Democratizing computational pathology: optimized Whole Slide Image representations for The Cancer Genome Atlas
Lazard, T.; Lerousseau, m.; Gardrat, S.; Vincent-Salomon, A.; Stern, M.-H.; Rodrigues, M.; Decenciere, E.; Walter, T.
Show abstract
Automatic analysis of hematoxylin and eosin (H&E) stained Whole Slide Images (WSI) bears great promise for computer assisted diagnosis and biomarker discovery. However, scarcity of annotated datasets leads to underperforming models. Furthermore, the size and complexity of the image data limit their integration into bioinformatic workflows and thus their adoption by the bioinformatics community. Here, we present Giga-SSL, a self-supervised method for learning WSI representations without any annotation. We show that applying a simple linear classifier on the Giga-SSL representations improves classification performance over the fully supervised alternative on five benchmarked tasks and across different datasets. Moreover, we observe a substantial performance increase for small datasets (average gain of 7 AUC point) and a doubling of the number of mutations predictable from WSIs in a pan-cancer setting (from 45 to 93). We make the WSI representations available, compressing the TCGA-FFPE images from 12TB to 23MB and enabling fast analysis on a laptop CPU. We hope this resource will facilitate multimodal data integration in order to analyze WSI in their genomic and transcriptomic context.
Matching journals
The top 7 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- TUGDA: Task uncertainty guided domain adaptation for robust generalization of cancer drug response prediction from in vitro to in vivo settings 96%
- ELISL: Early-Late Integrated Synthetic Lethality Prediction in Cancer 96%
- Predicting Spatially Resolved Gene Expression via Tissue Morphology using Adaptive Spatial GNNs 95%
Similar papers in this journal
- Features fusion or not: harnessing multiple pathological foundation models using Meta-Encoder for downstream tasks fine-tuning 96%
- CellFM: a large-scale foundation model pre-trained on transcriptomics of 100 million human cells 95%
- Generative AI Enables Medical Image Segmentation in Ultra Low-Data Regimes 95%
Similar papers in this journal
- SHEST: Single-cell-level artificial intelligence from haematoxylin and eosin morphology for cell type prediction and spatial transcriptomics reconstruction 96%
- An in-depth comparison of linear and non-linear joint embedding methods for bulk and single-cell multi-omics 96%
- scValue: value-based subsampling of large-scale single-cell transcriptomic data for machine and deep learning tasks 95%
Similar papers in this journal
- COSIME: Cooperative multi-view integration with Scalable and Interpretable Model Explainer 95%
- Inferring spatial single-cell-level interactions through interpreting cell state and niche correlations learned by self-supervised graph transformer 95%
- Generalized Radiograph Representation Learning via Cross-supervision between Images and Free-text Radiology Reports 95%
Similar papers in this journal
- Learning universal knowledge graph embedding for predicting biomedical pairwise interactions 95%
- Multivariate optimization of k for k-nearest-neighbor feature selection with dichotomous outcomes: complex associations, class imbalance, and application to RNA-Seq in Major Depressive Disorder 93%
- GrapHiC: An integrative graph based approach for imputing missing Hi-C reads 93%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.