the application of artificial neural network and wearable inertial sensor in kicking skill assessment.

The trade-off between speed and accuracy in process-oriented tests of fundamental motor skills development has always been a challenge in motor development screening plans. thus, this study was designed to evaluate the feasibility of using wearable inertial sensors (imus) based on artificial intelligence algorithms to assess kicking skill. thirteen children aged 4 to 10 years (age = 8 ±1.37) (boys = 58%) participated in this study. the subjects were asked to do at least ten repetitions of the kicking skill according to the tgmd-3. trials were captured with video recording and three wearable inertial sensors installed on the ankles and lower back. k-nearest neighbor artificial intelligence algorithms automatically classified the linear acceleration and angular velocity signals. the intraclass correlation coefficient (icc) was calculated between expert scores and the artificial intelligence algorithm. all tests were done at a 95% confidence interval. the classification accuracy of the knn algorithm (k=7) for kicking was 95%, icc =0.90 (ci=0.86-0.95). the scoring time was reduced from 5 minutes per trial (in an expert-oriented way) to less than 30 seconds (using artificial intelligence). as a result, this method was a reliable and practical way to assess the fundamental motor skills. also, by maintaining relative accuracy, it was possible to reduce test time for research, clinical, sports, and educational purposes.