<p>Sports injury is a core issue constraining professional athletes’ competitive performance and health, and the scientific management of training load is a key intervention for injury prevention. However, existing machine learning injury prediction models generally neglect the temporal dynamic characteristics and domain-specific semantics of the Acute:Chronic Workload Ratio (ACWR), resulting in limited prediction accuracy and clinical interpretability. This study proposes the ACWR-LSTM framework, which explicitly integrates dual-pathway ACWR feature engineering based on Rolling Average (RA) and Exponentially Weighted Moving Average (EWMA), four-level risk interval one-hot encoding, and a temporal attention mechanism into a two-layer stacked LSTM architecture, and employs Focal Loss to handle sample class imbalance. The primary methodological contribution lies in the systematic integration of sports-science domain knowledge—specifically the dual-path ACWR computation paradigm and evidence-based risk-zone encoding—into the input representation of a temporal deep learning model, a design that has not been previously validated in multi-sport injury prediction settings. Based on a multi-center dataset integrating 22 cohort studies (921 athletes, 657 injuries, covering four sports: soccer, tennis, rugby, and field hockey), the model was systematically evaluated via 5-fold temporal cross-validation and compared against six baseline methods: logistic regression, random forest, XGBoost, SVM, standard LSTM, and CNN-LSTM. ACWR-LSTM achieved AUC = 0.847 (95% CI: 0.831–0.863), sensitivity 78.1%, specificity 80.8%, F1 score 0.783, and Brier score 0.087 on the overall test set, significantly outperforming all baseline methods (<InlineEquation ID="IEq1"><EquationSource Format="TEX">$p < 0.001$</EquationSource><EquationSource Format="MATHML"><math><mrow><mi>p</mi><mo><</mo><mn>0.001</mn></mrow></math></EquationSource></InlineEquation>); AUC ranged from 0.794 to 0.862 across sports, and the inter-fold standard deviation of 5-fold cross-validation was below 0.022. Ablation studies confirmed that removing ACWR features caused an AUC drop of 0.084, making it the most critical source of model performance gain; temporal attention weight heatmaps revealed that the model autonomously focused on the high injury-risk window 7–10 days after workload spikes, highly consistent with sports medicine clinical consensus. Extensive new robustness analyses—including a <InlineEquation ID="IEq2"><EquationSource Format="TEX">$2\times 2$</EquationSource><EquationSource Format="MATHML"><math><mrow><mn>2</mn><mo>×</mo><mn>2</mn></mrow></math></EquationSource></InlineEquation> double-ablation matrix revealing a significant synergistic interaction between ACWR features and temporal attention (<InlineEquation ID="IEq3"><EquationSource Format="TEX">$+0.021$</EquationSource><EquationSource Format="MATHML"><math><mrow><mo>+</mo><mn>0.021</mn></mrow></math></EquationSource></InlineEquation> AUC, <InlineEquation ID="IEq4"><EquationSource Format="TEX">$p < 0.05$</EquationSource><EquationSource Format="MATHML"><math><mrow><mi>p</mi><mo><</mo><mn>0.05</mn></mrow></math></EquationSource></InlineEquation>), feature subset sensitivity, 10-seed data split sensitivity (bootstrap 95% CI [0.825, 0.868]), three-dimensional heterogeneity stratification (max <InlineEquation ID="IEq5"><EquationSource Format="TEX">$\Delta $</EquationSource><EquationSource Format="MATHML"><math><mi mathvariant="normal">Δ</mi></math></EquationSource></InlineEquation>AUC <InlineEquation ID="IEq6"><EquationSource Format="TEX">$= 0.018$</EquationSource><EquationSource Format="MATHML"><math><mrow><mo>=</mo><mn>0.018</mn></mrow></math></EquationSource></InlineEquation>), and a failure mode analysis—confirm that model performance is stable across operational variability and not contingent on specific design choices. By explicitly embedding sports science domain prior knowledge into a deep temporal modeling framework, ACWR-LSTM achieves substantial improvements in injury prediction accuracy, cross-sport generalization, and clinical interpretability, while important limitations remain regarding external validation and data heterogeneity across the 22 integrated cohort studies. The framework provides a replicable methodological paradigm for domain-knowledge-guided temporal deep learning in multi-sport injury prevention scenarios.</p>

Beyond threshold-based monitoring: A deep temporal learning framework with dual-path ACWR feature encoding for sports injury risk prediction

期刊论文
出版模式：

开放获取
发表日期：

2026年5月26日

查看全文

Yan Pan,
Junpeng Pang,
Binbin Jia,
Yongqi Lin,
Zaiying Chang,
Xuan Liu

摘要

Sports injury is a core issue constraining professional athletes’ competitive performance and health, and the scientific management of training load is a key intervention for injury prevention. However, existing machine learning injury prediction models generally neglect the temporal dynamic characteristics and domain-specific semantics of the Acute:Chronic Workload Ratio (ACWR), resulting in limited prediction accuracy and clinical interpretability. This study proposes the ACWR-LSTM framework, which explicitly integrates dual-pathway ACWR feature engineering based on Rolling Average (RA) and Exponentially Weighted Moving Average (EWMA), four-level risk interval one-hot encoding, and a temporal attention mechanism into a two-layer stacked LSTM architecture, and employs Focal Loss to handle sample class imbalance. The primary methodological contribution lies in the systematic integration of sports-science domain knowledge—specifically the dual-path ACWR computation paradigm and evidence-based risk-zone encoding—into the input representation of a temporal deep learning model, a design that has not been previously validated in multi-sport injury prediction settings. Based on a multi-center dataset integrating 22 cohort studies (921 athletes, 657 injuries, covering four sports: soccer, tennis, rugby, and field hockey), the model was systematically evaluated via 5-fold temporal cross-validation and compared against six baseline methods: logistic regression, random forest, XGBoost, SVM, standard LSTM, and CNN-LSTM. ACWR-LSTM achieved AUC = 0.847 (95% CI: 0.831–0.863), sensitivity 78.1%, specificity 80.8%, F1 score 0.783, and Brier score 0.087 on the overall test set, significantly outperforming all baseline methods ($p < 0.001$ $p < 0.001$ ); AUC ranged from 0.794 to 0.862 across sports, and the inter-fold standard deviation of 5-fold cross-validation was below 0.022. Ablation studies confirmed that removing ACWR features caused an AUC drop of 0.084, making it the most critical source of model performance gain; temporal attention weight heatmaps revealed that the model autonomously focused on the high injury-risk window 7–10 days after workload spikes, highly consistent with sports medicine clinical consensus. Extensive new robustness analyses—including a $2\times 2$ $2 \times 2$ double-ablation matrix revealing a significant synergistic interaction between ACWR features and temporal attention ($+0.021$ $+ 0.021$ AUC, $p < 0.05$ $p < 0.05$ ), feature subset sensitivity, 10-seed data split sensitivity (bootstrap 95% CI [0.825, 0.868]), three-dimensional heterogeneity stratification (max $\Delta $ $Δ$ AUC $= 0.018$ $= 0.018$ ), and a failure mode analysis—confirm that model performance is stable across operational variability and not contingent on specific design choices. By explicitly embedding sports science domain prior knowledge into a deep temporal modeling framework, ACWR-LSTM achieves substantial improvements in injury prediction accuracy, cross-sport generalization, and clinical interpretability, while important limitations remain regarding external validation and data heterogeneity across the 22 integrated cohort studies. The framework provides a replicable methodological paradigm for domain-knowledge-guided temporal deep learning in multi-sport injury prevention scenarios.