Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Chikontwe, Philip | - |
dc.contributor.author | Sung, Hyun Jung | - |
dc.contributor.author | Jeong, Jaehoon | - |
dc.contributor.author | Kim, Meejeong | - |
dc.contributor.author | Go, Heounjeong | - |
dc.contributor.author | Nam, Soo Jeong | - |
dc.contributor.author | Park, Sang Hyun | - |
dc.date.accessioned | 2022-11-07T08:30:13Z | - |
dc.date.available | 2022-11-07T08:30:13Z | - |
dc.date.created | 2022-06-29 | - |
dc.date.issued | 2022-08 | - |
dc.identifier.issn | 1361-8415 | - |
dc.identifier.uri | http://hdl.handle.net/20.500.11750/17060 | - |
dc.description.abstract | In digital pathology, segmentation is a fundamental task for the diagnosis and treatment of diseases. Existing fully supervised methods often require accurate pixel-level annotations that are both time-consuming and laborious to generate. Typical approaches first pre-process histology images into patches to meet memory constraints and later perform stitching for segmentation; at times leading to lower performance given the lack of global context. Since image level labels are cheaper to acquire, weakly supervised learning is a more practical alternative for training segmentation algorithms. In this work, we present a weakly supervised framework for histopathology segmentation using only image-level labels by refining class activation maps (CAM) with self-supervision. First, we compress gigapixel histology images with an unsupervised contrastive learning technique to retain high-level spatial context. Second, a network is trained on the compressed images to jointly predict image-labels and refine the initial CAMs via self-supervised losses. In particular, we achieve refinement via a pixel correlation module (PCM) that leverages self-attention between the initial CAM and the input to encourage fine-grained activations. Also, we introduce a feature masking technique that performs spatial dropout on the compressed input to suppress low confidence predictions. To effectively train our model, we propose a loss function that includes a classification objective with image-labels, self-supervised regularization and entropy minimization between the CAM predictions. Experimental results on two curated datasets show that our approach is comparable to fully-supervised methods and can outperform existing state-of-the-art patch-based methods. https://github.com/PhilipChicco/wsshisto © 2022 Elsevier B.V. | - |
dc.language | English | - |
dc.publisher | Elsevier BV | - |
dc.title | Weakly supervised segmentation on neural compressed histopathology with self-equivariant regularization | - |
dc.type | Article | - |
dc.identifier.doi | 10.1016/j.media.2022.102482 | - |
dc.identifier.wosid | 000871059300009 | - |
dc.identifier.scopusid | 2-s2.0-85131945996 | - |
dc.identifier.bibliographicCitation | Medical Image Analysis, v.80 | - |
dc.description.isOpenAccess | FALSE | - |
dc.subject.keywordAuthor | Class activation map | - |
dc.subject.keywordAuthor | Compressed histopathology | - |
dc.subject.keywordAuthor | Deep learning | - |
dc.subject.keywordAuthor | Image segmentation | - |
dc.subject.keywordAuthor | Weakly supervised learning | - |
dc.citation.title | Medical Image Analysis | - |
dc.citation.volume | 80 | - |
dc.description.journalRegisteredClass | scie | - |
dc.description.journalRegisteredClass | scopus | - |
dc.relation.journalResearchArea | Computer Science; Engineering; Radiology, Nuclear Medicine & Medical Imaging | - |
dc.relation.journalWebOfScienceCategory | Computer Science, Artificial Intelligence; Computer Science, Interdisciplinary Applications; Engineering, Biomedical; Radiology, Nuclear Medicine & Medical Imaging | - |
dc.type.docType | Article | - |
There are no files associated with this item.