| Description |
This two-part video series on the fundamentals of artificial intelligence in neuro-ophthalmology provides an introduction tailored for clinicians and trainees in the field. In Part 1, we cover how AI models are trained, focusing on dataset curation, labeling, and supervised vs. unsupervised learning approaches. In Part 2, we discuss model testing and evaluation, explaining key performance metrics like ROC and precision-recall curves, the differences between retrospective and prospective testing, and the importance of subgroup analyses. Together, the series equips neuro-ophthalmologists with the knowledge to evaluate AI tools and understand their limitations. |
| References |
1. Vasseneix C, Nusinovici S, Xu X, et al. Deep Learning System Outperforms Clinicians in Identifying Optic Disc Abnormalities. J Neuroophthalmol. 2023;43(2):159-167. doi:10.1097/WNO.0000000000001800. 2. Li, Anfei MD, PhD; Tandon, Anika K. MD; Sun, Grace MD; Dinkin, Marc J. MD; Oliveira, Cristiano MD. Early Detection of Optic Nerve Changes on Optical Coherence Tomography Using Deep Learning for Risk-Stratification of Papilledema and Glaucoma. Journal of Neuro-Ophthalmology 44(1):p 47-52, March 2024. | DOI: 10.1097/WNO.0000000000001945. 3. Chen D, Geevarghese A, Lee S, et al. Transparency in Artificial Intelligence Reporting in Ophthalmology-A Scoping Review. Ophthalmol Sci. 2024;4(4):100471. Published 2024 Jan 18. doi:10.1016/j.xops.2024.100471. 4. Nakayama LF, Mitchell WG, Shapiro S, et al Sociodemographic disparities in ophthalmological clinical trials BMJ Open Ophthalmology 2023;8:e001175. doi: 10.1136/bmjophth-2022-001175. 5. Zhou Y, Chia MA, Wagner SK, et al. A foundation model for generalizable disease detection from retinal images. Nature. 2023;622(7981):156-163. doi:10.1038/s41586-023-06555-x. 6. Jiang X, Dong L, Luo L, Zhang K, Li D. Retinal Photograph-based Deep Learning System for Detection of Thyroid-Associated Ophthalmopathy. J Craniofac Surg. 2024;35(2):e164-e167. doi:10.1097/SCS.0000000000009919. 7. Wu E, Wu K, Daneshjou R, Ouyang D, Ho DE, Zou J. How medical AI devices are evaluated: limitations and recommendations from an analysis of FDA approvals. Nat Med. 2021;27(4):582-584. doi:10.1038/s41591-021-01312-x. 8. Chuter Bm Huynh J, Hallaj S, Walker E, Liebmann J, Fazio M, Girkin C, Weinreb R, Christopher M, Zangwill L, Evaluating a foundation AI model for glaucoma detection in a diverse study population using color fundus photographs, Ophthalmol Science, 2024, https://doi.org/10.1016/j.xops.2024.100623. |
