A Decade of Progress in Artificial Intelligence for Fundus Image-Based Diabetic Retinopathy Screening (2014–2024): A Bibliometric Analysis

  1. Yuting Huang
  2. Yunfei Qi
  3. Chi Liu
  4. Fengshi Jing
  5. Changjin Li
  6. Minghao Wang
  7. Congcong Zhu
  8. Peng Gui
  9. Zongyuan Ge
  10. Xiaotong Han
Read the full article See related articles

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Background/Aims

Diabetic retinopathy (DR) screening using artificial intelligence (AI) has evolved significantly over the past decade. This study aimed to analyze research trends, developments, and patterns in AI-based fundus image DR screening from 2014 to 2024 through bibliometric analysis.

Methods

The study analyzed 1,172 publications from the Web of Science Core Collection database using CiteSpace and Microsoft Excel. The analysis included publication trends, citation patterns, institutional collaborations, and keyword emergence analysis.

Results

Publications showed consistent growth from 2014-2022, with a peak in 2021. India (26%), China (20.05%), and USA (9.98%) dominated research output. IEEE ACCESS was the leading publication venue with 44 articles. Research evolved from traditional image processing to deep learning approaches, with recent emphasis on multimodal AI models. The analysis identified three distinct phases: CNN-based systems (2014-2020), Vision Transformers and innovative learning paradigms (2020-2022), and large foundation models (2022-2024).

Conclusion

The field shows mature development in traditional AI approaches while transitioning toward multimodal learning technologies. Future directions indicate increased focus on telemedicine integration, innovative AI algorithms, and real-world implementation.

What is already known on this topic

  • AI-based DR screening has been developing since the 1960s, with significant acceleration after 2014 due to deep learning advances.

  • Traditional manual analysis of fundus images is time-consuming and error-prone.

What this study adds

  • Comprehensive mapping of research evolution in AI-based DR screening over the past decade.

  • Identification of research concentration in specific geographical areas and emerging trends in multimodal AI approaches.

How this study might affect research, practice or policy

  • Highlights the need for increased international collaboration and technology sharing.

  • Suggests focus areas for future research, including multimodal learning and real-world implementation.

  • Provides direction for healthcare organizations and researchers in adopting and developing AI-based DR screening technologies.

Article activity feed

  1. Version published to 10.1101/2024.11.02.24316635v1 on medRxiv