About

Glenn C. Yiu, M.D., Ph.D., is a professor in the Department of Ophthalmology at UC Davis Health and serves as the Director of Tele-Ophthalmology. He is a board-certified vitreoretinal specialist and surgeon at the UC Davis Eye Center. Dr. Yiu received his medical training at Harvard Medical School, where he earned a dual MD-PhD degree with research exploring mechanisms of nervous system regeneration. His residency was completed at Harvard, followed by a vitreoretinal surgery fellowship at Duke University, where he also engaged in ocular imaging research focused on age-related macular degeneration (AMD) and diabetic retinopathy. His clinical expertise encompasses the medical and surgical management of retinal diseases such as macular degeneration, diabetic retinopathy, retinal vascular conditions, retinal detachments, and macular diseases including epiretinal membranes and macular holes. As a federally-funded clinician-scientist, his research interests include AMD, diabetic retinopathy, gene therapy, ocular imaging, and telemedicine. Dr. Yiu has pioneered innovative research, including the first use of CRISPR-based genome editing for wet AMD, and has developed novel methods for gene delivery using microneedles. He is actively involved in translational research spanning animal models to human patients, and he leads major national clinical trials for retinal disease treatments. His contributions to the field include numerous peer-reviewed publications, editorial roles, and international lectures, along with recognition through awards such as the Ronald G. Michels Fellowship, the Heed Fellowship, and the Retina Society Fellowship Research Award. Dr. Yiu is committed to advancing eye care through research, education, and the development of tele-ophthalmology programs to expand screening and treatment access.

Research topics

• Medicine
• Surgery
• Ophthalmology
• Internal medicine
• Optometry
• Endocrinology

Selected publications

• Simultaneous Segmentation of Geographic Atrophy in Longitudinally Acquired Fundus Autofluorescence Images

  Ophthalmology Science · 2026-02-07

  articleOpen access

  Purpose: Accurately measuring geographic atrophy (GA) progression in clinical trials is challenging, owing to its slow and variable nature.This study develops deep learning algorithms that can measure GA growth rates with improved accuracy and robustness, by performing automated annotation of preexisting versus newly expanding GA regions, based on simultaneous grading of fundus autofluorescence (FAF) image pairs.Design: Retrospective analysis of data acquired in two independent prospective clinical trials (i.e., 830 images AREDS2 and 273 images in METfoMIN).Study Participants: 174 AREDS2 and 44 METfoMIN participants. Methods:A standard DeepLabV3+ segmentation model that automatically segments GA from a single FAF image (DL STATIC ) was enhanced to input a longitudinally acquired, spatially aligned FAF image pair and simultaneously segment regions corresponding to existing and expanding GA pixels (DL JOINT ).This model was further modified to ensure longitudinal consistency of existing GA pixels (DL JOINT+SCL ).Main Outcome Measures: Image pixel overlap (using precision, recall, and Dice), GA area, and GA growth rates were measured with manual contours as the gold standard.The study also quantified the effect of longitudinal alignment of images to GA growth measurements. Results:When compared to manual expert annotations, the DL JOINT+SCL model exhibited the highest performance, with precision (mean stdev) = 0.86 0.13, recall = 0.92 0.11, and Dice = 0.89 0.11, highest GA area correlation (R 2 = 0.97), and highest growth rate correlation (R 2 = 0.82).This shows the utility of simultaneously segmenting images when computing growth.Comparatively, the DL STATIC model exhibited precision = 0.84 0.13, recall = 0.93 0.15, and Dice = 0.87 0.13, area correlation R 2 = 0.94, J o u r n a l P r e -p r o o f and growth rate correlation R 2 =0.75.Spatial alignment via image registration was a key step that enabled the algorithm implementation; this yielded accuracy improvements when measuring GA growth by mapping regions to a common spatial coordinate system. Conclusion:The simultaneous segmentation of longitudinal FAF images yielded improvements in accuracy and robustness by maintaining the longitudinal consistency of measurements.Such improvements are critical to the translation of automatic algorithms and minimizing measurement variability to effectively assess clinical trial outcomes in GA.

