas unpicking in Python is used to load the pickle file. The relevance of this research is that it evaluates disorders and includes all the characteristics that cause the condition, allowing the disease's maximum impact to be detected. We undertake a thorough search of all known feature variables within the KAGGLE dataset to construct models for cardiovascular, prediabetes, and diabetes identification. Using several time-frames and feature sets for the data (based on laboratory data), the Support Vector Machine algorithm is used to forecast illnesses with greater accuracy.

Keywords

Multi-disease Prediction using Machine Learning. G. Sangeetha, K. Meenatchi. 2023. IJIRCT, Volume 9, Issue 5. Pages 1-3. https://www.ijirct.org/viewPaper.php?paperId=2309003