The Pan-American Journal of Ophthalmology (Jan 2023)

Automated ophthalmic imaging analysis in the era of Generative Pre-Trained Transformer-4

  • Ethan Waisberg,
  • Joshua Ong,
  • Mouayad Masalkhi,
  • Sharif Amit Kamran,
  • Nasif Zaman,
  • Prithul Sarker,
  • Andrew G Lee,
  • Alireza Tavakkoli

DOI
https://doi.org/10.4103/pajo.pajo_62_23
Journal volume & issue
Vol. 5, no. 1
pp. 46 – 46

Abstract

Read online

AI is just beginning to be integrated into clinical medicine, and will continue to be integrated into clinical care. ChatGPT has already been used in ophthalmology to triage symptoms, write operative notes, answer ophthalmic board-exam level questions and for medical education. GPT-4 builds upon the abilities of the prior GPT models by delivering more sophisticated natural language processing capacities, making it an AI technology with high potential application for medical imaging analysis. We wanted to examine GPT-4's ability to interpret ophthalmic images. Although GPT-4 remains a promising start to the future of ophthalmic imaging analysis, further research and modifications will be necessary before the technology can be used for high stakes clinical encounters in human patients.

Keywords