Junior Researcher Smart & Personalized AI Coaching Rehabilitation

The project

Acquired brain injury (ABI), including stroke, imposes a growing burden on Dutch healthcare. Every year thousands of patients face long-term limitations demanding rehabilitation and secondary prevention. Meanwhile, rehabilitation professionals face rising demand and staff shortages. Traditional rehabilitation programmes rely on in-clinic supervision, while current home-based training options are unstructured and poorly monitored. This leaves patients uncertain about exercise intensity and progress, therapists without timely information, and informal caregivers with limited support—undermining safety, efficiency, and outcomes.

TrAIn@home aims to design, integrate, and validate an AI-driven coaching platform for real-time, personalized, and safety-bounded home rehabilitation in adults with ABI. Multimodal data from wearables, context, and patient-reported outcomes are translated into actionable insights using AI-tools. Built-in safety rules and professional oversight ensure reliability and acceptance. Rather than replacing therapists, the platform strengthens and supports their role by providing structured information while giving patients guidance for safe, effective exercise at home.

By reducing unnecessary clinic visits, shortening waiting lists, and enabling more efficient allocation of therapist time, the TrAIn@home platform eases pressure on healthcare while lowering treatment costs. At the same time, it strengthens patient recovery through timely, personalized, and explainable feedback that enhances safety, motivation, and autonomy in home rehabilitation. Finally, TrAIn@home delivers validated multimodal datasets and integrates explainable, safety-bounded AI into clinical workflows.

This advances understanding of dose–response relations and movement quality in ABI rehabilitation, while establishing transferable methods and infrastructures for other chronic conditions. TrAIn@home thus provides a foundation for sustained innovation in evidence-based, AI-assisted rehabilitation.

The targeted starting date is 1 April 2026.

• Lead Work Packages 2 and 3 within the TrAIn@home project and coordinate project management and collaboration with partners (AUMC, HvA, SMEs).
• Design and develop study protocols and develop/evaluate AI algorithms (e.g., intensity, adherence) for home-based rehabilitation.
• Analyze movement data, validate findings in lab and living-lab studies, and conduct reliability, validity, and therapist–AI agreement analyses.
• Work with wearables, integrate movement data into usable models, and ensure secure data sharing under FAIR and GDPR principles.
• Recruit patients via rehabilitation centers and report findings through scientific publications, conferences, and structured meetings.
• Collaborate with interdisciplinary partners on ethics, implementation, and stakeholder engagement.