How lekhan-blaze/oncoaifusion works
OncoAI Fusion is a browser-based medical-image classification platform: users upload an image, the API detects or accepts a cancer type, a corresponding classifier returns predictions and probabilities, and users can request a structured diagnostic report. The report route can augment its template with Groq-generated content. The graph preserves all eight documented classifier services and the main API/UI modules. Edges reflect the supplied source index, excerpts, or README; unsampled modules are included where substantial, but are not assigned speculative connections.
Main entry points: predict.py Prediction endpoints · main.py FastAPI application · App.jsx Diagnostic dashboard · report.py Report endpoints
Components 20
Web interface
- Diagnostic dashboard
frontend/src/App.jsx - React entry
frontend/src/main.jsx
API routing
- FastAPI application
backend/main.py - Health endpoint
backend/app/routes/health.py - Prediction endpoints
backend/app/routes/predict.py - Report endpoints
backend/app/routes/report.py
Image classification
- Brain classifier
backend/app/services/brain_cancer_inference.py - Breast classifier
backend/app/services/breast_cancer_inference.py - Cervical classifier
backend/app/services/cervical_cancer_inference.py - Colon classifier
backend/app/services/colon_cancer_inference.py - Kidney classifier
backend/app/services/kidney_cancer_inference.py - Lung classifier
backend/app/services/lung_cancer_inference.py - Lymphoma classifier
backend/app/services/lymphoma_inference.py - Oral classifier
backend/app/services/oral_cancer_inference.py - Inference service
backend/app/services/inference.py - ResNet50 model
backend/app/models/resnet50.py
Reports and configuration
- LLM report generator
backend/app/services/llm_service.py - Application settings
backend/app/config/settings.py
Other components
- User browser
- Groq API
Connections 16
- User browser → Diagnostic dashboard: uses
README.md - React entry → Diagnostic dashboard: renders
frontend/src/main.jsx - Diagnostic dashboard → FastAPI application: HTTP requests
README.md - FastAPI application → Health endpoint: registers
backend/main.py - FastAPI application → Prediction endpoints: registers
backend/main.py - FastAPI application → Report endpoints: registers
backend/main.py - Prediction endpoints → Brain classifier: dispatches
backend/app/routes/predict.py - Prediction endpoints → Breast classifier: dispatches
backend/app/routes/predict.py - Prediction endpoints → Cervical classifier: dispatches
backend/app/routes/predict.py - Prediction endpoints → Colon classifier: dispatches
backend/app/routes/predict.py - Prediction endpoints → Kidney classifier: dispatches
backend/app/routes/predict.py - Prediction endpoints → Lung classifier: dispatches
backend/app/routes/predict.py - Prediction endpoints → Lymphoma classifier: dispatches
backend/app/routes/predict.py - Prediction endpoints → Oral classifier: dispatches
backend/app/routes/predict.py - Report endpoints → LLM report generator: requests analysis
backend/app/routes/report.py - LLM report generator → Groq API: generates report
backend/app/services/llm_service.py