Presented by:
Guillaume Jaume
- Post-doctoral researcher at Harvard Medical School and Brigham and Women's Hospital
[email protected]
Welcome to this tutorial session on Computational Pathology. This hands-on session is created to showcase simple but efficient methods for classifying whole-slide-images (WSIs) of lung cancer. Although illustrated in lung cancer subtyping, similar approaches can be applied to other tasks (e.g., grading, metastasis detection), and sites (e.g., breast, prostate, etc).
This tutorial is largely based on Lu et al., Data-efficient and weakly supervised computational pathology on whole-slide images Nature BME 2021, with code derived from CLAM.
We recommend using Google Colab for running the main notebook. All default Colab packages can be used for this session.
The workshop will take place on the 6th and 7th of July from 9:00 to 18:00 CET.
| Time | Title |
|---|---|
| 9:00-10:30 | Session 1: Intro to QuPath and OpenSlide |
| 10:30-11:00 | Break |
| 11:00-12:30 | Session 2: Multiple Instance Learning (MIL) |
| 12:30-14:30 | Break |
| 14:30-16:00 | Session 3: Interpretability of MIL methods |
| 16:00-16:30 | Break |
| 16:30-18:00 | Session 4: Broader considerations |