  PublisherOA PDFDOI

• Factors Associated with Machine Learning–Based Predictions of Retinal Aging Using Teleretinal Screening Images from Patients with Diabetes

  Ophthalmology Science · 2026-02-07

  articleOpen accessSenior author

  Purpose: To identify factors associated with accelerated retinal aging based on machine learning predictions of age using fundus images from teleretinal screening of patients with diabetes. Design: Cross-sectional study of retinal images. Subjects: Ten thousand, five hundred thirty eye images from 2939 patients with diabetes who underwent teleretinal screening at the University of California clinics. Methods: We trained a vision transformer (ViT) model to predict chronological age from retinal fundus photographs of 2939 patients with diabetes who underwent teleretinal screening as part of the Collaborative University of California Teleophthalmology Initiative (CUTI), and validated it using images from the Artificial Intelligence Ready and Exploratory Atlas for Diabetes Insights data set. We collected demographic, lifestyle, and systemic health factors, and analyzed their association with prediction errors, known as the retinal age gap. Main Outcome Measures: Association between demographic, lifestyle, and systemic factors with retinal age gap. Results: < 0.001). Key limitations include the cross-sectional study design and potential biases in medical record data. Conclusions: Machine learning predictions of retinal aging using teleretinal images from patients with diabetes may predict cardiovascular risk and are accelerated by systemic comorbidities. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

  PublisherDOI

• Applying Lincoff’s Rule to Central Serous Chorioretinopathy: The Macula Society International CSCR Research Network (MICRoN) Report-15

  Ophthalmology Science · 2026-04-17

  articleOpen access
  PublisherDOI

• THE “SCOLEX” SIGN

  Retina · 2026-01-05

  article

  PURPOSE: To describe a novel optical coherence tomography finding called "scolex sign" in cases of central serous chorioretinopathy. METHODS: This retrospective multicenter study included patients with central serous chorioretinopathy with serous pigment epithelial detachments (PEDs) greater than 100 µ m, with or without subretinal fluid. Eyes showing a distinct hyperreflective focus on the PED wall (scolex sign) were analyzed. An equal number of age-matched controls with PEDs but without the scolex sign were included. Multimodal imaging data were reviewed. A subgroup analysis based on subretinal fluid status was also performed. RESULTS: Of 291 eyes with large serous PEDs, 52 eyes exhibited the "scolex sign" and were compared with 52 age-matched controls. Baseline characteristics including sex, systemic comorbidities, best-recorded visual acuity and optical coherence tomography parameters were similar in both the groups. However, eyes with the "scolex" sign exhibited a greater number of central PEDs and shorter distance from the foveal center. On follow-up, eyes with the "scolex" sign showed a higher rate of PED flattening and subretinal fluid resolution trended to be higher compared with controls. CONCLUSION: The "scolex" sign represents a novel, benign optical coherence tomography feature seen in a subset of central serous chorioretinopathy cases. While not associated with poorer outcomes or adverse sequelae, it may reflect ongoing reparative changes, indicating a resolving stage of the disease.

  PublisherDOI

• Demography, Clinical Characteristics and Long-Term Outcomes of Central Serous Chorioretinopathy in Women. MICRoN Report Number Fourteen

