Objective To analyze the risk of return to sport in athletes using their gait data following anterior cruciate ligament re-construction(ACLR). Methods From May to June,2023,39 athletes after ACLR were recruited in Wuhan Sports University.Their data on sta-ble gait and tandem gait were recorded using a three-dimensional motion capture system,surface electromyogra-phy and a three-dimensional ergometer table.Additionally,return-to-sport scores were calculated using the K-STARTS test.The relationship between each gait indicator and the total score of the K-STARTS test was ana-lyzed with Pearson correlation analysis.Furthermore,the key indicators related to the risk of return to sport were analyzed using linear regression. Results In the stable gait test,the step time was negatively correlated with the total score of K-STARTS(r=-0.479,P=0.002),and the peak amplitude symmetry index of rectus femoris(r=0.448,P=0.004)and vastus lateralis(r=0.595,P=0.001)were positively correlated with the total score of K-STARTS.In the tandem gait test,the lateral displacement distance of the center of gravity was negatively correlated with the total score of K-STARTS(r=-0.341,P=0.034),and the time symmetry index of peak amplitude of vastus lateralis was positively correlated with the total score of K-STARTS(r=0.320,P=0.047).Regression analysis showed that the interpretation of the model based on stable gait(F=15.818,P=0.001,R2=0.650)was better than that based on tandem gait(F=7.692,P=0.001,R2=0.397). Conclusion In stable gait,gait rhythm variability and symmetry are correlated with return to sport risk.In tandem gait,gait balance and symmetry indexes are correlated with return-to-sport risk.Compared with tandem gait,the inter-pretation of return-to-sport risk assessment model based on stable gait information is better,and may be more suitable as a simple return-to-sport risk test method.

BACKGROUND:The sports medicine community has widely called for the use of machine learning technology to efficiently process the huge and complicated sports data resources,and construct intelligent sports injury prediction models,enabling accurate early warning of sports injuries.It is of great significance to comprehensively summarize and review such research results so as to grasp the direction of early warning model improvement and to guide the construction of sports injury prediction models in China. OBJECTIVE:To systematically review and analyze relevant research on sports injury prediction models based on machine learning technology,thereby providing references for the development of sports injury prediction models in China. METHODS:Literature search was conducted on CNKI,Web of Science and EBSCO databases,which mainly searched for literature related to machine learning techniques and sports injuries.Finally,61 articles related to sports injury prediction models were included for analysis. RESULTS AND CONCLUSION:(1)In terms of external risk feature indicators,there is a lack of competition scenario indicators,and the inclusion of related feature indicators needs to be further improved to further enrich the dimensions of the dataset for model training.In addition,the inclusion feature weighting methods of the sports injury prediction model are mainly based on filtering methods and the use of embedding and wrapping weighting methods needs to be strengthened in order to enhance the analysis of the interaction effects of multiple risk factors.(2)In terms of model body training,supervised learning algorithms become the mainstream choice.Such algorithms have higher requirements for the completeness of sample labeling information,and the application scenarios are easily limited.Therefore,the application of unsupervised and semi-supervised algorithms can be increased in the later stage.(3)In terms of model performance evaluation and optimization,the current studies mainly adopt two verification methods:HoldOut crossover and k-crossover.The range of AUC values is(0.76±0.12),the range of sensitivity is(75.92±11.03)%,the range of specificity is(0.03±4.54)%,the range of F1 score is(80.60±10.63)%,the range of accuracy is(69.96±13.10)%,and the range of precision is(70±14.71)%.Data augmentation and feature optimization are the most common model optimization operations.The accuracy and precision of the current sports injury prediction model are about 70%,and the early warning effect is good.However,the model optimization operation is relatively single,and data augmentation methods are often used to improve model performance.Further adjustments to the model algorithm and hyperparameters are needed to further improve model performance.(4)In terms of model feature extraction,most of the internal risk profile indicators included are mainly based on anthropometrics,training load,years of training,and injury history,but there is a lack of sports recovery and physical function indicators.

ObjectiveTo investigate the difference in bilateral lower limb muscle synergy mode during gait in patients after unilateral anterior cruciate ligament reconstruction. MethodsElectromyography from bilateral lower limb muscles during gait were collected from twelve male and eight female patients after unilateral anterior cruciate ligament reconstruction in Affiliated Hospital of Wuhan Sports University, from April to June, 2023. The data were analyzed using non-negative matrix decomposition algorithm to extract the number of muscle synergies in the affected and unaffected legs, the time to peak activation of muscle synergies and the relative weights of the muscles. ResultsSix types of muscle synergy were identified in the unaffected leg of males during gait, while five types were identified in the affected leg, lacking synergy 2 that mainly from the tibialis anterior muscle. Six types of muscle synergy were identified in both legs in females during gait. There was no significant difference in the time to peak activation of muscle synergies between both legs in males (P > 0.05). However, the time to peak activation of muscle synergies increased in females in the affected leg for synergy 3 and synergy 5 (P < 0.05). The relative weight of the rectus femoris was lower in synergy 1 in the affected leg in males (P < 0.05). For female, the relative weight of the vastus lateralis was higher and the relative weight of the biceps femoris was lower in synergy 2 in the affected leg in females (P < 0.05); while the relative weight of the rectus femoris was lower in synergy 3 (P < 0.05), and the relative weight of the biceps femoris was lower in synergy 6 (P < 0.05). ConclusionMales would freeze the muscle synergy dominating ankle dorsiflexion in affected leg to enhance ankle stability, and reduce the relative weight of rectus femoris during the loading response phase to weaken the knee landing cushioning. However, females would delay the activation of synergies dominating in loading response phase and the mid-stance phase, enhance the relative weight of vastus lateralis during the loading response phase, and reduce the relative weights of rectus femoris in the loading response phase and the relative weight of biceps femoris in the mid-stance phase, to limit knee flexion.