跳到正文
原文
Nature 健康护理主题·· 2 天前AI 评分38

基于多模态自适应学习的青光眼诊断:眼底照片与视野数据缺失情况下的模型开发

Modality-adaptive learning for glaucoma diagnosis with incomplete availability of fundus photographs and visual field data

AI 导读

CGMA-Net模型在5,181个样本(3,262名参与者)的两中心回顾性队列中实现青光眼分类,内部测试集AUC为0.997,准确率97.6%,敏感性98.9%,特异性94.9%,且在仅眼底照片、仅视野检查、配对数据三种模态条件下均保持高判别能力。该研究为技术可行性证明,尚需在包含诊断疑难病例的连续队列中进一步验证后方可考虑临床应用。

正文

Abstract

Multimodal diagnostic models often assume that all input modalities are available, whereas clinical datasets commonly contain incomplete examination records. We developed CGMA-Net to evaluate glaucoma diagnosis under three modality-availability conditions: fundus photography only, visual field testing only, and paired fundus and visual field data. CGMA-Net combines modality-specific encoders, an explicit availability mask, and gated adaptive fusion to prevent absent inputs from contributing to the fused representation without imputing missing examinations. The model was evaluated in a retrospective two-center cohort comprising 5,181 samples from 3,262 participants, with participant-level separation of training, validation, and test sets. On the internal test set, CGMA-Net achieved an area under the receiver operating characteristic curve of 0.997, with 97.6% accuracy, 98.9% sensitivity, and 94.9% specificity, and maintained high discrimination across modality-availability conditions. In ablation analyses, removing gated fusion produced a larger performance decrease than removing cross-attention. Post hoc temperature scaling improved probability calibration without changing classification metrics. Both learned uncertainty and Monte Carlo dropout uncertainty were higher for misclassified than correctly classified samples. These results demonstrate the technical feasibility of modality-adaptive glaucoma classification across CFP-only, VF-only, and paired CFP–VF settings in a selected retrospective cohort, but do not establish performance in an unselected clinical population. Validation in a consecutive cohort that includes diagnostically ambiguous cases, relevant differential diagnoses, and examinations of variable quality is required before clinical use.

Acknowledgements

We are very grateful to the clinicians from the Eye Center at the Second Affiliated Hospital of Zhejiang University School of Medicine (Zhejiang, China) for participating in data collection, annotation and the exploratory reader assessment. We also thank the Eye Center of Peking University Third Hospital (Beijing, China) for contributions to data collection. In particular, we would like to thank clinicians YY and XDL for their participation in data collection and annotation, and clinicians AS, YMS, JXZ, ZYG, and YFZ for their participation in the reader assessment. We thank Shuangshuang Liu at the Core Facilities of Zhejiang University School of Medicine for technical support.

Funding

This work was supported in part by STI 2030—Major Projects (2021ZD0200401), Key R&D Program of Zhejiang Province (2024C03150, 2025C01119), Fundamental Research Funds for the Central Universities (226-2026-00003).

Author information

Authors and Affiliations

  1. Department of Neurology of the Second Affiliated Hospital of Zhejiang University, School of Medicine, State Key Laboratory of Extreme Photonics and Instrumentation & Liangzhu Laboratory, College of Optical Science and Engineering, Zhejiang University, Hangzhou, 310027, China

    Jiazhu Zhu, Yao Zheng, Ke Si & Wei Gong

  2. Nanhu Brain-computer Interface Institute, Hangzhou, 311100, China

    Jiazhu Zhu & Wei Gong

  3. MOE Frontier Science Center for Brain Science & Brain-Machine Integration, NHC and CAMS Key Laboratory of Medical Neurobiology, School of Brain Science and Brain Medicine, Zhejiang University, Hangzhou, 310058, China

    Jiajia Chen, Ke Si & Wei Gong

  4. Eye Center of Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, China

    Juan Ye

  5. Zhejiang Provincial Key Laboratory of Ophthalmology, Hangzhou, Zhejiang, China

    Juan Ye

  6. Zhejiang Institute of Medical Device Testing, Hangzhou, 310018, China

    Wei Gong

  7. Lingang Laboratory, Shanghai, 200031, China

    Wei Gong

Authors

  1. Jiazhu Zhu
  2. Jiajia Chen
  3. Yao Zheng
  4. Juan Ye
  5. Ke Si
  6. Wei Gong

Corresponding authors

Correspondence to Juan Ye, Ke Si or Wei Gong.

Ethics declarations

Competing interests

The authors declare no competing interests.

Disclosures

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhu, J., Chen, J., Zheng, Y. et al. Modality-adaptive learning for glaucoma diagnosis with incomplete availability of fundus photographs and visual field data. Sci Rep (2026). https://doi.org/10.1038/s41598-026-81016-9

Download citation

  • Received: 14 August 2026

  • Accepted: 12 November 2026

  • Published: 18 November 2026

  • DOI: https://doi.org/10.1038/s41598-026-81016-9

来源:Nature 健康护理主题 · nature.com