Intelligent Monitoring for Neuro-Degenerative diseases through connected insoles and self-supervised AI
Duration and funding: 6 months – €71.6 k
MIND-SCAN aims to explore the feasibility of inferring pelvic and leg motricity solely from plantar activation data, leveraging self-supervised AI to model human movement while integrating biomechanical constraints. By accurately estimating joint positions, this research could revolutionize non-intrusive monitoring of elderly and disabled individuals.
CONTEXT
Increasing life expectancy and an aging population pose major challenges for maintaining the quality of life and autonomy of the elderly and disabled individuals. At a certain stage in life, depending on a person’s state of health and level of dependency, a decision has to be made as to whether they can remain at home or need to be placed in a specialized facility. This transition is often perceived as a break with the past and the memories associated with it, making the process emotionally difficult for the elderly. However, home care requires the use of medical devices to monitor physical and mental well-being. As life expectancy increases and the population ages, the need to develop solutions that promote independence and maintain quality of life for the elderly becomes ever more urgent.
CHALLENGES
Existing human motor analysis systems, whether used at home or in specialized facilities, are often perceived as intrusive due to their reliance on surveillance cameras or motion capture technologies. Despite their high accuracy, this limitation affects their acceptance. As a result, the scientific community is increasingly exploring foot pressure analysis through connected insoles. However, most research in this area focuses on specific metrics such as ground contact time, step length, or cadence. While these parameters provide valuable insights into physical activity, a comprehensive assessment of psychomotor disorders, crucial for predicting and evaluating health decline, requires a more detailed analysis of an individual’s motor signature for a reliable diagnosis. This project, therefore, explores the feasibility of extracting more complex information from connected insoles with a reliability comparable to vision-based systems.
RESULTS
Coming Soon