  American Journal of Ophthalmology · 2026-02-27

  articleOpen access

  PURPOSE: To evaluate the characteristics and longitudinal outcomes of chronic central serous chorioretinopathy (CSCR) in women compared to an age-matched cohort of men with CSCR. DESIGN: Retrospective, multicenter clinical cohort study from the Macula Society CSCR Study Group. PARTICIPANTS: This study included 426 eyes (213 women and 213 age-matched men) with a diagnosis of CSCR. METHODS: Baseline and final best-recorded visual acuity (BRVA) and multimodal imaging parameters such as area of retinal pigment epithelium (RPE) alterations, choroidal macular thickness (CMT), sub-foveal choroidal thickness (SFCT), subretinal fluid (SRF), pigment epithelium detachment (PED), double layer sign (DLS), hyperreflective dots (HRD), as well as the presence of choroidal neovascularization (CNV) and subretinal hyperreflective material (SHRM) were assessed. Regression analysis was used to evaluate baseline predictors of final visual acuity. MAIN OUTCOME MEASURES: Longitudinal changes in BRVA and imaging parameters in men and women stratified for age; factors affecting subretinal fluid (SRF) persistence, and change in BRVA. RESULTS: A total of 426 eyes (213 women and 213 age-matched men) with CSCR were analyzed. Women showed better BRVA at presentation (0.25 ± 0.24 vs 0.31 ± 0.35 logMAR; P = .05), and exhibited smaller areas of RPE alterations (2.37 ± 2.64 vs 1.59 ± 1.55 disc areas; P = .003), less frequent peripapillary RPE changes (13.6% vs 7.5%; P < .001), shorter DLS (1353.9 ± 970.2 vs 1071.6 ± 888.7 µm; P = .039), and smaller PEDs (644.9 ± 546.4 vs 442.1 ± 278.9 µm; P = .022). During follow-up, women exhibited higher rates of complete SRF resolution (P = .001) while persistence and the number of recurrences were significantly more common in men (P = .006 and P = .02, respectively). Logistic regression analysis revealed that persistent SRF was independently associated with complex CSCR, male gender, baseline PROS irregularities, worse BRVA, SHRM, and CNV, while PDT was protective. CONCLUSION: Women had better visual outcomes and more favorable structural evolution while men tended to present with more complex anatomical alterations and experience higher rates of persistent SRF.

  PublisherOA PDFDOI

• Demography, Clinical Characteristics and Long-Term Outcomes of Central Serous Chorioretinopathy in Women. MICRoN report number Fourteen.

  Open MIND · 2026-03-03

  articleOpen access

  Purpose To evaluate the characteristics and longitudinal outcomes of chronic central serous chorioretinopathy (CSCR) in women compared to an age-matched cohort of men with CSCR.Design Retrospective, multicenter clinical cohort study from the Macula Society CSCR Study Group.Participants This study included 426 eyes (213 women and 213 age-matched men) with a diagnosis of CSCR.Methods Baseline and final best-recorded visual acuity (BRVA) and multimodal imaging parameters such as area of retinal pigment epithelium (RPE) alterations, choroidal macular thickness (CMT), sub-foveal choroidal thickness (SFCT), subretinal fluid (SRF), pigment epithelium detachment (PED), double layer sign (DLS), hyperreflective dots (HRD), as well as the presence of choroidal neovascularization (CNV) and subretinal hyperreflective material (SHRM) were assessed. Regression analysis was used to evaluate baseline predictors of final visual acuity.Main Outcome Measures Longitudinal changes in BRVA and imaging parameters in men and women stratified for age; factors affecting subretinal fluid (SRF) persistence, and change in BRVA.Results A total of 426 eyes (213 women and 213 age-matched men) with CSCR were analyzed. Women showed better BRVA at presentation (0.25 ± 0.24 vs 0.31 ± 0.35 logMAR; p = 0.05), and exhibited smaller areas of RPE alterations (2.37 ± 2.64 vs 1.59 ± 1.55 disc areas; p = 0.003), less frequent peripapillary RPE changes (13.6% vs 7.5%; p &lt; 0.001), shorter DLS (1353.9 ± 970.2 vs 1071.6 ± 888.7 µm; p = 0.039), and smaller PEDs (644.9 ± 546.4 vs 442.1 ± 278.9 µm; p = 0.022). During follow-up, women exhibited higher rates of complete SRF resolution (p=0.001) while persistence and the number of recurrences were significantly more common in men (p=0.006 and p=0.02, respectively). Logistic regression analysis revealed that persistent SRF was independently associated with complex CSCR, male gender, baseline PROS irregularities, worse BRVA, SHRM, and CNV, while PDT was protective.Conclusion Women had better visual outcomes and more favorable structural evolution while men tended to present with more complex anatomical alterations and experience higher rates of persistent SRF.

