Performing specific physical exercises regularly is an important part of patient recovery after injury. To ensure patient compliance, these rehabilitation exercises are usually monitored by physiotherapists who also motivate the patient. In order to further improve outcome, promote patients' self-motivation and reduce costs, previous work focused on encouraging therapeutic exercises with computer games using human computer interaction devices.

Developing therapeutic games is a challenging task due to the fact that it requires an interdisciplinary team with clinical, engineering, and game design background. The aim of this project was to simplify the development process of therapeutic games by developing a software framework to unify tracking data of skeletal motion features by external sensors through abstraction. The proposed framework defines interaction device plugins which integrate manufacturer APIs into the framework in order to provide a unified view of interaction devices and tracking data for the game. The information of joint positions and orientations are extracted and transferred to the game via the User Datagram Protocol (UDP). Along with the framework, a game design cycle is put forward.

The framework was implemented in C# and device plugins for Microsoft Kinect 2.0 and Sixense Razer Hydra were developed to interface with the corresponding manufacturer API. Unity 5 was selected as the game engine, and game layer components were implemented. Standard visual components for limb segments were created, together with a range-of-motion assessment component. The proposed game development cycle was applied to the area of upper arm exercises after shoulder injury. It was shown that the framework can simplify inter-disciplinary therapy game development by providing a unified platform for team members' contributions, which can thus bring state-of-the-art game technology to therapeutic games.

I collaborated with Daniel Gonçalves and the ETH Zürich Rehabilitation Engineering Laboratory on this project.

2015

• Publication in Swiss Medical Informatics, Volume 31 (2015)