The following is a summary of “A comparative study on the knowledge levels of artificial intelligence programs in diagnosing ophthalmic pathologies and intraocular tumors evaluated their superiority and potential utility,” published in the October 2023 issue of Ophthalmology by Sensoy et al.
Researchers performed a retrospective study to evaluate ChatGPT, Bing, and Bard AI chatbots for ophthalmic pathology and tumor diagnosis accuracy and superiority.
The study involved 36 questions sourced from the American Academy of Ophthalmology’s 2022-2023 Basic and Clinical Science Course in Ophthalmic Pathology and Intraocular Tumor section. Each question was individually posed to three artificial intelligence programs, namely ChatGPT, Bing, and Bard. The responses from these programs were categorized as either correct or incorrect. The study examined the statistical relationship between the accuracy rates of responses from artificial intelligence programs.
The result demonstrated that ChatGPT provided correct answers to 58.6% of the questions, Bing to 63.9%, and Bard to 69.4%. There was no statistical significance observed in the differences in correct response rates among the three artificial intelligence programs (P = 0.705, Pearson Chi-square test).
The study found AI chatbots can provide information on ophthalmic pathologies and tumors, but their answers should be used with caution.
Source: link.springer.com/article/10.1007/s10792-023-02893-x
