Perspective from Derek MacDonald, OD, FAAO
Source/Disclosures
Disclosures: The authors were funded by the NIH, NEI grants and in part by an unrestricted grant from Research to Prevent Blindness and BrightFocus Foundation.
June 02, 2020
1 min read
Save

AI-enabled dashboard provides map for visual field loss in glaucoma

Perspective from Derek MacDonald, OD, FAAO
Source/Disclosures
Disclosures: The authors were funded by the NIH, NEI grants and in part by an unrestricted grant from Research to Prevent Blindness and BrightFocus Foundation.
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

An AI-enabled dashboard designed to monitor glaucomatous functional loss showed potential for providing clinicians with a user-friendly tool for determining the severity of glaucoma-related vision deficit, the spatial extent of damage and means for monitoring progression.

“Detecting functional vision changes due to glaucoma is critical to making treatment decisions with the goal of preserving vision and maintaining quality of life,” Siamak Yousefi, PhD, from the University of Tennessee Health Science Center, and colleagues wrote. “However, most of the approaches for functional assessment of glaucoma through [visual field maps] have several limitations that pose critical challenges to their clinical utility.”

The study comprised 31,591 visual field maps from 8,077 patients. Yousefi and colleagues entered the visual field data from the most recent visits of both patients with and without glaucoma into a “pipeline” that included principal component analysis, manifold learning and unsupervised clustering to identify eyes with similar global, hemifield and local patterns of visual field loss

Unsupervised machine learning identified 32 dense clusters after excluding 38% of the mapped visual fields located on shallow regions of the dashboard. The researchers decomposed the visual field maps at each cluster to 17 different archetypal patterns and identified the archetypal patterns of visual field loss that are prevalent in each cluster to specify the characteristic local patterns of loss at different dashboard regions.

Data revealed that different regions of the dashboard have varying local visual field patterns. Each region of the dashboard was representative of the following: a particular global functional severity, an extent of visual field loss into different hemifields, and characteristic local patterns of visual field loss.

“The AI provided an explainable and clinician-friendly tool with multiple layers of glaucoma knowledge on a simple interpretable 2-dimensional map, which we have termed the ‘dashboard.’ The dashboard not only represents the visual functional assessment of a patient with glaucoma but also overlays the trajectories on visual functional information from thousands of past patients with glaucoma,” Yousefi and colleagues concluded. – by Talitha Bennett