This repo documents my self-study path into computational pathology — notes on papers I've read, and code/experiments as I work through a curriculum I built for myself.
notebooks/— Code and experiments (Jupyter notebooks) as I work through each phase.papers/— Notes and summaries on research papers I've read, with my own takeaways.
I'm following a self-designed curriculum, broken into phases:
- Phase 1 — Pathology-Specific Technical Skills ← currently here
(WSIs, tiling, Multiple Instance Learning, stain normalization, foundation models like UNI/CONCH/Virchow) - Phase 2 — Pathology & Biology Domain Knowledge
(histology, pathology basics, field literacy through papers) - Phase 3 — Applied Compath Projects
(real datasets like Camelyon/TCGA, a full applied project, writeup, outreach to labs)
Phase 1: getting comfortable tiling whole-slide images and understanding why Multiple Instance Learning is the standard approach in this field.