Melanoma detection machine learning model for EECS 582 (Computer Science Capstone), Spring 2025.
This was a collaborative capstone project. I served as a team lead responsible for coordinating project artifacts, tracking deadline readiness, and communicating with our supervisor to make sure deliverables stayed aligned with course and project expectations.
This fork is included on my GitHub as portfolio evidence of that team leadership and project coordination work, not as a claim of sole ownership over the full model, backend, or interface implementation.
- Coordinating a machine learning capstone with documentation, artifacts, and deadlines.
- Building a Flask interface around an image-classification model.
- Training and packaging a TensorFlow/Keras melanoma classification prototype.
- Documenting model architecture, preprocessing decisions, limitations, and medical-risk disclaimers.
- Maintaining project artifacts such as model parameters, progress logs, and team-facing documentation.
| Path | Purpose |
|---|---|
backend/ |
Flask application, templates, upload flow, image normalization, and packaged model artifact. |
backend/templates/ |
Web pages for the prototype interface, education content, contact page, and disclaimer. |
model/ |
Model training/testing scripts, parameters, and progress tracking CSV. |
docs/ |
Model tracker and TensorFlow/preprocessing documentation. |
LICENSE |
MIT license inherited from the upstream project. |
docs/ARCHITECTURE.md: backend, inference, and training flow.docs/MODEL_CARD.md: intended use, limitations, data notes, and evaluation context.docs/PROJECT_GOVERNANCE.md: team-lead responsibility and contribution framing.SECURITY.md: upload, privacy, and repository hygiene risks.
This project is for educational and research purposes only. By using this tool, you acknowledge and accept all associated risks.
No security or privacy systems are in place. User uploads are at your own risk.
For full details, see the melanoma_model_tracker.md file in the docs folder.
- Some preprocessing and normalization steps/documentation notes are adapted from the TensorFlow image loading tutorial.
The dataset used in this project is provided by the International Skin Imaging Collaboration (ISIC) Link: ISIC Archive