Category: Tics/Stereotypies
Objective: To develop a machine learning tool to distinguish between facial/head tics in people with Tourette syndrome and spontaneous movements in healthy controls.
Background: Distinguishing tics in people with tic disorders from extra movements in healthy controls can be difficult [1]. Also, rating tics based on video recordings is time consuming and effortful. Machine learning has the potential to assist with these challenges by distinguishing between tics and other extra movements and by supporting clinical ratings.
Method: The study used a dataset of 63 videos of people with tic disorders to train a Random Forest classifier for second-wise tic detection. The classifier utilized facial landmarks as input and defined tic seconds as those with tics of equal or greater severity than a pre-defined threshold. The trained classifier was then used to predict the presence of tics in patients and extra movements in healthy controls. These predictions were utilized to calculate several features, such as the number of tics per minute, the maximum duration of a continuous non-tic segment, the maximum duration of a continuous tic, the average duration of tic-free segments, the number of changes from tic to non-tic segments and vice versa per minute, the average size of a tic-cluster, and the number of clusters per minute. These features were combined into a single tic detection score using logistic regression. The model parameters were obtained by training on a dataset of 124 videos of individuals with tic disorders and 162 videos of healthy controls. To assess the accuracy of this score in classifying patients and healthy controls, a test dataset of 50 videos of patients and 50 videos of healthy controls was used.
Results: The test set achieved a classification accuracy of 83%. Paired samples T-tests revealed significant differences between the two groups in all features derived from the tic predictions.
Conclusion: The machine learning algorithm is useful to detect tics and distinguish between tics and other extra movements. It could be developed into a clinically applicable tool. To improve classification accuracy, our next step is to fine-tune the tic detection score. Also, we aim to analyze the significance of each feature to determine, which characteristics are most helpful in differentiating between the two groups. The algorithm might also be helpful to distinguish between tics and functional movements.
References: 1. Bartha, S. et al. Extra Movements in Healthy People: Challenging the Definition and Diagnostic Practice of Tic Disorders. Ann. Neurol. 93, 472–478 (2023).
To cite this abstract in AMA style:
L. Becker, R. Schappert, N. Brügge, G. Sallandt, F. Li, J. Friedrich, T. Bäumer, C. Frings, C. Beste, R. Stenger, V. Roessner, H. Handels, S. Fudickar, A. Münchau. New machine learning approaches in tic detection: Seeking to learn more about the characteristic of tics [abstract]. Mov Disord. 2023; 38 (suppl 1). https://www.mdsabstracts.org/abstract/new-machine-learning-approaches-in-tic-detection-seeking-to-learn-more-about-the-characteristic-of-tics/. Accessed November 21, 2024.« Back to 2023 International Congress
MDS Abstracts - https://www.mdsabstracts.org/abstract/new-machine-learning-approaches-in-tic-detection-seeking-to-learn-more-about-the-characteristic-of-tics/