Movi

Abstract

Background
One in three people requires rehabilitation, with the majority (90%) undergoing treatment at home. However, around 50% of patients do not fully adhere to their rehabilitation plans. IMU-based motion capture can assist both patients and rehabilitation specialists by addressing common challenges such as lack of motivation, time constraints, and incorrect execution of exercises. Integrating a visualisation tool into a smart garment using smart textile principles can offer an effective solution to these problems.

Project goal
The goal of this project was to create a proof of concept for an accessible smart shirt that integrates IMUs to capture and visualise upper-body movements for physical rehabilitation. The focus was on seamless integration, real-time motion visualisation, and aligning the design with patient needs.

Method
To ensure the concept aligned with the rehabilitation context and user needs, the project involved several key activities: desk research, expert interviews, benchmarking existing smart shirts, and prototyping. Insights were gathered from two rehabilitation doctors and two physiotherapists to understand rehabilitation practices and the needs of both patients and specialists. A benchmark of existing IMU-based smart garments helped identify ideas and challenges that needed addressing. A rehabilitation journey was developed to define the context, and a function diagram along with a list of requirements outlined the design criteria.

Prototype Development
The prototype development was divided into sections: motion capture visualisation, IMU integration, upper back design, wiring with conductive thread, and aesthetics. The motion capture visualisation focused on the electronics setup, rotational data gathering, and the use of OpenSense or Blender for visualisation. IMU integration addressed the placement and stability of the IMUs on the arms and their integration into the shirt. The upper back design explored the placement of electronics and the main node. Wiring with elastic thread focused on the conductive thread’s placement and sensor connections. These individual concepts were combined into the final design.

Results
Movi is the working prototype resulting from this project. It features five BNO055 IMUs, a TCA multiplexer, and a Xiao Seeeduino wired with conductive thread integrated in elastic for seamless integration. The system provides real-time movement visualisation through Blender. The shirt is designed with a snug fit, using a sturdy sports shirt as the base, and is costs 220 euro for the materials. It is designed for easy separation of electronics and textiles at the end of life.

To evaluate Movi a performance and user test were executed. The visualisation is clear, stable, and offers a 0.23-second delay. While the precision can be improved with calibration, a method has been proposed that patients can perform themselves. Feedback from users indicated that the shirt is comfortable and easy to wear, with a good fit for women (sizes S and M) and a slightly tighter fit for men.

Conclusion
Movi demonstrates the feasibility of using IMU-based motion capture garments for real-time visualisation within the rehabilitation sector. It offers a more affordable and integrated solution compared to existing systems. The smart shirt allows patients to use it independently during exercises, improving their rehabilitation process at home. The patient gets visual feedback in real-time. Movi can be further developed into various applications, such as motivation tools, gaming, exercise guidance, progress tracking, and discussion tools for rehabilitation specialists, ultimately enhancing both the patient experience and the specialist’s ability to track progress remotely.