Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Brief Communication
  • Published:

Can ‘Deep Research’ agents and general AI agentic systems autonomously perform systematic review and meta-analysis?

Eye volume 40, pages 147–149 (2026)Cite this article

Subjects

Recent advances in large language models (LLMs) have expanded their use across domains, including academic research [1]. Following this, new ‘Deep Research’ agents have emerged, having the ability to actively search and evaluate web content from multiple sources to synthesise insights through the integration of Retrieval-Augmented Generation (RAG) [2]. Extending this concept, agentic systems are designed to coordinate multiple agents, theoretically enabling autonomous execution of evidence synthesis.

Systematic reviews and meta-analyses (SRMAs) are a cornerstone of evidence-based medicine but are resource-intensive. AI agents may potentially support parts of the SRMA workflow through autonomous task execution. Nevertheless, the ability of such AI agents to independently conduct end-to-end SRMAs remains untested. Our study aimed to evaluate and compare the performances of different AI agents in autonomously conducting SRMAs.

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

Access options

Buy this article

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

USD 39.95

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

Data availability

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.

References

  1. Chen H, Jiang Z, Liu X, Xue CC, Yew SME, et al. Can large language models fully automate or partially assist paper selection in systematic reviews? Br J Ophthalmol. https://doi.org/10.1136/bjo-2024-326254

  2. Google. Gemini Deep Research. Accessed 15 April 2024, https://gemini.google/overview/deep-research/?hl=en

  3. OpenAI. Introducing Deep Research. Accessed 15 April 2025, https://openai.com/index/introducing-deep-research/

  4. Perplexity. What is Deep Research? Accessed 28 April 2025, https://www.perplexity.ai/help-center/en/articles/10738684-what-is-deep-research

  5. Ha A, Lee YJ, Lee M, Shim SR, Kim YK. Digital screen time and myopia: a systematic review and dose-response meta-analysis. JAMA Netw Open. 2025;8:e2460026.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Ophthalmology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore

    Wan Ting Loke, Sahana Srinivasan, Ke Zou & Yih-Chung Tham

  2. Centre for Innovation and Precision Eye Health, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore

    Wan Ting Loke, Sahana Srinivasan, Ke Zou & Yih-Chung Tham

  3. Singapore Eye Research Institute, Singapore National Eye Centre, Singapore, Singapore

    Gabriel Dawei Yang & Yih-Chung Tham

  4. Moorfields Eye Hospital NHS Foundation Trust, London, UK

    Ariel Yuhan Ong

  5. Institute of Ophthalmology, University College London, London, UK

    Ariel Yuhan Ong & Pearse A. Keane

  6. NIHR Biomedical Research Centre, Moorfields Eye Hospital NHS Foundation Trust, London, UK

    Ariel Yuhan Ong & Pearse A. Keane

  7. Centre for Eye Research Australia, Ophthalmology, University of Melbourne, Melbourne, VIC, Australia

    Lisa Zhuoting Zhu

  8. Department of Surgery (Ophthalmology), University of Melbourne, Melbourne, VIC, Australia

    Lisa Zhuoting Zhu

  9. Cole Eye Institute, Cleveland Clinic, Cleveland, OH, USA

    Fares Antaki

  10. The CHUM School of Artificial Intelligence in Healthcare, Montreal, QC, Canada

    Fares Antaki

  11. Ophthalmology and Visual Science Academic Clinical Program (Eye ACP), Duke-NUS Medical School, Singapore, Singapore, Singapore

    Yih-Chung Tham

Authors
  1. Wan Ting Loke
    View author publications

    Search author on:PubMed Google Scholar

  2. Sahana Srinivasan
    View author publications

    Search author on:PubMed Google Scholar

  3. Gabriel Dawei Yang
    View author publications

    Search author on:PubMed Google Scholar

  4. Ke Zou
    View author publications

    Search author on:PubMed Google Scholar

  5. Ariel Yuhan Ong
    View author publications

    Search author on:PubMed Google Scholar

  6. Lisa Zhuoting Zhu
    View author publications

    Search author on:PubMed Google Scholar

  7. Fares Antaki
    View author publications

    Search author on:PubMed Google Scholar

  8. Pearse A. Keane
    View author publications

    Search author on:PubMed Google Scholar

  9. Yih-Chung Tham
    View author publications

    Search author on:PubMed Google Scholar

Contributions

WTL, SS, and YCT conceptualised and designed the study. WTL, SS, and GDWY evaluated and graded the generated outputs. WTL drafted the initial version and prepared the final version of the manuscript. KZ, AYO, LZZ, FA, and PAK critically reviewed and revised the manuscript.

Corresponding author

Correspondence to Yih-Chung Tham.

Ethics declarations

Competing interests

Y-CT is a member of the Eye editorial board. The remaining authors declare that they have no conflict of interest.

Declaration of Generative AI and AI-assisted Technologies in the Writing Process

After completing the first draft of the manuscript, the authors used ChatGPT-5 to vet for grammatical errors and improve the language and readability. It was not used for generation of new content. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the final version of the manuscript.

Additional information

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

Supplementary information

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Loke, W.T., Srinivasan, S., Yang, G.D. et al. Can ‘Deep Research’ agents and general AI agentic systems autonomously perform systematic review and meta-analysis?. Eye 40, 147–149 (2026). https://doi.org/10.1038/s41433-025-04138-w

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Version of record:

  • Issue date:

  • DOI: https://doi.org/10.1038/s41433-025-04138-w

Search

Advanced search

Quick links