HealthWithSevgi — ML Visualization Tool for Healthcare

SENG 430 · Software Quality Assurance Çankaya University · Spring 2025-2026 Instructor: Dr. Sevgi Koyuncu Tunç

An interactive, browser-based ML learning tool that guides healthcare professionals through a 7-step pipeline — from choosing a medical specialty to training an AI model and checking it for fairness — with no coding required.

🏥 What It Does

Step	Name	Description
1	Clinical Context	Introduces the medical problem the AI will tackle
2	Data Exploration	Upload patient data (CSV) or use a built-in dataset
3	Data Preparation	Handle missing values, normalise, split train/test
4	Model & Parameters	Pick one of 6 ML models and tune via sliders
5	Results	Accuracy, sensitivity, specificity, confusion matrix, ROC
6	Explainability	Feature importance + single-patient SHAP explanations
7	Ethics & Bias	Subgroup fairness audit + EU AI Act compliance checklist

20 clinical domains supported (Cardiology, Nephrology, Oncology, Neurology, Diabetes, Pulmonology, Sepsis/ICU, Fetal Health, Dermatology, Stroke Risk, and more).

6 ML models: KNN, SVM, Decision Tree, Random Forest, Logistic Regression, Naïve Bayes.

🛠 Tech Stack

Layer	Technology
Frontend	React 18 + Vite
Backend	FastAPI (Python)
ML Engine	scikit-learn
Styling	CSS (DM Sans / DM Mono / Fraunces)
Diagrams	Figma + draw.io
PM	Jira

📁 Repository Structure

HealthWithSevgi/
├── frontend/                # React 18 + Vite application
│   ├── public/
│   ├── src/
│   │   ├── components/      # Reusable UI components
│   │   ├── pages/           # Step pages (Step1–Step7)
│   │   ├── hooks/           # Custom React hooks
│   │   ├── utils/           # Helper functions
│   │   ├── styles/          # Global styles & theme
│   │   ├── assets/          # Static assets (icons, images)
│   │   ├── App.jsx
│   │   └── main.jsx
│   ├── package.json
│   └── vite.config.js
│
├── backend/                 # FastAPI REST API
│   ├── app/
│   │   ├── main.py          # FastAPI entry point
│   │   ├── routers/         # API route modules
│   │   ├── services/        # ML training, preprocessing, prediction
│   │   ├── models/          # Pydantic schemas
│   │   └── utils/           # Helpers (certificate gen, etc.)
│   ├── datasets/            # Built-in clinical datasets (CSV)
│   ├── tests/               # pytest test suite
│   ├── requirements.txt
│   └── Dockerfile
│
├── docs/                    # Reference documents & design specs
│   ├── ML_Tool_User_Guide.docx
│   ├── GENERAL_DESIGN_SAMPLE_HTML.html
│   └── 6_MODEL_VISUALISATION_PAGES__SAMPLE_HTML.html
│
├── .github/                 # GitHub templates & workflows
│   └── pull_request_template.md
│
├── .gitignore
├── SETUP.md                 # Local development setup guide
└── README.md                # ← You are here

🚀 Quick Start

See SETUP.md for full instructions.

# Backend
cd backend
python -m venv venv && source venv/bin/activate
pip install -r requirements.txt
uvicorn app.main:app --reload --port 8000

# Frontend
cd frontend
npm install
npm run dev

🌿 Branch Strategy

Branch	Purpose
main	Production-ready, protected
develop	Integration branch for sprint work
feature/US-XXX	One branch per user story

Rules:

• All changes go through Pull Requests
• PRs require at least 1 approval
• main and develop are protected — no direct pushes

👥 Team

Role	Name	Student ID
Product Owner + Developer	Efe Çelik	202128016
UX Designer	Burak Aydoğmuş	202128028
Lead Developer + Scrum Master	Batuhan Bayazıt	202228008
Developer	Berat Mert Gökkaya	202228019
QA / Documentation Lead	Berfin Duru Alkan	202228005

📋 Links

• Jira Board: [TBD]
• Figma Designs: [TBD]
• GitHub Wiki: Wiki →
• API Docs: http://localhost:8000/docs (when running locally)

📄 License

This project is developed as part of the SENG 430 course at Çankaya University. All rights reserved.