VIS4ION-FiT (Visually Impaired Smart Service System for Spatial Intelligence and Onboard Navigation for Fitness)

New York University School Of Medicine, New York NY

Investigators

Abstract

PROJECT SUMMARY/ABSTRACT Persons with blindness or low vision (PBLV) lead more sedentary lifestyles, with reduced mobility, lower physical activity, and an increase in associated non-communicable diseases. PBLV often fear travel/journeys, given frequent dependence on others and the looming threat of getting lost or stranded. PBLV also face substantial increases in mechanical trips, falls, and fractures. These barriers lower their health outcomes, some of which could be mediated by physical exercise. Despite an intensive evidence base supporting physical activity to promote health, there is a large practice gap for PBLV in almost every country, including urban/city centers, such as NYC. Digital technologies can close this access and implementation gap by providing wayfinding and scene understanding support to guide safe exercise within public parks in a manner that does not require additional physical infrastructure (e.g., signs, beacons). In this proposal, we will implement our wearable and smartphone application, VIS4ION-FiT (Visually Impaired Smart Service System for Spatial Intelligence and Onboard Navigation for ¬Fitness) and evaluate effectiveness to improve physical activity in PBLV. VIS4ION-FiT is a customizable personal mobility solution that uses artificial intelligence (AI) to process camera/sensor data, both locally on the system, and remotely on servers, to provide safe guiding instructions delivered as speech, audio alerts, and tactile feedback. Our central hypothesis is that PBLV who use VIS4ION-FiT will increase their ‘moderate-equivalent’ minutes of physical activity per week and therefore improve personal health metrics (blood pressure, resting heart rate, weight). The proposal has three aims to validate this approach. First, we will interview 10 PBLV to identify key access barriers to physical activity. Then, we will update our AI-based navigation technology with GPS to create a more robust and precise hybrid solution for wayfinding. We will map walking routes and places of interest in large public parks (e.g., Central Park in NYC). User testing (25 sighted, 25 PBLV) will include the evaluation of our navigation and scene understanding assistance. Second, we will conduct a randomized controlled crossover trial with 40 PBLV using VIS4ION-FiT for 6 months over a 12-month period, providing everyone with park access for exercise. We will measure physical activity metrics, adherence, and health-related changes across the 12-month period. Our primary outcome is moderate-equivalent minutes of physical activity per week, with daily steps, blood pressure, resting heart rate, and weight, as secondary outcomes. Users will also provide feedback on VIS4ION-FiT’s acceptability, appropriateness, and feasibility. Third, we will perform a process evaluation of our implementation strategy using the RE-AIM framework, gathering data on how the intervention facilitated physical activity and health. Collective results will guide future directions to enhance physical activity for PBLV in urban/city areas across the US.