Volume 1 - Issue 2, May - June 2026

This research project originated from an idea that one could find in many hospitals and even clinics where testing for certain disease involves spending quite a bit of time and resources. In the past, healthcare professionals relied on paperwork and personal experience, so patients were subjected to many lab tests regardless of whether they were necessary or not. The advancement in information technology made it possible to utilize computers in order to provide more efficient assistance. The Problem Statement is to develop software able to propose medical tests based on the set of symptoms (Phase 1) as well as determine disease probability based on lab test results. Phase 2 takes 10 lab features like Hemoglobin, ESR, ALT, Creatinine, etc., and predicts percentages for 6 conditions (Anaemia, TB, CAD, Hepatitis, CKD, Healthy) using multi-output regression. Techniques used include data preprocessing, train-test split, StandardScaler, MultiLabelBinarizer, Random Forest, SVM, GridSearchCV, and weighted ensemble idea. Results are stored in MongoDB with login system. Conclusion is that this tool can give quick guidance and keep history, but it still needs medical checking and bigger real data for stronger trust.

Suraj Kumar S S, Prashant Ankalkoti, "Symptom-Driven Medical Test and Disease Prediction Assistant" International Journal of Computer Science and Engineering Development, V1(2): Page(32-36) May-June 2026. ISSN: 3139-0862. www.ijcsed.com. Published by Scientific and Academic Research Publishing.