Research

Paper

AI LLM March 02, 2026

MMNavAgent: Multi-Magnification WSI Navigation Agent for Clinically Consistent Whole-Slide Analysis

Authors

Zhengyang Xu, Han Li, Jingsong Liu, Linrui Xie, Xun Ma, Xin You, Shihui Zu, Ayako Ito, Xinyu Hao, Hongming Xu, Shaohua Kevin Zhou, Nassir Navab, Peter J. Schüffler

Abstract

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.

Metadata

arXiv ID: 2603.02079
Provider: ARXIV
Primary Category: cs.CV
Published: 2026-03-02
Fetched: 2026-03-03 04:34

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.02079v1</id>\n    <title>MMNavAgent: Multi-Magnification WSI Navigation Agent for Clinically Consistent Whole-Slide Analysis</title>\n    <updated>2026-03-02T17:02:44Z</updated>\n    <link href='https://arxiv.org/abs/2603.02079v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.02079v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>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.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <published>2026-03-02T17:02:44Z</published>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Zhengyang Xu</name>\n    </author>\n    <author>\n      <name>Han Li</name>\n    </author>\n    <author>\n      <name>Jingsong Liu</name>\n    </author>\n    <author>\n      <name>Linrui Xie</name>\n    </author>\n    <author>\n      <name>Xun Ma</name>\n    </author>\n    <author>\n      <name>Xin You</name>\n    </author>\n    <author>\n      <name>Shihui Zu</name>\n    </author>\n    <author>\n      <name>Ayako Ito</name>\n    </author>\n    <author>\n      <name>Xinyu Hao</name>\n    </author>\n    <author>\n      <name>Hongming Xu</name>\n    </author>\n    <author>\n      <name>Shaohua Kevin Zhou</name>\n    </author>\n    <author>\n      <name>Nassir Navab</name>\n    </author>\n    <author>\n      <name>Peter J. Schüffler</name>\n    </author>\n  </entry>"
}