Multi-Task Learning in Histo-pathology for Widely Generalizable Model
This work addresses the problem of model generalization in computational pathology for medical researchers and clinicians, but it is incremental as it builds on existing multi-task learning approaches without introducing a new paradigm.
The authors tackled the challenge of developing a widely generalizable model for computational pathology by applying deep multi-task learning across 11 diverse tasks, including oral cancer classification and nuclei segmentation, but only reported preliminary results without concrete performance numbers.
In this work we show preliminary results of deep multi-task learning in the area of computational pathology. We combine 11 tasks ranging from patch-wise oral cancer classification, one of the most prevalent cancers in the developing world, to multi-tissue nuclei instance segmentation and classification.