  OA PDFDOI

• Two‑step segmentation of motion‑corrupted primate OCT using annotation‑efficient training

  2026-04-01

  article
  PublisherDOI

• Choroidal imaging biomarkers as predictors of recurrent central serous chorioretinopathy: A multicenter retrospective study – MICRoN report number five

  Graefe s Archive for Clinical and Experimental Ophthalmology · 2026-02-17

  article
  PublisherDOI

• Impact of Diet and Gut Microbiome on Nonhuman Primate Models of Age-Related Macular Degeneration

  Current Developments in Nutrition · 2025-05-01

  articleOpen accessSenior author
  PublisherDOI

• Genotype Prediction from Retinal Fundus Images Using Deep Learning in Eyes with Age-Related Macular Degeneration

  Ophthalmology Science · 2025-05-27

  articleOpen accessSenior author

  Purpose: To employ deep learning models to predict high-risk genetic variants associated with age-related macular degeneration (AMD) from retinal fundus photographs of patients with this condition. Design: Deep learning algorithm development to classify single-nucleotide polymorphism in the complement factor H (CFH) and age-related maculopathy susceptibility 2 (ARMS2) genes using retinal fundus images. Participants: Thirty-one thousand two hundred seventy-one retinal color fundus photographs of 1754 participants from the Age-Related Eye Disease Study. Methods: We trained deep learning models including convolution neural networks and vision transformers (ViTs) to classify patients into high-risk (homozygous high-risk alleles) or low-risk (heterozygous or homozygous low-risk alleles) genotypes for CFH or ARMS2, then evaluated algorithm performance on an independent test set. The complexity of genotype predictions was compared with AMD severity or gender classification tasks using V-usable information. Attribution mapping was performed to identify fundus regions used to predict genotype from phenotype. Main Outcome Measures: Area under the receiver operating characteristic curve (AUROC), balanced accuracy, and average precision for predicting high-risk genotypes. Results: Our trained ViT models predicted high-risk genotypes in CFH and ARMS2 with an AUROC of 0.719 and 0.741 across all eyes, respectively. For genotype predictions for ARMS2, model performance is improved in eyes with advanced AMD (AUROC 0.867), choroidal neovascularization (AUROC 0.833), and geographic atrophy (AUROC 0.957). Genotype predictions from fundus images appear more difficult than AMD severity or gender classification tasks, although saliency mapping supports biological plausibility by demonstrating attention to the central macula for genotype predictions. Conclusions: Deep learning can predict high-risk genotypes in CFH and ARMS2 from retinal fundus images of patients with AMD. Our findings provide proof of principle for inferring genotype from noninvasive eye imaging and reveal insights into genotype-phenotype relationships in AMD. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

  PublisherDOI

Recent grants

• Optogenetic Control of Oxidative Stress as a Model of Geographic Atrophy

  NIH · $312k · 2019–2021

• Soft drusen in rhesus macaques as a nonhuman primate model of early age-related macular degeneration

  NIH · $3.2M · 2021–2026

• A nanoparticle platform for targeted delivery of therapeutic agents to retinal pigment epithelial or choroidal endothelial cells

  NIH · $1.0M · 2016–2020

Frequent coauthors

• Sina Farsiu

  Duke University

  45 shared

• Ala Moshiri

  42 shared

• Sara M. Thomasy

  University of California, Davis

  29 shared

• Sally L. Baxter

  University of California, Davis

  26 shared

• Matthew Freeby

  University of California, San Diego

  26 shared

• Christine Thorne

  UC San Diego Health System

  26 shared

• George Su

  University of California, San Francisco

  25 shared

• Chhavi Gregg

  University of California, San Diego

  25 shared

Education

• Residency

  Massachusetts Eye and Ear Infirmary

• MD, PhD

  Harvard Medical School

• Fellowship

  Duke University Hospital

Awards & honors

• American Academy of Ophthalmology Achievement Award, 2019
• International Society for Eye Research Travel Fellowship, 20…
• American Society of Retina Specialists Honor Award, 2018
• Macula Society International Travel Grant, 2017
• NAEVR Emerging Vision Scientist, 2016