Can Machine Learning with IMUs Be Used to Detect Different Throws and Estimate Ball Velocity in Team Handball?

Roland Van Den Tillaar,Tom Stewart,Shruti Bhandurge

doi:10.3390/s21072288

Roland Van Den Tillaar, Tom Stewart + Show 1 more

Open Access

https://doi.org/10.3390/s21072288

Copy DOI

Abstract

Injuries in handball are common due to the repetitive demands of overhead throws at high velocities. Monitoring workload is crucial for understanding these demands and improving injury-prevention strategies. However, in handball, it is challenging to monitor throwing workload due to the difficulty of counting the number, intensity, and type of throws during training and competition. The aim of this study was to investigate if an inertial measurement unit (IMU) and machine learning (ML) techniques could be used to detect different types of team handball throws and predict ball velocity. Seventeen players performed several throws with different wind-up (circular and whip-like) and approach types (standing, running, and jumping) while wearing an IMU on their wrist. Ball velocity was measured using a radar gun. ML models predicted peak ball velocity with an error of 1.10 m/s and classified approach type and throw type with 80–87% accuracy. Using IMUs and ML models may offer a practical and automated method for quantifying throw counts and classifying the throw and approach types adopted by handball players.

Highlights

Handball is a popular sport with around 20 million competitors worldwide
The gradient boosting machine (GBM) model with the high-g features performed slightly better for classifying throw type, this was not a significant difference (+2.4% balanced accuracy, p = 0.092)
This study investigated whether an inertial measurement unit (IMU) and machine learning could detect different This study investigated whether an IMU and machine learning could detect different types of team handball throws and predict ball velocity

Summary

Introduction

The physical attributes of the sport include a high playing tempo, a high volume of throws, frequent physical contacts, collisions, and rapid changes of movement [1,2]. Athletes need to train frequently, often daily for elite players. Extensive training load is detrimental to player performance and heightens injury risk. This can cause cumulative tissue overload [3], which is a contributing factor to overuse injuries. Shoulder injuries are highly prevalent in handball players [4]. In baseball (another overhead sport with throwing that is somewhat comparable to handball), half of youth pitchers reported shoulder pain during the competition season [6,7]. It is reported that shoulder pain has an impact on athletes’

Objectives

Results

Conclusion