Artificial Intelligence in Vision and Ophthalmology

Analysis of the anatomical and functional ocular changes related to spaceflight

Gal Antman — 2026-04-17

Purpose: To describe novel early changes in ocular physiology following short-duration exposure to microgravity.

Observations: The subject was a 64-year-old astronaut who participated in the AX-1 mission that was sent to the International Space Station by AXIOM Space and the National Aeronautics and Space Administration (NASA) in April 2022 for 17 days. Comprehensive multimodal preflight and postflight ophthalmic examinations were performed, including anterior and posterior segment imaging and head magnetic resonance imaging (MRI). In addition, optical coherence tomography angiography (OCTA), which is not part of the standard NASA protocol, was conducted for the first time in this setting in the medical literature. Automated image processing was used to quantify flow signal pixels from the retina or choroid for separate analysis. The subject reported a new-onset need for reading glasses while in space. Mild widening of the optic nerve sheaths was found on MRI. OCTA studies demonstrated a significant postflight decrease in macular choroidal flow signals by 29% in the right eye and 11% in the left eye, and in retinal flow signals by 8.5% and 6.5%, respectively (p < 0.0001 for both factors for both eyes). Focal alterations were noted in choroidal thickness and Haller’s vessel diameter, which did not reach statistical significance.

Conclusion: A significant decrease in macular choroidal and retinal blood flow was observed in an astronaut after a short-duration spaceflight. These changes may serve as possible biomarkers of spaceflight-associated neuro-ocular syndrome and warrant further investigation. We recommend that future spaceflight evaluations include OCTA in the standard protocol.

Multimodal large language models for use in diabetic retinopathy screening

S. Saeed Mohammadi — 2026-03-11

Purpose: To evaluate the performance o f ChatGPT-4o and Gemini 2.5 Pro in detecting more-than-mild diabetic retinopathy (mtmDR) from fundus photography (FP) and diabetic macular edema (DME) from optical coherence tomography (OCT) using publicly available datasets.

Methods: A custom GPT (powered by ChatGPT-4o) was created and instructed to follow the LumineticsCore™ (IDx-DR) screening criteria for mtmDR, defined as an ETDRS level ≥ 35 and/or clinically significant diabetic macular edema (CSDME). Gemini 2.5 Pro was evaluated with the same criteria. Performance on FPs was assessed using 2 publicly available datasets: MESSIDOR-2 (n = 106; 66 mtmDR, 40 no mtmDR) and EyePACS (n = 99; 56 mtmDR, 43 non-mtmDR). To assess detection of DME, a separate OCT dataset (n = 48; 24 DME, 24 normal) was used to evaluate identification of intraretinal and/or subretinal fluid. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for detecting mtmDR on FP and DME on OCT were calculated for each multimodal large language model (LLM).

Results: On MESSIDOR-2 (n = 106), ChatGPT-4o achieved a sensitivity of 90.77%, specificity of 97.50%, PPV of 98.33%, and NPV of 86.67% for mtmDR detection. Gemini 2.5 Pro achieved a sensitivity of 80.30%, specificity of 97.50%, PPV of 98.15%, and NPV of 75.00%. On EyePACS (n = 99), ChatGPT-4o demonstrated a sensitivity of 94.64%, specificity of 86.05%, PPV of 89.83%, and NPV of 92.50%, while Gemini 2.5 Pro achieved a sensitivity of 89.29%, specificity of 88.37%, PPV of 90.91%, and NPV of 86.36%. For OCT-based DME detection (n = 48), ChatGPT-4o achieved a sensitivity of 95.83%, specificity of 100%, and PPV of 100%, while Gemini 2.5 Pro achieved a sensitivity of 95.83%, specificity of 95.65%, PPV of 95.83%, and NPV of 95.65%.

Conclusion: ChatGPT-4o and Gemini 2.5 Pro demonstrated high performance in detecting mtmDR and DME across multiple publicly available datasets. These findings support the potential of multimodal LLMs as cost-effective and accessible tools for diabetic retinopathy screening. Further validation in larger, more diverse real-world datasets is warranted.

Development and planning for future scenarios in ophthalmology: content generation using a modified Delphi process

Timothy P. Mayotte — 2026-04-14

Purpose: To identify the most influential drivers shaping the field of ophthalmology and develop expert-derived, consensus-driven future scenarios.

Design: A mixed-methods study.

Methods: A modified Delphi process was performed to develop and build expert consensus on future scenarios within ophthalmology. Initial faculty surveys were used to identify critical drivers in the field. Focus groups with key opinion leaders (KOLs) in the United States were conducted to discuss implications of these drivers within the field in the context of “aspirational”, “conventional”, and “bleak” state scenarios. Focus groups and surveys were qualitatively analyzed using grounded theory principles of coding to develop the future scenarios. Drafted scenarios were then sent back to KOLs for feedback to achieve consensus.

Results: Twenty-seven faculty responded to an initial survey, identifying five key drivers: artificial intelligence in eye care, health policy and financial reform, physician shortages, the aging population, and research funding. Thirty-one experts participated in five focus groups, yielding 276 coded quotations. Discussion centered most heavily on artificial intelligence (AI) and least on aging. Across all drivers, 51% of coded data reflected conventional projections, 30% aspirational, and 19% bleak. Aspirational futures emphasized whole-person preventive care, AI integration with proper safeguards, equitable workforce distribution, enhanced advocacy for research, and improved care access utilizing assistive technologies. Conversely, bleak futures involved drastic funding cuts, regulatory misalignment in AI, worsening physician shortages, and system overload from demographic pressures.

Conclusion: Scenario planning reveals that ophthalmology’s trajectory depends on proactive strategies to strengthen research advocacy, adopt population-based care models, optimize the workforce, and ensure responsible AI implementation.

SAIVO 2026 Annual Meeting Abstracts

KuglerPublications — 2026-04-30

Abstracts from the annual SAIVO meeting 2026