Blood Cancer Prediction — Flask GUI

📁 Folder Structure

gui/
├── app.py                  ← Main Flask application
├── requirements.txt        ← Python dependencies
├── templates/
│   └── index.html          ← Web interface
└── model/
    └── blood_cancer_model.keras  ← PUT YOUR MODEL FILE HERE

🚀 How to Run

1. Install dependencies

pip install -r requirements.txt

2. Add your model

Copy blood_cancer_model.keras (from Person 2) into the model/ folder.

3. Run the app

python app.py

4. Open in browser

Visit: http://127.0.0.1:5000

---

⚙️ Class Labels

The model predicts 4 classes (update in app.py if your dataset folders differ):

• Benign — Normal blood cells
• Early Pre-B ALL — Early stage leukemia
• Pre-B ALL — Pre-B leukemia
• Pro-B ALL — Pro-B leukemia

If your dataset folders have different names, update CLASS_LABELS in app.py to match the alphabetical order of your dataset subfolders.

---

🔗 Integration Notes

• Preprocessing: Images are resized to 224×224 and rescaled by 1/255 — matching training exactly.
• Model input shape: (1, 224, 224, 3)
• Model output: Softmax over 4 classes