Novel research on predicting functional loss in glaucoma patients in collaboration with University of Bern and Inselspital.

This month PeriVision founders Serife Kucur and Patrick Kessel travelled through the US to visit partners and potential customers for their virtual eye testing platform with an initial focus on glaucoma diagnosis and monitoring.

One of the highlights of the trip was attending the Imaging & Perimetry Society (IPS) conference in Berkeley, California. There Davide Scandella, a computer scientist at the University of Bern’s ARTORG Center for Biomedical Research, presented findings from a joint research project between the University of Bern, Inselspital Bern and PeriVision.

The team used machine learning techniques to investigate structure-function relationships in glaucoma patients. They predicted visual field loss (functional) based on optical coherence tomography (structural) data. One issue with visual field data is that it is very noisy, i.e. often unreliable due to fatigue effects, the uncomfortable body position or anxiety. Naturally, such noisy data makes predictions harder.

PeriVision’s novel approach to perform visual fields portable virtual reality headsets and a suite of AI algorithms to make the measurements faster and more reliable will help obtain more frequent and higher-quality functional data in the future.