Researchers develop collision-warning device for visually impaired patients

Reviewed by Shrinivas Pundlik, PhD, and Gang Luo, PhD.

The real-world evaluation of a novel device designed to reduce collisions and thus improve daily walking showed that it decreased the rate of collisions by 37% when used by blind patients and those whose vision was severely impaired,¹ according to lead author Shrinivas Pundlik, PhD, and senior author Gang Luo, PhD.

Pundlik and Luo are both from the Schepens Eye Research Institute of Massachusetts Eye and Ear and the Department of Ophthalmology of Harvard Medical School, Boston, where the device is being developed.

Pundlik and Luo also pointed out that there is little evidence available about the real-world mobility benefit of electronic mobility aids.

Testing the device

The device includes a video camera and a microcomputer enclosed in a slingback bag. When users wear the bag, the camera is level with their chest. The authors conducted a double-masked randomized clinical trial from January 2018 to December 2019 to determine the usefulness of the collision warning system for visually impaired patients during daily walks over the course of 4 weeks.

During the trial, the device switched itself randomly from active mode (referred to as the intervention condition), which provides alerts to the wearer via 2 vibrotactile wristbands when the potential for a collision arises, and silent mode, the control condition that detects the threat of collisions without providing warnings.

The 31 study participants (18 men; median age, 61 years; range, 25-73) were volunteers who were independently mobile and used a long cane or guide dog during daily activities; all had severely impaired vision and some were blind with restricted peripheral visual fields.

The main outcome measure—which was compared between the 2 modes for each participant—was the rate of contact with obstacles normalized by the number of hazards and hours of walking.

The investigators explained that the device records videos when it sends the collision warning. Two independent reviewers evaluated and scored potential hazards detected by the device, including direct contact with obstacles. The participants and reviewers did not know the mode in which the device operated at any given time.

Findings

Nineteen of the 31 participants (61%) had light perception or worse vision; almost all (90%) used a long cane as a mobility aid.

The results showed that the median number of contacts was lower when the device was in active mode compared with silent mode (9.3; interquartile range [IR], 6.6-14.9 compared with 13.8; IR, 6.9-24.3, respectively), a difference that reached significance (P<0.001).

When the investigators controlled the analysis for demographic characteristics, such as vision and a history of falls, the rate of contacts decreased significantly when the device was in active mode compared with silent mode (rate ratio = 0.63; 95% confidence interval, 0.54-0.73; P<0.001).

The authors pointed out that the ability to review the contacts recorded in the videos and to switch between 2 modes added value to their findings because actual use and collision data could be collected, the correctness of which did not rely on self-reporting from participants or the bias of reviewers.

“In this study involving 31 visually impaired participants, the collision warnings were associated with a reduced rate of contacts with obstacles during daily mobility, indicating the potential of the device to augment habitual mobility aids.”

Shrinivas Pundlik, PhD

E: shrinivas_pundlik@meei.harvard.edu

Pundlik and are Luo listed as inventors on a patent related to this collision-warning technology, which is assigned to Massachusetts Eye and Ear, and are cofounders of EyeNexo, a company involved in mobile vision measurement technology that is unrelated to the present research.

Reference

Pundlik S, Baliutaviciute V, Moharrer M, Bowers AR, Luo G. Home-use evaluation of a wearable collision warning device for individuals with severe vision impairments: a randomized clinical trial. JAMA Ophthalmol. Published online July 22, 2021. doi:10.1001/jamaophthalmol.2021.2624

Related Content: Ophthalmology | Ophthalmic News | Imaging