Presenting at this year’s American Academy of Optometry meeting was Dr. Donald Hood, Professor of Ophthalmic Science at Columbia University. Dr. Hood has published extensively on the optimal use of spectral-domain OCT (SD-OCT) data in glaucoma assessment, leading to what is now commonly known as the “Columbia University Method.”¹ This method allows clinicians to diagnose or exclude glaucoma based on OCT findings alone, by applying a structured approach grounded in retinal nerve fiber layer (RNFL) anatomy. The process is guided by three central questions that help the clinician classify each case as glaucoma, no glaucoma, or uncertain.² At the 2025 meeting, Dr. Hood introduced a significant refinement to this framework—the recognition of arcuate area prototypes within both the superior and inferior RNFL. These archetypal patterns are used in a three-dimensional assessment of the OCT report, allowing simultaneous evaluation of multiple retinal regions rather than isolated segments.
According to Dr. Hood, this approach improves diagnostic specificity and reduces false positives and negatives—an important advance for clinicians who rely on OCT data in the early detection of glaucoma.
He also reported progress on an artificial intelligence (AI) initiative under development at Columbia. The AI model is designed to assist in interpreting OCT reports by identifying patterns consistent with glaucomatous damage. Importantly, Dr. Hood emphasized that AI will not replace the clinician, but rather serve as a supplemental analytic tool to enhance diagnostic confidence. Preliminary data demonstrated excellent sensitivity and specificity, with formal publication expected in 2026.³ Finally, Dr. Hood shared encouraging news for ZEISS Cirrus OCT users: a version of the Hood Report—already available on Topcon and Heidelberg platforms—may soon be released for Cirrus after early 2026. This will bring the Columbia approach to a much broader user base and represents a welcome step forward for clinicians committed to evidence-based glaucoma detection.Clinical Insight:
Dr. Hood’s latest refinements highlight the ongoing evolution of OCT analysis—from pattern recognition rooted in RNFL anatomy to prototype-driven interpretation and AI-supported accuracy. For clinicians, these advances promise fewer missed diagnoses and a more precise understanding of structural damage.References
-
Hood DC, Raza AS, de Moraes CG, et al. Glaucomatous damage of the macula. Prog Retin Eye Res. 2013;32:1–21.
-
Hood DC, De Moraes CG, Liebmann JM, Ritch R. Improving the structure–function relationship in glaucoma using OCT and visual field data. Prog Retin Eye Res. 2021;81:100882.
-
Hood DC et al. Prototype-based analysis and AI integration in glaucoma OCT interpretation. (Conference presentation, AAO 2025; forthcoming publication, 2026).