Publications by authors named "Qiehe Sun"
Shapley Values-enabled Progressive Pseudo Bag Augmentation for Whole-Slide Image Classification.
IEEE Trans Med Imaging
September 2024
Article Synopsis
- Whole-slide image (WSI) classification in computational pathology is challenging due to its high resolution and scarcity of detailed annotations, making it difficult to extract precise instance-level information from broader labels.
- To enhance instance importance estimation, a new method using Shapley values from cooperative game theory is proposed, which improves the accuracy of identifying crucial instances while speeding up the computation using attention mechanisms.
- The proposed framework also focuses on progressively assigning pseudo bags to promote balanced attention distributions in multiple-instance learning (MIL), demonstrating superior performance in experiments on several datasets compared to existing methods, with plans to release the code upon acceptance.
Motivation: With the sites of antigen expression different, the segmentation of immunohistochemical (IHC) histopathology images is challenging, due to the visual variances. With H&E images highlighting the tissue structure and cell distribution more broadly, transferring more salient features from H&E images can achieve considerable performance on expression site agnostic IHC images segmentation.
Methods: To the best of our knowledge, this is the first work that focuses on domain adaptive segmentation for different expression sites.