Olga Fourkioti, Matt De Vries, Chen Jin, Daniel C. Alexander, Chris Bakal
The visual examination of tissue biopsy sections is fundamental for cancer diagnosis, with pathologists analyzing sections at multiple magnifications to discern tumor cells and their subtypes. However, existing attention-based multiple instance learning (MIL) models used for analyzing Whole Slide Images (WSIs) in cancer diagnostics often overlook the contextual information of tumor and neighboring tiles, leading to misclassifications. To address this, we propose the Context-Aware Multiple Instance Learning (CAMIL) architecture. CAMIL incorporates neighbor-constrained attention to consider dependencies among tiles within a WSI and integrates contextual constraints as prior knowledge into the MIL model. We evaluated CAMIL on subtyping non-small cell lung cancer (TCGA-NSCLC) and detecting lymph node (CAMELYON16 and CAMELYON17) metastasis, achieving test AUCs of 97.5\%, 95.9\%, and 88.1\%, respectively, outperforming other state-of-the-art methods. Additionally, CAMIL enhances model interpretability by identifying regions of high diagnostic value.
| Task | Dataset | Metric | Value | Model |
|---|---|---|---|---|
| Multiple Instance Learning | CAMELYON16 | ACC | 0.917 | CAMIL |
| Multiple Instance Learning | CAMELYON16 | AUC | 0.959 | CAMIL |
| Multiple Instance Learning | CAMELYON16 | ACC | 0.91 | CAMIL (CAMIL-L) |
| Multiple Instance Learning | CAMELYON16 | AUC | 0.953 | CAMIL (CAMIL-L) |
| Multiple Instance Learning | CAMELYON16 | ACC | 0.891 | CAMIL (CAMIL-G) |
| Multiple Instance Learning | CAMELYON16 | AUC | 0.95 | CAMIL (CAMIL-G) |