Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.02079 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917306088554496 |
|---|---|
| author | Xu, Zhengyang Li, Han Liu, Jingsong Xie, Linrui Ma, Xun You, Xin Zu, Shihui Ito, Ayako Hao, Xinyu Xu, Hongming Zhou, Shaohua Kevin Navab, Nassir Schüffler, Peter J. |
| author_facet | Xu, Zhengyang Li, Han Liu, Jingsong Xie, Linrui Ma, Xun You, Xin Zu, Shihui Ito, Ayako Hao, Xinyu Xu, Hongming Zhou, Shaohua Kevin Navab, Nassir Schüffler, Peter J. |
| contents | Recent AI navigation approaches aim to improve Whole-Slide Image (WSI) diagnosis by modeling spatial exploration and selecting diagnostically relevant regions, yet most operate at a single fixed magnification or rely on predefined magnification traversal. In clinical practice, pathologists examine slides across multiple magnifications and selectively inspect only necessary scales, dynamically integrating global and cellular evidence in a sequential manner. This mismatch prevents existing methods from modeling cross-magnification interactions and adaptive magnification selection inherent to real diagnostic workflows. To these, we propose a clinically consistent Multi-Magnification WSI Navigation Agent (MMNavAgent) that explicitly models multi magnification interaction and adaptive magnification selection. Specifically, we introduce a Cross-Magnification navigation Tool (CMT) that aggregates contextual information from adjacent magnifications to enhance discriminative representations along the navigation path. We further introduce a Magnification Selection Tool (MST) that leverages memory-driven reasoning within the agent framework to enable interactive and adaptive magnification selection, mimicking the sequential decision process of pathologists. Extensive experiments on a public dataset demonstrate improved diagnostic performance, with 1.45% gain of AUC and 2.93% gain of BACC over a non-agent baseline. Code will be public upon acceptance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_02079 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | MMNavAgent: Multi-Magnification WSI Navigation Agent for Clinically Consistent Whole-Slide Analysis Xu, Zhengyang Li, Han Liu, Jingsong Xie, Linrui Ma, Xun You, Xin Zu, Shihui Ito, Ayako Hao, Xinyu Xu, Hongming Zhou, Shaohua Kevin Navab, Nassir Schüffler, Peter J. Computer Vision and Pattern Recognition Recent AI navigation approaches aim to improve Whole-Slide Image (WSI) diagnosis by modeling spatial exploration and selecting diagnostically relevant regions, yet most operate at a single fixed magnification or rely on predefined magnification traversal. In clinical practice, pathologists examine slides across multiple magnifications and selectively inspect only necessary scales, dynamically integrating global and cellular evidence in a sequential manner. This mismatch prevents existing methods from modeling cross-magnification interactions and adaptive magnification selection inherent to real diagnostic workflows. To these, we propose a clinically consistent Multi-Magnification WSI Navigation Agent (MMNavAgent) that explicitly models multi magnification interaction and adaptive magnification selection. Specifically, we introduce a Cross-Magnification navigation Tool (CMT) that aggregates contextual information from adjacent magnifications to enhance discriminative representations along the navigation path. We further introduce a Magnification Selection Tool (MST) that leverages memory-driven reasoning within the agent framework to enable interactive and adaptive magnification selection, mimicking the sequential decision process of pathologists. Extensive experiments on a public dataset demonstrate improved diagnostic performance, with 1.45% gain of AUC and 2.93% gain of BACC over a non-agent baseline. Code will be public upon acceptance. |
| title | MMNavAgent: Multi-Magnification WSI Navigation Agent for Clinically Consistent Whole-Slide Analysis |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.02079 |