A prototype suprachoroidal retinal prosthesis enables improvement in a tabletop object detection task

Abstract

Purpose: The Bionic Vision Australia consortium has been developing a suprachoroidal retinal prosthesis, which has an easily accessible surgical location, good device stability and minimal adverse events over a 2-year period. We have previously reported that this device improves light localisation and optotype recognition. The aim of this study was to investigate device efficacy using a tabletop object recognition task, which forms part of the Low Vision Assessment of Daily Activities (LoVADA) protocol.

Methods: Two subjects with bare light perception from retinitis pigmentosa were implanted with a 24-channel suprachoroidal retinal prosthesis prototype. Four white household items (placemat, dinner plate, coffee mug, fork) were placed in random locations on a black table. Performance was scored by an occupational therapist, on the number of items (a) correctly identified and (b) accurately touched; results of both providing a overall performance score. Time was recorded for all assessments, and not restricted. Assessment was completed at baseline (no device), at the first ADL testing session, and again six months later. Device on and device off conditions were randomly presented. Usage of the device was restricted to the laboratory, thus limiting opportunity to learn between sessions.

Results: The rate of accurate touch improved at each session with device-on. By the six-month session, one subject was able to accurately visually locate and touch 100% of the items with device on (vs. 0% with device off), whilst the other subject touched 75% with device on and 25% with device off. However, subjects were not able to reliably identify the household items by name. The task took more time to complete with the device on (average 4.5 minutes) vs. device off (average 2 mins) & baseline (average 42 seconds).

Conclusions: The prosthesis improved the ability to visually locate and touch tabletop items. Whilst not able to name items, subjects believed they would be able to better interpret the information with more practice and training. These results are consistent with other retinal prosthesis groups, and show that the suprachoroidal implant can provide useful visual information in a static real world setting. Clinical testing of a second generation “take-home” device is planned, where it is anticipated that functional outcomes will be improved through increased experience with the device and neuroplasticity.