Believing end-stage glaucoma management was lacking an appropriate characterization of central visual field (VF) loss, researchers recently used artificial intelligence (AI) to quantitatively identify and systematically study central VF loss patterns and their shifts in end-stage glaucoma. An unsupervised AI model determined 14 central VF patterns, most of which preserve the inferotemporal regions entirely or partially.

The retrospective cohort study collected data over a span of 15 years from a total of 2,912 reliable 10-2 VFs of 1,103 eyes from 1,010 patients measured after end-stage 24-2 VFs. The model determined the representative central VF patterns in end-stage glaucoma, including patterns of nearly intact VF, nearly total loss, total loss, temporal sparing, quadrant sparing, nasal loss and central islands.

For end-stage glaucoma,  the team found that VF damage in more vulnerable zones could take different forms. At one-year and two-year follow-up, new defects mostly occurred in the more vulnerable, superonasal, zone than in the less vulnerable inferotemporal zone. The researchers discovered that initial encroachments on an intact central visual field at follow-up in end-stage glaucoma were more likely to be nasal loss. One of the nasal loss patterns had a substantial chance at two-year follow-up to shift to total loss, whereas others did not.

“Enhanced understanding of patterns of central VF defects and their development over time is of great significance to improve management of end-stage glaucoma,” the researchers wrote in their paper.

Wang M, Tichelaar J, Pasquale LR, et al. Characterization of central visual field loss in end-stage glaucoma by unsupervised artificial intelligence. JAMA Ophthalmol. January 2, 2020. [Epub ahead of print].