SegMatch: semi-supervised surgical instrument segmentation
Abstract Surgical instrument segmentation is recognised as a key enabler in providing advanced surgical assistance and improving computer-assisted interventions. In this work, we propose SegMatch, a semi-supervised learning method to reduce the need for expensive annotation for laparoscopic and robo...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-04-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-025-94568-z |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850172456779644928 |
|---|---|
| author | Meng Wei Charlie Budd Luis C. Garcia-Peraza-Herrera Reuben Dorent Miaojing Shi Tom Vercauteren |
| author_facet | Meng Wei Charlie Budd Luis C. Garcia-Peraza-Herrera Reuben Dorent Miaojing Shi Tom Vercauteren |
| author_sort | Meng Wei |
| collection | DOAJ |
| description | Abstract Surgical instrument segmentation is recognised as a key enabler in providing advanced surgical assistance and improving computer-assisted interventions. In this work, we propose SegMatch, a semi-supervised learning method to reduce the need for expensive annotation for laparoscopic and robotic surgical images. SegMatch builds on FixMatch, a widespread semi-supervised classification pipeline combining consistency regularization and pseudo-labelling, and adapts it for the purpose of segmentation. In our proposed SegMatch, the unlabelled images are first weakly augmented and fed to the segmentation model to generate pseudo-labels. In parallel, images are fed to a strong augmentation branch and consistency between the branches is used as an unsupervised loss. To increase the relevance of our strong augmentations, we depart from using only handcrafted augmentations and introduce a trainable adversarial augmentation strategy. Our FixMatch adaptation for segmentation tasks further includes carefully considering the equivariance and invariance properties of the augmentation functions we rely on. For binary segmentation tasks, our algorithm was evaluated on the MICCAI Instrument Segmentation Challenge datasets, Robust-MIS 2019 and EndoVis 2017. For multi-class segmentation tasks, we relied on the recent CholecInstanceSeg dataset. Our results show that SegMatch outperforms fully-supervised approaches by incorporating unlabelled data, and surpasses a range of state-of-the-art semi-supervised models across different labelled to unlabelled data ratios. |
| format | Article |
| id | doaj-art-ebc09dc17534461d9bb0beed217b3281 |
| institution | OA Journals |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-ebc09dc17534461d9bb0beed217b32812025-08-20T02:20:05ZengNature PortfolioScientific Reports2045-23222025-04-0115111710.1038/s41598-025-94568-zSegMatch: semi-supervised surgical instrument segmentationMeng Wei0Charlie Budd1Luis C. Garcia-Peraza-Herrera2Reuben Dorent3Miaojing Shi4Tom Vercauteren5School of Biomedical Engineering & Imaging Sciences, King’s College LondonSchool of Biomedical Engineering & Imaging Sciences, King’s College LondonSchool of Biomedical Engineering & Imaging Sciences, King’s College LondonHarvard Medical School, Harvard UniversityCollege of Electronic and Information Engineering, Tongji UniversitySchool of Biomedical Engineering & Imaging Sciences, King’s College LondonAbstract Surgical instrument segmentation is recognised as a key enabler in providing advanced surgical assistance and improving computer-assisted interventions. In this work, we propose SegMatch, a semi-supervised learning method to reduce the need for expensive annotation for laparoscopic and robotic surgical images. SegMatch builds on FixMatch, a widespread semi-supervised classification pipeline combining consistency regularization and pseudo-labelling, and adapts it for the purpose of segmentation. In our proposed SegMatch, the unlabelled images are first weakly augmented and fed to the segmentation model to generate pseudo-labels. In parallel, images are fed to a strong augmentation branch and consistency between the branches is used as an unsupervised loss. To increase the relevance of our strong augmentations, we depart from using only handcrafted augmentations and introduce a trainable adversarial augmentation strategy. Our FixMatch adaptation for segmentation tasks further includes carefully considering the equivariance and invariance properties of the augmentation functions we rely on. For binary segmentation tasks, our algorithm was evaluated on the MICCAI Instrument Segmentation Challenge datasets, Robust-MIS 2019 and EndoVis 2017. For multi-class segmentation tasks, we relied on the recent CholecInstanceSeg dataset. Our results show that SegMatch outperforms fully-supervised approaches by incorporating unlabelled data, and surpasses a range of state-of-the-art semi-supervised models across different labelled to unlabelled data ratios.https://doi.org/10.1038/s41598-025-94568-z |
| spellingShingle | Meng Wei Charlie Budd Luis C. Garcia-Peraza-Herrera Reuben Dorent Miaojing Shi Tom Vercauteren SegMatch: semi-supervised surgical instrument segmentation Scientific Reports |
| title | SegMatch: semi-supervised surgical instrument segmentation |
| title_full | SegMatch: semi-supervised surgical instrument segmentation |
| title_fullStr | SegMatch: semi-supervised surgical instrument segmentation |
| title_full_unstemmed | SegMatch: semi-supervised surgical instrument segmentation |
| title_short | SegMatch: semi-supervised surgical instrument segmentation |
| title_sort | segmatch semi supervised surgical instrument segmentation |
| url | https://doi.org/10.1038/s41598-025-94568-z |
| work_keys_str_mv | AT mengwei segmatchsemisupervisedsurgicalinstrumentsegmentation AT charliebudd segmatchsemisupervisedsurgicalinstrumentsegmentation AT luiscgarciaperazaherrera segmatchsemisupervisedsurgicalinstrumentsegmentation AT reubendorent segmatchsemisupervisedsurgicalinstrumentsegmentation AT miaojingshi segmatchsemisupervisedsurgicalinstrumentsegmentation AT tomvercauteren segmatchsemisupervisedsurgicalinstrumentsegmentation |