PI Probaligence Gmbh

Smart Injury Prevention

Artificial intelligence recognizes the risk of injury during sports
The “Smart Injury Prevention” research project starts with the aim of using artificial intelligence (AI) to detect potential running injuries in a timely manner and thereby prevent them. For this purpose, the specialist for body-hugging sensors Humotion GmbH and the software expert PI Probaligence GmbH, together with the University of Hamburg, are developing intelligent software for the personalized prediction of running injuries. The project will be funded over the next three years with a funding volume of around 530,000 euros from the Central Innovation Program for SMEs (ZIM).

19 million active athletes make running one of the most popular sports in Germany. There is a high risk of injury, especially due to overuse, which causes running-related injuries in 80% of active participants once a year. In order to be able to recognize the usually gradual development and prevent an injury, the analysis of biomechanics or the individual running style represents a promising approach. The current analysis method – usually on the treadmill as part of video-supported laboratory diagnostics – is possible due to the limitations of the artificial environment and the time limit does not provide effective injury prevention.

The latest research results on the assessment of individual injury risks indicate a rethink, away from individual risk factors and towards individual injury patterns. However, the technical requirements for a continuous recording of running biomechanics are currently lacking, both in the area of data collection and parameter calculation as well as in the area of data analysis and the development of suitable prediction models.

Real-time analysis of running biomechanics
For this reason, the three research partners are jointly developing “Smart Injury Prevention”, an AI to prevent running-related sports injuries. The focus of development is on novel machine learning methods from partner PI Probaligence for processing large, highly multimodal and non-linear amounts of data, which are used to identify injury-relevant parameters and to estimate the individual risk of injury. In addition to classic data such as GPS and heart rate, the biomechanical parameters are to be recorded based on high-resolution sensor data from Humotion, transferred to a server system and used there for ongoing training of the machine learning networks. This is intended to enable real-time analysis of running biomechanics for athletes during training and competitions and to create the database for personalized movement analysis and running style recognition. The systems trained in this way will later run on common sports watches and apps for smartphones and serve as a basis for calculating the individual risk of injury and optimizing training. As part of the research project, the University of Hamburg is developing a monitoring system for the identification, classification and monitoring of running injuries and is carrying out the series of tests with the runners.

The idea for the “Smart Injury Prevention” project came about as part of the innovation network INTELLUS – Intelligent Support Systems, which is funded by the Central Innovation Program for SMEs (ZIM). As part of membership, partners are actively supported in implementing R&D projects and securing financing. INTELLUS is supported by IWS GmbH, which also takes over the application management of the cooperation projects and intensively supports the members in the development of new technologies.

Stay Updated

Join Our Newsletter

Get regular updates about PI’s new developments and activities in the field of AI and machine learning.


Social Networks

PI Probaligence GmbH

(c) 2023 All Rights Reserved