Performance metrics of ML models.

Accurate assessment of training status in team sports is crucial for optimising performance and reducing injury risk. This pilot study investigates the feasibility of using machine learning (ML) models to estimate oxygen uptake (VO2) with wearable sensors during team sports activities. Six healthy male team sports athletes participated in the study. Data were collected using inertial measurement units (IMU), heart rate monitors, and breathing rate sensors during incremental fitness tests. The performance of different ML models, including multiple linear regression (MLR), XGBoost, and deep learning models (LSTM, CNN, MLP), was compared using raw and engineered features from IMU data. Results indicate that while LSTM models with raw IMU data provided the most accurate predictions (RMSE: 4.976, MAE: 3.698 ), MLR models remained competitive, especially with engineered features. Multi-sensor configurations, particularly those including sensors on the torso and limbs, enhanced prediction accuracy. The findings demonstrate the potential of ML models to monitor VO2 noninvasively in real-time, offering valuable insights into the internal physiological demand during team sports activities.

团队项目运动中训练状态的精准评估，对于优化运动表现、降低运动损伤风险至关重要。本预实验探究了在团队项目运动场景下，借助可穿戴传感器利用机器学习（machine learning, ML）模型估算摄氧量（oxygen uptake, VO2）的可行性。共有6名健康男性团队项目运动员参与本研究。研究过程中，通过惯性测量单元（inertial measurement unit, IMU）、心率监测仪以及呼吸速率传感器，在递增负荷体能测试期间采集相关数据。本研究对比了多种机器学习模型的性能，包括多元线性回归（multiple linear regression, MLR）、XGBoost以及深度学习模型（长短期记忆网络LSTM、卷积神经网络CNN、多层感知机MLP），对比所用特征涵盖IMU原始数据与经特征工程生成的特征两类。研究结果显示，尽管采用IMU原始数据的LSTM模型可实现最精准的预测（均方根误差（Root Mean Squared Error, RMSE）：4.976，平均绝对误差（Mean Absolute Error, MAE）：3.698），但MLR模型在使用经特征工程生成的特征时仍具备相当的竞争力。多传感器配置方案，尤其是在躯干与四肢部署传感器的配置，可有效提升预测精度。本研究结果证实，机器学习模型具备无创实时监测VO2的潜力，可为团队项目运动过程中的内部生理负荷评估提供极具价值的参考依据。