Impact of Artificial Intelligence on the Knowledge, Attitude, and Performance of Ophthalmology Residents: A Systematic Review.

Abstract

This systematic review investigated the role of artificial intelligence (AI) in the knowledge, attitude, and performance of ophthalmology residents. We conducted a comprehensive systematic search in international databases including PubMed, Web of Science, Scopus, CINAHL (Cumulative Index to Nursing and Allied Health Literature), Education Resources Information Center (ERIC) using keywords "artificial intelligence", "deep learning", "ophthalmology", "ocular surgery", and "education" and their synonyms. The keywords were extracted from medical research studies published from January 1, 2018 to April 15, 2024. The quality of these studies was evaluated by using the STORBE, JADA, and JBI appraisal tools. Six studies were selected based on the defined criteria. Specifically, five of these studies investigated the effectiveness of AI interventions on the performance of ophthalmology residents in diagnosing myopia, corneal diseases (using a confocal microscope), staging of diabetic retinopathy, abnormal findings in posterior segment ultrasonography, including retinal detachment, posterior vitreous detachment, and vitreous hemorrhage, and 13 fundus diseases. One study investigated the residents' attitudes about the application of an AI model for providing feedback in cataract surgery. All six studies showed positive results. Due to the small number of studies found through our systematic search and the variations in the investigated outcomes and study settings, it was not possible to conduct a meta-analysis. Despite the positive reports on improving the diagnostic performance of residents and their attitude toward the usability of AI models in cataract surgery, it is recommended that more studies be conducted in this area. These studies should replicate previous investigations using similar study settings while maintaining high quality standards and addressing existing limitations.