# AI based tool for monitoring intensity and fatigue in elite women handball

## TL;DR

This paper introduces an AI tool that helps track and predict performance and fatigue in elite women handball players using wearable data.

## Contribution

The novel contribution is an AI tool combining predictive accuracy and explainability for monitoring athlete intensity and fatigue.

## Key findings

- The AI tool accurately predicts key performance indicators like running distance and speed.
- The model provides actionable insights for optimizing training and lineup strategies.
- The approach can be extended to predict other physiological KPIs in handball.

## Abstract

We propose an AI-based tool to predict and monitor Key Performance Indicators (KPIs) for player’s activity such as running distance and speed from wearable devices. These KPIs serve as proxies for intensity and fatigue levels in professional athletes. Applied to a women’s professional handball team competing at the EHF Champions League level, our model helps predict player workload and physiological stress, enabling accurate monitoring of player condition. By combining predictive accuracy with explainability methods, our tool not only forecasts fatigue and intensity metrics but also provides actionable insights for coaching staff to optimize training and lineup strategies. This work demonstrates the potential of advanced machine learning methods and can be extended to the prediction of any physiological KPI to support handball performance monitoring.