Thread by @fayyazhere

"Whole slide images are graphs": Our paper on effectiveness of graph modelling of WSIs in #cpath & Graph Neural Networks (#GNN) for receptor status prediction of #breastcancer from routine WSIs is accepted in Medical Image Analysis.
Preperint: arxiv.org/abs/2110.06042

TL;DR
1. Motivation: Whole slide images are big and holistic modelling of interactions between different tissue components using only slide level labels is difficult with conventional "patch-then-aggregate" approaches used for weakly supervised learning in CPath.

2. We introduce a flexible #GNN based #ML framework (#SlideGraph+) that can holistically model WSIs for different large scale prediction problems
3. For HER2 status prediction, the method gives AUC 0.75-0.8 and can be used for case triaging & advanced ordering of tests.

4. To demonstrate the flexibility of SlideGraph+, our @tiatoolbox implementation uses different local features For ER status predictionwith an AUC of 0.88
5. Once you extract local features, the method is lightning fast allowing large scale experiments (<1 hour for TCGA-BRCA)!

A (slightly dated) video on "Whole Slide Images are Graphs" is available at:
A big thank you to Wenqi Lu, Michael Toss, M. Dawood, Emad Rakha, @nmrajpoot, @TIAwarwick & @sea_raza for integration into @tiatoolbox and @WarwickDCS.

Unrolled: threadreaderapp.com/thread/1528397…