Evaluating the Reliability of Kinotek: An AI-Driven 3D Motion Capture Tool for Weight-Bearing Ankle Dorsiflexion Assessment.

Background: The range of motion (ROM) in ankle dorsiflexion during weight-bearing activities has important functional implications, including an elevated risk of injury when ROM is restricted. The integration of advanced digital technologies, particularly those utilizing artificial intelligence (AI), is becoming increasingly prevalent across professional domains. These tools have demonstrated effectiveness in enhancing diagnostic accuracy and improving patient outcomes in physical rehabilitation. This study aimed to assess the reliability of a portable 3D motion capture platform incorporating AI (Kinotek) compared to a standard plastic goniometer in evaluating weight-bearing ankle dorsiflexion ROM.Methods: Twenty-four healthy participants (mean age: 29 ± 12 years; height: 172.7 ± 10.2 cm; weight: 70.3 ± 15 kg) were recruited. Each participant completed two test-retest trials of weight-bearing ankle dorsiflexion during a forward lunge. Intertrial reliability was evaluated using intraclass correlation coefficients (ICC(2, k)) with 95% confidence intervals (CI), comparing measurements obtained from the Kinotek system and the goniometer.Results: The mean ± standard deviation (standard error of the mean) ROM values were 18.8 ± 6.67 (0.99) degrees for Kinotek and 14.61 ± 5.72 (0.86) degrees for the goniometer. The ICC (95% CI) values were 0.90 (0.82–0.94) for Kinotek and 0.79 (0.65–0.89) for the goniometer. The Pearson correlation coefficient (r) was 0.55.Conclusion: The findings indicate that the Kinotek system demonstrates good-to-excellent reliability, whereas the goniometer exhibits moderate-to-good reliability in assessing weight-bearing ankle dorsiflexion ROM. The strong association observed supports the potential utility of AI-driven motion capture systems as reliable tools in both clinical and research contexts for evaluating weight-bearing ankle dorsiflexion.