An antibiotic prediction application MEDIXA is an explainable clinical decision-support system designed to reduce antibiotic failure and slow antimicrobial resistance at the most critical moment: the first prescription. In emergency wards and resource-limited clinics, antibiotics are often prescribed before lab results arrive. A wrong first choice can harm patients and accelerate long-term resistance. MEDIXA addresses this gap by predicting the risk of antibiotic failure before prescription.
🔍 What MEDIXA Does
Evaluates whether a chosen antibiotic is likely to fail for a specific patient Frames the problem as binary risk prediction: Low Risk / High Risk Flags high-risk prescriptions only, avoiding alert fatigue Provides clear, human-readable explanations for every alert
📊 How It Works
MEDIXA uses: Patient history Infection context Prior antibiotic exposure Local resistance patterns The system does not recommend drugs—it supports clinicians by highlighting risk, not replacing judgment.
⚡ Key Features
Explainable AI (clinician-friendly reasoning) Minimal data entry, fast workflow Secure login & clean dashboard Designed for high-pressure, real-world clinical settings
🎯 Impact
Improves first-line antibiotic decisions Reduces unnecessary antibiotic misuse Helps slow antibiotic resistance before it starts “Helping doctors prioritize, predict, and decide — without replacing them.”
cd frontend
npm install
npm run devBackend Setup
cd backend
pip install -r requirements.txt
uvicorn main:app --reload
- Start the Application
Run the frontend and backend servers.
Open the frontend in your browser.
2.You Can login by 2 different users(doctors & Patients)
Feel free to submit issues or pull requests to improve the system.
Team Members: Aasma Gupta, Khushi Singh, Manisha Kumari, Vanshika Chhabra