Classification of chronic ankle instability using machine learning technique based on ankle kinematics during heel rise in delivery workers

Abstract

Objective: Ankle injuries in delivery workers (DWs) are often caused by trips, and high recurrence rates of ankle sprains are related to chronic ankle instability (CAI). Heel rise requires joint angles and moments similar to those of the terminal stance phase of walking that the foot supinates. Thus, our study aimed to develop, determine, and compare the predictive performance of statistical machine learning models to classify DWs with and without CAI using ankle kinematics during heel rise.

Methods: In total, 203 DWs were screened for eligibility. Seven predictors were included in our study (age, work duration, body mass index, calcaneal stance position angle [CSPA] in the initial and terminal positions during heel rise, calcaneal movement during heel rise [CM_HR], and plantar flexion angle during heel rise). Six machine learning algorithms, including logistic regression, decision tree, AdaBoost, Extreme Gradient boosting machines, random forest, and support vector machine, were trained.

Results: The random forest model (area under the curve [AUC], 0.967 [excellent]; F1, 0.889; accuracy, 0.925) confirmed the best predictive performance in the test datasets among the six machine learning models. For Shapley Additive Explanations, old age, low CMHR, high CSPA in the initial position, high PFA, long work duration, low CSPA in the terminal position, and high body mass index were the most important predictors of CAI in the random forest model.

Conclusion: Ankle kinematics during heel rise can be considered in the classification of DWs with and without CAI.

Keywords: Exercise; machine learning; musculoskeletal; rehabilitation; risk factors.

Figure 1.

Ankle kinematic measurements using two-dimensional video analysis during heel rise. (A) Measurement of the calcaneal stance position angle in terminal position rise and (B) Measurement of plantar flexion angle.

Figure 2.

Scatter plot and multiaxis linear projection for classification of DWs with and without CAI. (A) scatter plot and (B) multiaxis linear projection, blue dot = DW with CAI, red dot = DW without CAI, dot size is based on CM_HR (the smaller the dot size, the greater the calcaneal inversion movement).

Figure 3.

Receiver operating characteristic (ROC) curves of six machine learning algorithms. (A) In the training set and (B) in the test set.

Figure 4.

(A) Feature permutation importance of random forest model in the training set; (B) Shapley Additive Explanation analyses of random forest model in the training set; (C) feature permutation importance of support vector machine model in the training set; (D): Shapley Additive Explanation analyses of support vector machine model in the training set.