Exercise and nursing care needs—can AI train elderly people?

In this episode of the Digital Public Health Podcast, host Rasmus Cloes talks to Dr. Thomas Cordes, sports scientist at the University of Bremen, about a topic that is central to the future of elderly care: How can targeted training slow down the decline in older people in need of care – and what role can artificial intelligence play in this?

[Translate to Englisch:] Podcast-Thumbnail zur Episode 045. Links steht der Text „Bewegung gegen Pflegebedürftigkeit: Kann KI alte Menschen trainieren?“. Rechts ist ein Porträt von Dr. Thomas Cordes zu sehen. In das Porträt ist eine transparente Szene eingeblendet, die eine Trainingssituation mit älteren Personen und einer unterstützenden Person zeigt.

Thomas explains that his doctoral thesis shows how even very elderly, multimorbid residents in inpatient care facilities benefit from structured training programs. Although there are no major leaps in performance, motor skills, cognition, and psychosocial well-being remain stable, while people without training decline significantly over the same period.

In the AI-aktiv project, Thomas is now investigating how a chatbot can be used as a digital training assistant. The AI is intended to support nursing and care staff in planning and adapting exercises – not as a replacement for human expertise, but as an assistance system that makes suggestions that are then reviewed by experts.

At the same time, it becomes clear where today's systems still have their limitations. The lack of memory over longer periods of time is particularly problematic. Individual training thrives on the fact that previous stresses, progress, or setbacks remain present—this is precisely where the greatest weakness currently lies.

Nursing staff are responding surprisingly openly to this new technology. Although many have had little experience with tablets or chatbots to date, they report curiosity and interest in interviews. They see AI less as an additional burden and more as an opportunity to improve their own actions in terms of training science.

In the end, Thomas paints a vision in which caregivers use AI to access exercise videos in seconds, receive alternatives for residents who have fallen or have limited mobility, and even receive direct audiovisual feedback on their exercise performance. This makes prevention scalable – and finally puts exercise where it belongs: in the everyday routine of elderly care.

Watch/listen now and gain exciting insights:YouTube, Spotify, Apple, Dezer

More information:Website, Instagram