Weakly Supervised Multi-Task Learning for Cell Detection and Segmentation
This work addresses a critical bottleneck in digital pathology analysis by reducing annotation burden for researchers and clinicians, though it is incremental as it builds on existing weakly supervised and multi-task learning approaches.
The paper tackled the labor-intensive problem of obtaining pixel-level ground truth for single cell segmentation in digital pathology by developing an end-to-end deep learning algorithm that uses only point labels for both detection and segmentation, showing significant improvement over state-of-the-art methods without increasing annotation efforts.
Cell detection and segmentation is fundamental for all downstream analysis of digital pathology images. However, obtaining the pixel-level ground truth for single cell segmentation is extremely labor intensive. To overcome this challenge, we developed an end-to-end deep learning algorithm to perform both single cell detection and segmentation using only point labels. This is achieved through the combination of different task orientated point label encoding methods and a multi-task scheduler for training. We apply and validate our algorithm on PMS2 stained colon rectal cancer and tonsil tissue images. Compared to the state-of-the-art, our algorithm shows significant improvement in cell detection and segmentation without increasing the annotation efforts.