Corrigendum: A novel approach for sports injury risk prediction: based on time-series image encoding and deep learning.

[This corrects the article DOI: 10.3389/fphys.2023.1174525.].

A novel approach for sports injury risk prediction: based on time-series image encoding and deep learning.

The rapid development of big data technology and artificial intelligence has provided a new perspective on sports injury prevention. Although data-driven algorithms have achieved some valuable results in the field of sports injury risk assessment, the lack of sufficient generalization of models and the inability to automate feature extraction have made it challenging to deploy research results in the real world. Therefore, this study attempts to build an injury risk prediction model using a combination of time-series image encoding and deep learning algorithms to address this issue better. This study used the time-series image encoding approach for feature construction to represent relationships between values at different moments, including Gramian Angular Summation Field (GASF), Gramian Angular Difference Field (GADF), Markov Transition Field (MTF), and Recurrence Plot (RP). Deep Convolutional Auto-Encoder (DCAE) learned the image-encoded data for representation to obtain features with good discrimination, and the classifier was performed using Deep Neural Network (DNN). The results from five repeated experiments show that the GASF-DCAE-DNN model is overall better in the training (AUC: 0.985 ± 0.001, Gmean: 0.930 ± 0.007, Sensitivity: 0.997 ± 0.003, Specificity: 0.868 ± 0.013) and test sets (AUC: 0.891 ± 0.026, Gmean: 0.830 ± 0.027, Sensitivity: 0.816 ± 0.039, Specificity: 0.845 ± 0.022), with good discriminative power, robustness, and generalization ability. Compared with the best model reported in the literature, the AUC, Gmean, Sensitivity, and Specificity of the GASF-DCAE-DNN model were higher by 23.9%, 27.5%, 39.7%, and 16.2%, respectively, which confirmed the validity and practicability of the model in injury risk prediction. In addition, differences in injury risk patterns between the training and test sets were identified through shapley additivity interpretation. It was also found that the training volume was an essential factor that affected injury risk prediction. The model proposed in this study provides a powerful injury risk prediction tool for future sports injury prevention practice.

A novel lower extremity non-contact injury risk prediction model based on multimodal fusion and interpretable machine learning.

The application of machine learning algorithms in studying injury assessment methods based on data analysis has recently provided a new research insight for sports injury prevention. However, the data used in these studies are primarily multi-source and multimodal (i.e., longitudinal repeated-measures data and cross-sectional data), resulting in the models not fully utilising the information in the data to reveal specific injury risk patterns. Therefore, this study proposed an injury risk prediction model based on a multi-modal strategy and machine learning algorithms to handle multi-source data better and predict injury risk. This study retrospectively analysed the routine monitoring data of sixteen young female basketball players. These data included training load, perceived well-being status, physiological response, physical performance and lower extremity non-contact injury registration. This study partitions the original dataset based on the frequency of data collection. Extreme gradient boosting (XGBoost) was used to construct unimodal submodels to obtain decision scores for each category of indicators. Ultimately, the decision scores from each submodel were fused using the random forest (RF) to generate a lower extremity non-contact injury risk prediction model at the decision-level. The 10-fold cross-validation results showed that the fusion model was effective in classifying non-injured (mean Precision: 0.9932, mean Recall: 0.9976, mean F2-score: 0.9967), minimal lower extremity non-contact injuries risk (mean Precision: 0.9317, mean Recall: 0.9167, mean F2-score: 0.9171), and mild lower extremity non-contact injuries risk (mean Precision: 0.9000, mean Recall: 0.9000, mean F2-score: 0.9000). The model performed significantly more optimal than the submodel. Comparing the fusion model proposed with a traditional data integration scheme, the average Precision and Recall improved by 8.2 and 20.3%, respectively. The decision curves analysis showed that the proposed fusion model provided a higher net benefit to athletes with potential lower extremity non-contact injury risk. The validity, feasibility and practicality of the proposed model have been confirmed. In addition, the shapley additive explanation (SHAP) and network visualisation revealed differences in lower extremity non-contact injury risk patterns across severity levels. The model proposed in this study provided a fresh perspective on injury prevention in future research.

Prediction of thoracic injury severity in frontal impacts by selected anatomical morphomic variables through model-averaged logistic regression approach.

This study resulted in a model-averaging methodology that predicts crash injury risk using vehicle, demographic, and morphomic variables and assesses the importance of individual predictors. The effectiveness of this methodology was illustrated through analysis of occupant chest injuries in frontal vehicle crashes. The crash data were obtained from the International Center for Automotive Medicine (ICAM) database for calendar year 1996 to 2012. The morphomic data are quantitative measurements of variations in human body 3-dimensional anatomy. Morphomics are obtained from imaging records. In this study, morphomics were obtained from chest, abdomen, and spine CT using novel patented algorithms. A NASS-trained crash investigator with over thirty years of experience collected the in-depth crash data. There were 226 cases available with occupants involved in frontal crashes and morphomic measurements. Only cases with complete recorded data were retained for statistical analysis. Logistic regression models were fitted using all possible configurations of vehicle, demographic, and morphomic variables. Different models were ranked by the Akaike Information Criteria (AIC). An averaged logistic regression model approach was used due to the limited sample size relative to the number of variables. This approach is helpful when addressing variable selection, building prediction models, and assessing the importance of individual variables. The final predictive results were developed using this approach, based on the top 100 models in the AIC ranking. Model-averaging minimized model uncertainty, decreased the overall prediction variance, and provided an approach to evaluating the importance of individual variables. There were 17 variables investigated: four vehicle, four demographic, and nine morphomic. More than 130,000 logistic models were investigated in total. The models were characterized into four scenarios to assess individual variable contribution to injury risk. Scenario 1 used vehicle variables; Scenario 2, vehicle and demographic variables; Scenario 3, vehicle and morphomic variables; and Scenario 4 used all variables. AIC was used to rank the models and to address over-fitting. In each scenario, the results based on the top three models and the averages of the top 100 models were presented. The AIC and the area under the receiver operating characteristic curve (AUC) were reported in each model. The models were re-fitted after removing each variable one at a time. The increases of AIC and the decreases of AUC were then assessed to measure the contribution and importance of the individual variables in each model. The importance of the individual variables was also determined by their weighted frequencies of appearance in the top 100 selected models. Overall, the AUC was 0.58 in Scenario 1, 0.78 in Scenario 2, 0.76 in Scenario 3 and 0.82 in Scenario 4. The results showed that morphomic variables are as accurate at predicting injury risk as demographic variables. The results of this study emphasize the importance of including morphomic variables when assessing injury risk. The results also highlight the need for morphomic data in the development of human mathematical models when assessing restraint performance in frontal crashes, since morphomic variables are more "tangible" measurements compared to demographic variables such as age and gender.