Artificial Intelligence Shows Promise For Disseminating Myopia Information

Artificial intelligence demonstrates the potential to provide accurate and quick myopia-related information, but the technology still can deliver inaccurate responses, according to a study published in Ophthalmic and Physiological Optics. In some instances, these chatbots may help patients to avoid unnecessary expenses associated with in-office visits and save clinicians time associated with answering common questions, the report suggests.

A 5-member expert team consisting of optometric teachers and researchers created 11 questions pertaining to myopia, which included questions regarding the disorder’s cause, prevention, symptoms, treatments, and prognosis. The team entered each question into the artificial intelligence language model 5 times, rated the quality of information provided on a 5-point Likert scale (1, very poor; 5, very good), and compared agreement among themselves. 

Overall, the artificial intelligence application provided good information for 10 of the questions (median score, 4.0) and an acceptable response for 1 of the questions (median score, 3.0), according to the report. Investigators rated 24% of responses as very good, 49% as good, 22% as acceptable, 3.6% as poor and 1.8% as very poor. Agreement was good between test items (Cronbach’s α, 0.807) and the investigators’ ratings demonstrated slight agreement (Fleiss’s κ, 0.005) with significant differences in scoring (P < .001).

“[Artificial intelligence] may be the tool for the future, and this is an opportunity to understand the utility of [artificial intelligence] in interacting with patients and providing valuable information on eye diseases,” according to researchers. “[Artificial intelligence] holds the promise to lower the cost of healthcare by reducing consultation time and provide valuable patient information.”

Study limitations include the application’s inability to predict real-life questions in a clinical setting or critically analyze information provided by the scientific literature.