Problem

Physiotherapy often relies on in-person sessions and subjective patient reporting, which can lead to gaps in adherence and real-time feedback. Patients struggle to perform prescribed exercises with correct form at home, delaying recovery and risking aggravation of injuries. Accessibility barriers such as transportation, scheduling, and geographic location further reduce consistent engagement with therapy plans.

We identified a need for a scalable, patient-centered solution that provides immediate guidance and objective metrics during home-based rehabilitation. Without robust monitoring, clinicians have to rely on sporadic assessments, making it difficult to tailor programs or detect compensatory movements early. To address this, we proposed an AI-powered AR platform that can guide exercises, monitor performance, and support remote care. AR feedback could be a game changer for adherence and outcomes.

Solution

We built an AI-powered AR-based physiotherapy website that leverages MediaPipe for real-time exercise monitoring and analysis. The system tracks sets, repetitions, timing, and form accuracy, providing patients with immediate visual and audio cues to correct movements during sessions. This enables consistent practice and helps ensure exercises align with the prescribed protocol. The platform also logs progress, enabling clinicians to adjust plans based on objective data.

Our solution delivers a data-driven feedback loop: patients see how they performed, clinicians access quantitative progress, and the AR interface adapts to individual needs. By combining motion tracking with exercise libraries, we can scale remote rehabilitation while maintaining high-quality care. The result is a more engaging, accountable, and efficient therapy experience for users and providers alike.

Impact

X% improvement in exercise adherence over baseline
Y% reduction in time-to-recovery compared to traditional care
Z% increase in objective accuracy of movements during home sessions

Quote of how incredible of an impact this had.

Team

Team Member #1 - AI Engineer
- Designed real-time motion analysis pipeline using MediaPipe
- Implemented AR feedback mechanisms and data logging
Team Member #2 - Product Designer
- Defined user workflows and accessibility considerations
- Shaped the UX for intuitive guidance and progress visualization

AI AR Physiotherapy for Remote Rehab