Jumping is one of the standardised tests for evaluating the functional performance of the lower extremities. Due to its complexity, it has been found useful in many sports/fitness and medical applications. Typical assessments include visual observation of the performed jumping task and evaluation with standard kinematic parameters. The need for designing more sophisticated assessment approaches has been widely recognised; however, there is no available solution on the market that provides objective visualisation and quantification of jumping patterns and techniques based on data recorded with simple and affordable wearable sensors.

The CanIJUMP project examines the concept of using wearable inertial sensors and intelligent analysis for the visualisation and quantification of jumping movement patterns. Artificial intelligence has been applied to detect valid jumps. An innovative approach based on multi-dimensional angle diagrams is used to visualise and interpret jumping patterns. New metrics were explored for quantifying jumping patterns and designed in a form that is understandable and intuitive for potential end-users. The project also examined how the proposed scientific concept behaves on untrained movement patterns recorded in a small cohort of untrained subjects. The proposed concept was integrated into a new, easy-to-use tool with an interactive interface suitable for various applications and end users.

This project is financed within the Proof of Concept Program of the Science Fund of the Republic of Serbia, #14908, supported by the SAIGE project of the World Bank.