How pras097/brain_tumour works
This is a Flask-based brain-tumour image screening demo. Staff log in, upload an MRI image, and receive a classifier result with a tumour profile and care guidance. The runtime path shown is image preprocessing and feature extraction, followed by scaling and model prediction, then result rendering. The sampled source also shows label-encoder loading and profile lookup. There is no README, and the source excerpt is partial, so the graph includes only relationships visible in the excerpt; notebook and CSV artifacts are not treated as runtime components.
Main entry points: app.py Scan upload and routing · Scaler.pkl Feature scaler · Best_BrainTumor_Model.pkl Tumour classifier · uploads Uploaded scans
Components 13
Staff access
- Staff authentication
app.py
Image screening
- Scan upload and routing
app.py - Uploaded scans
static/uploads - Image preprocessing
app.py - Feature extraction
app.py - Feature scaler
Scaler.pkl - Tumour classifier
Best_BrainTumor_Model.pkl - Label encoder
LabelEncoder.pkl
Results and guidance
- Tumour profiles
app.py - Care guidance
app.py
Web interface
- Login page
templates/login.html - Scan results page
templates/index.html
Other components
- Staff user
Connections 12
- Staff user → Staff authentication: submits credentials
app.py - Staff authentication → Login page: renders
app.py - Staff user → Scan upload and routing: uploads scan
app.py - Scan upload and routing → Uploaded scans: saves image
app.py - Scan upload and routing → Image preprocessing: analyzes image
app.py - Image preprocessing → Feature extraction: supplies grayscale image
app.py - Feature extraction → Feature scaler: transforms features
app.py - Feature scaler → Tumour classifier: supplies scaled features
- Tumour classifier → Label encoder: decodes prediction
- Scan upload and routing → Tumour profiles: looks up prediction
app.py - Tumour profiles → Care guidance: provides care details
app.py - Scan upload and routing → Scan results page: renders result
app.py