Intelligent Systems Research Group
KIARA
AI-based generation of 4D avatars for joint angle detection with smartphones for orthopedic applications
Team:
Prof. Dr. Astrid Laubenheimer
Robin Baumann M.Sc.
Funding: MWK
Duration: 2022 - 2024
Overview
Orthopedic diagnostics and therapies often require the capture of joint angles in dedicated motion sequences. While clinical research in the university environment has laboratories equipped with motion capturing (MoCap) systems, such cost- and space-intensive systems are not available in other areas - in particular in outpatient physiotherapy but also in the use of weight training equipment in home training.
In the KIARA project, an AI method for markerless joint angle acquisition for orthopedic applications is being developed. It will combine both the acquisition of the required video data and the calculation of the joint angles on RGB-D capable smartphones, thus realizing privacy-by-design principles. The main methodological innovation compared to existing MoCap systems lies in the use of data sparingly generated morphable models, which can significantly reduce sensor data quality requirements.
Partners & Funding
In the project, the ISRG is responsible for method research and development as well as project management. inovex GmbH will handle the app development including the porting of the researched methods to smartphones. Vitracom GmbH supports the project in an advisory capacity with its expertise in the field of time-synchronized sensor systems for data acquisition. The Clinic for Orthopedics and Trauma Surgery at Freiburg University Hospital is an associated partner in the project and allows the project partners access to its laboratory to record annotated Mocap data as part of a current "Return-to-Play" study with soccer players.
The project is funded by the Ministry of Science, Research and Arts of Baden-Württemberg (MWK).
