Mobility improvement of patients with peripheral visual field losses using novel see-through digital spectacles

• Published in: PloS one Vol. 15; no. 10; p. e0240509
• Main Authors: Sayed, Ahmed M, Shousha, Mohamed Abou, Baharul Islam, M D, Eleiwa, Taher K, Kashem, Rashed, Abdel-Mottaleb, Mostafa, Ozcan, Eyup, Tolba, Mohamed, Cook, Jane C, Parrish, 2nd, Richard K
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 14-10-2020; Public Library of Science (PLoS)
• Subjects: Algorithms; Augmented reality; Binocular vision; Biology and Life Sciences; Biomedical engineering; Calibration; Cameras; Clinical trials; Computer engineering; Computer science; Deactivation; Design defects; Eye; Eye movements; Eyeglasses; Eyewear; Female; Glaucoma; Health aspects; Human locomotion; Humans; Low vision aids; Male; Maneuverability; Measurement; Medicine and Health Sciences; Mobility; Object recognition; Parameters; Patients; Physical Sciences; Placebo effect; Placebos; Prospective Studies; Research and Analysis Methods; Scanning; Social Sciences; Testing; Virtual Reality; Vision Disorders - physiopathology; Vision Disorders - rehabilitation; Visual aids; Visual field; Visual Field Tests; Visual fields; Visual Fields - physiology; Walking; Walking - physiology; Wearable computers; United States; United States > US; Egypt; California
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0240509

To evaluate see-through Augmented Reality Digital spectacles (AR DSpecs) for improving the mobility of patients with peripheral visual field (VF) losses when tested on a walking track.
Prospective Case Series.
21 patients with peripheral VF defects in both eyes, with the physical ability to walk without assistance.
We developed the AR DSpecs as a wearable VF aid with an augmented reality platform. Image remapping algorithms produced personalized visual augmentation in real time based on the measured binocular VF with the AR DSpecs calibration mode. We tested the device on a walking track to determine if patients could more accurately identify peripheral objects.
We analyzed walking track scores (number of recognized/avoided objects) and eye tracking data (six gaze parameters) to measure changes in the kinematic and eye scanning behaviors while walking, and assessed a possible placebo effect by deactivating the AR DSpecs remapping algorithms in random trials.
Performance, judged by the object detection scores, improved with the AR DSpecs (P<0.001, Wilcoxon rank sum test) with an average improvement rate of 18.81%. Two gaze parameters improved with the activated algorithm (P<0.01, paired t-test), indicating a more directed gaze on the central path with less eye scanning. Determination of the binocular integrated VF with the DSpecs correlated with the integrated standard automated perimetry (R = 0.86, P<0.001), mean sensitivity difference 0.8 ± 2.25 dB (Bland-Altman).
AR DSpecs may improve walking maneuverability of patients with peripheral VF defects by enhancing detection of objects in a testing environment.