MSCA PF candidates to develop and test the EvaLife healthy ageing initiative: an AI -based, personalised exercise recommendation platform for seniors

Over the next 50 years, the percentage of people over 65 years of age will nearly double in Europe, posing serious challenges to our healthcare and senior care systems. Physical exercise is a proven, cost-effective means of contributing to healthy ageing, but is often seen as a challenge by older adults. Delivering materially and financially accessible supervised exercise recommendations adapted to individual capacities and needs would substantially improve the quality of live and the autonomy off elderly adults.

University Jean Monnet in Saint-Etienne, France has developed the EvaLife system, an innovative exercise recommendation platform through collaboration between exercise physiologists at the Interuniversity Laboratory of the Biology of Motricity (LIBM Lab) and specialists in data mining and machine learning from the Hubert Curien Laboratory (LabHC) (Le Mat, Yann et al., 2024). The EvaLife system has five components: (1) a mobile assessment station for measuring key physical capacities (strength, balance, flexibility, aerobic fitness), (2) an AI-based algorithm that generates personalised exercise programmes based on individual capacities and constraints, and user feedback, (3) a smartphone application delivers exercise guidelines and collects user feedback (4) a secure data server that stores EvaLife resources and user data, with (5) a monitoring interface so that coaches and healthcare personnel can monitor user activity and adjust exercise recommendations if needed.

This two-year project will target adults aged 65 and over, including both relatively healthy individuals and sedentary seniors with co-morbidities, adapting supervision as required. The project will be multi-phased, starting with agile participant feedback home-based intervention to identify barriers, facilitators and user adherence factors, allowing to enhance the end-user smartphone interface and questionnaires, as well as intervention design. Following system updates, the main phase will enrol over 200 participants using different intervention strategies in different contexts, co-designing interventions with local partners according to local opportunities. The mobile assessment station can be move for site to site for demonstrations an receruitment. Implementation design and techniques should be recorded and compared to assess relative efficacy in different contexts. Participants will undergo a 24-week randomised controlled trial comparing three groups: (i) a home-based programme delivered through EvaLife, (ii) a supervised gym-based exercise programme, and (iii) a control group maintaining usual habits. Outcomes indicators will measure quality of life, fatigue, physical activity levels, mental health and key functional capacities. A complementary mechanistic sub-study will investigate the physiological and neuromuscular mechanisms underlying observed improvements, using advanced assessments such as neuromuscular function testing, body composition analysis and cardiopulmonary exercise testing. The project is expected to generate both robust scientific evidence and practical guidelines supporting the wider deployment of digital exercise solutions for ageing populations.