Skip to content
Conference Open access

AI-Powered Rehabilitation Exercise Monitoring: A Deep Learning Framework with RAG-Enhanced Feedback and Computer Vision

Sep 2026 · 2026 Forth International Conference on Networks, Multimedia and Information Technology (NMITCON) · pp. 1-7 · 0 citations · 16 references

Abstract

Musculoskeletal disorders impact more than 1.3 billion people globally, and rehabilitation in the home setting is crucial to the process. Due to the lack of professional oversight, patient compliance usually dips under 35%. This study describes RehabSense, an AI-powered software framework aiming to fill this void. MediaPipe pose recognition coupled with a combination of CNN and LSTM results in excellent performance at exercise classification and repetitions counting. The unique pipeline based on the Google Gemini API performs retrieval augmented generation (RAG), enabling us to convert biomechanical indicators into coaching recommendations that are natural and contextually relevant. The infrastructure of this project includes FastAPI, React.js, and WebSockets, allowing less than 100 ms end-to-end delay that ensures instantaneous feedback needed for form correction. The tests performed with seven users showed that this solution was much more efficient at improving the users form and self-confidence compared to the case when patients train on their own.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.