In physical therapy, especially when dealing with therapeutic small devices such as bands or dumbbells, correct exercise execution is crucial. Qualified feedback is important for maintaining or restoring physical health. Compared to the frequency with which the exercises should be performed, supervised exercise sessions are usually limited due to time and cost reasons.

Objectives and Approach

The research team involved in the project THERESA aims to make a significant contribution to therapy support by developing a system that uses artificial intelligence (AI) to analyze videos showing physiotherapeutic exercises. The AI recognizes both the exercises themselves and the therapeutic small devices, and precisely determines the training intensity based on the forces generated during the exercises. Thus, the system does not merely count movements and repetitions, but also analyzes the biomechanical processes for individual performance evaluation. This enables tailored training, which is enhanced with live feedback and playful, thus motivating instructions.

Innovations and Perspectives

The system developed in the project supports patients during rehabilitation even outside of clinical settings. Improved exercise analysis and innovative gamification approaches help in forming long-term habits. Therapists can monitor developments at any time and provide targeted oversight. This remote monitoring capability enhances patient adherence and engagement with their rehabilitation exercises, leading to potentially better outcomes in their recovery process.

Project Partners

Fundings

Federal Ministry of Education and Research (BMBF, Germany) in the project THERESA (project number 16SV9257).