Incorporating intratumoral heterogeneity into weakly-supervised deep learning models via variance pooling
This work addresses a critical challenge in computational pathology for cancer diagnosis and prognosis, though it is incremental as it builds on existing multi-instance learning frameworks.
The paper tackled the problem of predicting cancer survival from whole slide images by incorporating intratumoral heterogeneity into weakly-supervised deep learning models, resulting in improved survival prediction performance for five cancer types as shown in an empirical study with 4,479 images.
Supervised learning tasks such as cancer survival prediction from gigapixel whole slide images (WSIs) are a critical challenge in computational pathology that requires modeling complex features of the tumor microenvironment. These learning tasks are often solved with deep multi-instance learning (MIL) models that do not explicitly capture intratumoral heterogeneity. We develop a novel variance pooling architecture that enables a MIL model to incorporate intratumoral heterogeneity into its predictions. Two interpretability tools based on representative patches are illustrated to probe the biological signals captured by these models. An empirical study with 4,479 gigapixel WSIs from the Cancer Genome Atlas shows that adding variance pooling onto MIL frameworks improves survival prediction performance for five cancer types.