Recova

Recova Inspiration Orthopedic recovery often fails during the unsupervised months between clinic visits. We noticed that patients over 55 frequently lose progress because they lack real-time feedback. Recova was built to turn the home environment into a safe and supervised clinical space. What it does Recova uses browser-based computer vision to analyze patient movement in real time. It performs kinematic analysis on joint angles and provides immediate corrective cues to ensure proper form. The system generates automated performance digests for clinicians to monitor progress asynchronously. How we built it Frontend: Next.js and Tailwind CSS for a high-contrast and accessible interface. Computer Vision: MediaPipe Pose for real-time skeletal tracking and coordinate extraction. Intelligence: Claude API for transforming raw joint data into natural language clinical feedback. Infrastructure: Node.js with RabbitMQ to handle movement data ingestion and PostgreSQL for storing numerical performance metrics. Motion: Framer Motion for fluid and responsive UI transitions. Challenges we ran into Calculating precise 3D joint angles from a 2D camera feed required significant geometric normalization. We also focused heavily on reducing latency to ensure the feedback loop felt instantaneous for the user. Building a UI that felt premium yet remained simple enough for users with limited technical literacy was a constant balancing act. Accomplishments that we're proud of We successfully implemented a clinical-grade tracking system that requires zero specialized hardware. Our skeletal mapping remains stable even in low-light home environments. The feedback engine provides specific corrections that align with actual physical therapy standards. What we learned We gained deep insights into the UX requirements of the 55+ demographic and the importance of high-contrast design. We also refined our understanding of kinematic data processing and the ethical requirements of building non-diagnostic AI tools in the healthcare space. What's next for Recova We plan to expand our library of supported exercises to include complex multi-joint movements. Future iterations will integrate directly with Electronic Health Record systems to automate clinician billing for remote therapeutic monitoring. We are also exploring mobile optimization to support recovery on the go.