Disease Prediction: A Case Study for Healthcare Communities Using Machine Learning

A disease is a condition, that negatively act on the function of all or a part of the organs in the human producing symptoms that affects internally and are not immediately caused due to any physical injury. With tremendous growth in bigdata in the field of medical diagnostic and healthcare communit...

Full description

Saved in:

Bibliographic Details
Published in:	2022 IEEE 2nd Mysore Sub Section International Conference (MysuruCon) pp. 1 - 6
Main Authors:	Karthik, M, Anuradha, S G, Raghu Kumar, K S, Srusti, U M, Sanjeev Kumar, S
Format:	Conference Proceeding
Language:	English
Published:	IEEE 16-10-2022
Subjects:	Bigdata Analysis Healthcare History Machine Learning Medical diagnosis Medical diagnostic imaging Medical Diagnostics Predictive models Random forests Support vector machines Training
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	A disease is a condition, that negatively act on the function of all or a part of the organs in the human producing symptoms that affects internally and are not immediately caused due to any physical injury. With tremendous growth in bigdata in the field of medical diagnostic and healthcare communities demands accurate analysis of the medical data facilitates early disease diagnosis, patient care and in community services to reduce the global mortality rate. However, Machine Learning models seeks to replicate human nature of healthcare system and researchers now a days are facing challenges in the medical field to analyze the substantial amount of data, administer the patient history and further diagnose the disease in the early stage. We propose a robust machine learning model that can efficiently predict the disease associated with patient based on his/her symptoms. The key contribution of our work is to build an integrated machine learning model with Support Vector Machine, Random Forest, Multi-Layer Perceptron, Naïve Bayes, K-Nearest Neighbour, Decision Tree and predict a disease. The experimental work is carried on the medical dataset with 132 symptoms recorded and classified into 42 disease categories respectively. The results reveal that the combined model outperforms the training and testing accuracy with the result of 100%.
DOI:	10.1109/MysuruCon55714.2022.9972697