README.md

AI-Powered Differential Diagnosis (Educational)

What this Space does

• You enter at least 3 comma-separated symptoms.
• The app compares your input to a dataset using TF-IDF similarity.
• It returns a differential list with scaled scores (0..1), example matches, and structured reasoning outlines (deterministic).

Files to upload to your Hugging Face Space

• app.py (this file)
• requirements.txt
• differential_diagnosis_dataset.csv (your dataset)
• (Optional) README.md

Dataset assumptions

• One column named diagnosis (lower/upper case both ok).
• Symptom columns contain the word symptom, feature, sign. If none found, all non-diagnosis columns are treated as symptoms.

Run locally

pip install -r requirements.txt
python app.py

Open http://127.0.0.1:7860

Safety

• This tool is for education and prototyping.
• It is not a substitute for a clinician.
• For urgent issues, contact emergency services.