International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 05 | May 2025
p-ISSN: 2395-0072
www.irjet.net
AI Powered Medical Diagnosis and Disease Support System Akshay J1, Anirudh S Bhat2, Bhimesh B3, Gururaj4, Mr. Akhilesh Sathyanarayan5 . 1,2,3,4 Students, Computer Science and Engineering, Jyothy Institute of Technology, Bangalore, India
5 Assistant Professor , Computer Science and Engineering, Jyothy Institute of Technology, Bangalore, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Machine learning's use in healthcare has
patients wait until symptoms develop, even though early disease discovery can significantly enhance treatment outcomes. The requirement for an approachable platform that can provide initial insights based on symptom analysis is what spurred this research.
created new opportunities to support medical diagnosis. In this study, a system that uses user-provided symptoms to forecast diseases is presented. A dataset of symptoms and related disorders is used to train the model using supervised machine learning methods including Support Vector Machine (SVM), Random Forest, and Decision Tree. Users can enter symptoms and get immediate forecasts thanks to an intuitive interface created using Streamlit. By serving as an initial diagnostic tool, this system aims to raise users' awareness of possible health problems. Through the provision of an easy-to-use platform, the system illustrates how machine learning may be incorporated into healthcare processes to enhance health literacy and early diagnosis. Additionally, the modular architecture enables future improvements, enabling it to be tailored to new datasets and medical requirements.
Machine learning models are used to provide customers with quick, data-supported forecasts that encourage additional medical research. Additionally, offering a straightforward web approach lessens dependency on unreliable internet sources that could provide false information. The system encourages users to seek prompt professional medical care by providing a systematic and scientific approach to symptom analysis.
1.2 Objective Create a system that uses machine learning to forecast illnesses based on their symptoms.
Key Words: AI diagnostics, symptom checker, disease prediction, machine learning, healthcare technology, clinical decision support, medical AI, patient triage, Streamlit application, open-source healthcare
1.Train and assess several models, including SVM, Random Forest, and Decision Tree. 2.Use Streamlit to create a web interface that is both interactive and lightweight.
1.INTRODUCTION Medical diagnosis frequently calls for in-depth knowledge and meticulous analysis of clinical findings, patient history, and symptoms. However, machine learning has a chance to help with faster and more accurate disease diagnosis given the abundance of health data already available. The goal of this project is to create a system that predicts diseases based on user-input symptoms using machine learning techniques. The system serves as a first point of reference, directing consumers toward qualified medical assistance as necessary, rather than trying to replace doctors.
3.Give users prompt, recommendations.
5.Allow for quick model upgrades and improvement as new data becomes available.
1.3 Scope Using a preprocessed and structured dataset, this system is intended to make disease predictions based just on symptom input. It does not take the role of thorough clinical evaluation, even though it covers a wide spectrum of common disorders. Although the present version uses a static dataset, it may be extended in the future to include more diseases, symptoms, or individualized health information.
Due to uncertainty or delayed access to medical services, people frequently put off getting medical counsel. Many
Impact Factor value: 8.315
ongoing
6.The ultimate goal is to offer a preliminary diagnostic instrument that can help people comprehend potential health issues.
1.1 Motivation
|
diagnostic
4.Evaluate the performance of the models and choose the best one to implement.
Three machine learning models—SVM, Random Forest, and Decision Tree—were put into practice and their accuracy and efficiency were evaluated. Users may easily interact with the models thanks to the solution's deployment via Streamlit. The strategy focuses on offering a diagnostic tool that is easy to use, scalable, and accessible.
© 2025, IRJET
trustworthy
|
ISO 9001:2008 Certified Journal
|
Page 1078