LOUISA - Learning Model for Multidimensional Quantitative Movement Analysis

General overview

Musculoskeletal (MSK) diseases represent one of the main health risks in the workplace that result in chronic pain that significantly reduces the quality of life to those affected and can lead to disabilities. The aim of LOUISA is to motivate and support people in the workplace. Employers actively prevent and improve employees' MSK health in order to keep people healthy. 

Motivation

The well–being of the individual is inextricably linked to her physical health. Diseases of the musculoskeletal system represent one of the main disease risks in the workplace and often result in chronic pain conditions that significantly reduce the quality of life of those affected and lead to disability.

Way of Working

LOUISA researches, develops and validates a learning model for multidimensional quantitative movement analysis. We expect synergy effects between the individual dimensions that advance the state of research and technology in both movement and pain analysis. New machine learning methods will lead to much more precise expert systems when analyzing abnormal movement. Testing and validation takes place under laboratory conditions as well as in real ecosystems with employees, employers, health providers and insurance companies.

Innovations and Perspectives

LOUISA analyzes the entire musculoskeletal system and any pain objectively using the 2D camera of a mobile phone and uses artificial intelligence (AI) to calculate a weighted movement score.

Ethical implications

Transparency, coherent privacy protection, and secure data, as well as user-centered development form the basis for gaining our user's trust. They are core components for common good technology development that will be accepted by users at scale. Among the several ethical and philosophical questions risen by the LOUISA project are:

• Can pain be recognized and objectified in the movement?
• If so, can that be transferred to a mobile APP?
• Are there predictive parameters for indicating risk of pain?
• Can they be measured?
• Does this analysis lead to actions and impacts on the individuals’ well-being?