AI based rehabilitation: the way forward in addressing unmet needs in musculoskeletal disease

Musculoskeletal (MSK) conditions are the leading cause of disability worldwide and, in European Union countries, account for up to 17% of years lived with disability and around 2% of gross domestic product (GDP) in direct and indirect costs. Despite universal health coverage and a doubling of public rehabilitation prescriptions in the past decade, unmet rehabilitation needs persist in Portugal, alongside growing regional disparities, long waiting times, and a heavy reliance on private services for physical rehabilitation. These factors undermine both clinical and economic outcomes. International and national evidence indicates that rehabilitation delivered through AI-enabled programmes is feasible and potentially effective, can be deployed at scale, and may reduce barriers related to geography, scheduling, and limited rehabilitation facilities. Such solutions may help improve continuity of care, shorten waiting times, and address unmet needs, but large-scale adoption requires robust frameworks for clinical evaluation and validation, patient selection, professional training, and outcome monitoring, often within hybrid models of care. By explicitly addressing potential benefits, risks, limitations, and clinical criteria, the rehabilitation community can facilitate responsible and ethical integration of AI-supported and digital models into rehabilitation practice and research, while managing the organisational and cultural changes needed to incorporate these models as complementary interventions within health systems. Drawing on WHO and OECD recommendations and on recent Portuguese implementation experience, this perspective examines how AI-driven rehabilitation may support more equitable, timely, and efficient responses to MSK rehabilitation needs, particularly for physician-prescribed care delivered under medical supervision in the home setting.