MovAI

Years: 2022-2024 Location: Uppsala, Sweden

In collaboration with: Uppsala University, Northern Arizona University’s Playful Health Technology Lab, KTH Royal Institute of Technology and Umeå University.

Movement sensing technologies have opened up physical rehabilitation and training to the inclusion of modern machine learning (ML) techniques for personalization. The paramount standard for evaluating the success of a ML model has long been its predictive power and accuracy, but even a gold-standard accuracy benchmark fails when it inappropriately misrepresents disabled and minority bodies, and people with diverse physical capabilities.

In MovAI, we seek to reframe the roles of ML for physical rehabilitation through two case studies. In the first case, we engaged in a participatory work co-creating exergames by employing ML and its training as a source of play and motivation rather than an accurate diagnostic tool for children with and without Sensory Based Motor Disorder.

In the second case study (still ongoing), we explore the potential of ML to support personalization of physiotherapy in ways that do not aim at replacing the physiotherapist’s expertise, but rather enhance it. This case study involves a cohort of physiotherapists, interaction designers and ML as expert informants.

PUBLICATIONS:

Laia Turmo Vidal and Jared Duval. 2024. Ambiguity as a Resource to Design for a Plurality of Bodies. In Halfway to the Future Symposium, 2024.

Jared Duval, Laia Turmo Vidal, Elena Márquez Segura, and Annika Waern. 2023. Reimagining Machine Learning’s Role in Assistive Technology by Co-Designing Exergames with Children Using a Participatory Machine Learning Design Probe. In ACM ASSETS 2023.