跳到正文
原文
Nature 健康护理主题·· 27 天前AI 评分7

回复:基于基础模型的血红蛋白浓度估算及视网膜眼底图像贫血风险评估评论

Response to: ‘Comment on: ‘Foundation model-based haemoglobin concentration estimation and anaemia risk assessment from retinal fundus images”

AI 导读

Kwak等回应了关于视网膜眼底图像估算血红蛋白的临床价值问题。研究团队指出该研究仅评估适配预训练基础模型的可行性与可解释性,并未在特定工作流程中建立临床效用。因此无法确定该方法能否减少采血、改变转诊决策或改善下游临床管理。

正文
  • Correspondence
  • Published: 24 October 2026

Eye (2026) Cite this article

We thank Arif for highlighting that the clinical value of retinal haemoglobin prediction depends on the pathway in which it is used. We agree that discrimination alone is insufficient, and that disease prevalence, referral capacity, access to confirmatory testing, and the consequences of false-positive and false-negative results should guide the selection of operating thresholds.

The scope of the present study was to evaluate the feasibility and interpretability of adapting pretrained foundation models for haemoglobin estimation from fundus photographs, rather than to establish clinical utility within a specific workflow. In this context, the health-promotion programme served as a source of paired retinal imaging and reference-standard laboratory data for model development and evaluation, rather than representing the intended deployment pathway. Accordingly, the present study does not establish whether retinal haemoglobin prediction can reduce blood sampling, alter referral decisions, or improve downstream clinical management.

This is a preview of subscription content, access via your institution

Access options

Access through your institution

Subscribe to this journal

Receive 18 print issues and online access

251.40 € per year only 13.97 € per issue

Learn more

Buy this article

  • Purchase on SpringerLink
  • Instant access to the full article PDF.

39,95 €

Prices may be subject to local taxes which are calculated during checkout

Additional access options:

Data availability

No new datasets were generated or analysed for this correspondence. Data underlying the original article are available from the corresponding author on reasonable request, subject to approval by the relevant institutional review board.

Author information

Author notes

  1. These authors contributed equally: Sangmin Kwak, Jaewon Kim.

Authors and Affiliations

  1. College of Medicine, Seoul National University, Seoul, Republic of Korea

    Sangmin Kwak

  2. Department of Biomedical Sciences, Seoul National University Graduate School, Seoul, Republic of Korea

    Jaewon Kim & Sang Min Park

  3. XAIMED Co. Ltd., Seoul, Republic of Korea

    Sang Min Park

Authors

  1. Sangmin Kwak
  2. Jaewon Kim
  3. Sang Min Park

Contributions

Sangmin Kwak and Jaewon Kim contributed equally as co-first authors. Sangmin Kwak: Conceptualisation, writing—original draft, writing—review and editing. Jaewon Kim: Conceptualisation, writing—review and editing, supervision. Sang Min Park: Writing—review and editing, supervision.

Corresponding author

Correspondence to Sang Min Park.

Ethics declarations

Competing interests

Sang Min Park is the founder and CEO of XAIMED Co. Ltd. The author confirms that this relationship did not influence the study design, data collection, analysis, or interpretation of the results. The remaining authors declare no competing interests in relation to the work described.

Additional information

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

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Kwak, S., Kim, J. & Park, S.M. Response to: ‘Comment on: ‘Foundation model-based haemoglobin concentration estimation and anaemia risk assessment from retinal fundus images”. Eye (2026). https://doi.org/10.1038/s41433-026-04982-4

Download citation

  • Received: 29 August 2026

  • Revised: 04 September 2026

  • Accepted: 07 October 2026

  • Published: 24 October 2026

  • Version of record: 24 October 2026

  • DOI: https://doi.org/10.1038/s41433-026-04982-4

来源:Nature 健康护理主题 · nature.com