Build · Projects

Projects and experiments

These projects document how I test workflows, explore tooling, and translate pathology questions into code. Some are structured experiments, others are scrappy prototypes, but all are learning exercises.

Each card summarises the problem, who I am building for, and the stack I am currently evaluating so collaborators can see how things evolve.

Work in progress

This site is a personal learning project in pathology informatics.

I am an MD preparing for anatomical pathology residency. I am not a board-certified pathologist or informatics subspecialist.

All notes, cases, and projects are for study and discussion only and must not be used for diagnosis, treatment, or clinical decisions.

ML / Deep Learning

1 project

Reproducible workflow for tiling, training, and reporting attention-based slide classifiers on invasive breast carcinoma.

Started · 11/1/2023
Tags
  • pytorch
  • whole-slide-imaging
  • qa
  • Standardised patch extraction and stain normalisation using reproducible configs.
  • Slide-level attention pooling with integrated attribution overlays for case review.
  • Validation dashboards aligned with CAP/CLIA quality metrics and internal QA checks.