The Human Tic Detector: An automatic approach to tic characterization using wearable sensors. (February 2022)

• Record Type: Journal Article
• Authors: Cernera, Stephanie; Pramanik, Leena; Boogaart, Zachary; Cagle, Jackson N.; Gomez, Julieth; Moore, Katie; Au, Ka Loong Kelvin; Okun, Michael S.; Gunduz, Aysegul; Deeb, Wissam
• Is Part Of: Clinical neurophysiology. Volume 134(2022)
• Page Start: 102
• Page End: 110
• Publication Date: 2022-02
• Subjects: Tourette syndrome -- Tics -- Wearable sensors -- Objective clinical measures -- Electromyography; ACC accelerometer data -- ∑-ACC sum of ACC power spectrum -- ADHD attention-deficit hyperactivity disorder -- DSM-5 diagnostic and statistical manual edition 5 -- EMG Electromyography -- ∑-EMG sum of the EMG power spectrum -- M mRVTRS protocol -- mRVTRS modified Rush Video Tic Rating Scale -- OCD obsessive-compulsive disorder -- R rest period -- SHAP SHapley Additive exPlanation -- SVM Support Vector Machine -- T tic-freely period -- TS Tourette Syndrome -- TSW acronym for study subject -- V voluntary movement period -- YGTSS Yale Global Tic Severity Scale; Neurophysiology -- Periodicals; Electroencephalography -- Periodicals; Electromyography -- Periodicals; Neurology -- Periodicals; 612.8
• DOI: 10.1016/j.clinph.2021.10.017
• Languages: English
• ISSNs: 1388-2457

Abstract

Highlights: Tourette syndrome scales are limited by recall bias or brief observation; thus, we aimed at developing a sensor-based algorithm. The algorithm was 96.69% successful at detecting and classifying tics versus voluntary movements on a test dataset. Using validation data, movement classification accuracy was 94.23% and tic detection accuracy compared to experts was 85.63%. Abstract: Objective: Current rating scales for Tourette syndrome (TS) are limited by recollection bias or brief assessment periods. This proof-of-concept study aimed to develop a sensor-based paradigm to detect and classify tics. Methods: We recorded both electromyogram and acceleration data from seventeen TS patients, either when voluntarily moving or experiencing tics and during the modified Rush Video Tic Rating Scale (mRVTRS). Spectral properties of voluntary and tic movements from the sensor that captured the dominant tic were calculated and used as features in a support vector machine (SVM) to detect and classify movements retrospectively. Results: Across patients, the SVM had an accuracy, sensitivity, and specificity of 96.69 ± 4.84%, 98.24 ± 4.79%, and 96.03 ± 6.04%, respectively, when classifying movements in the test dataset. Furthermore, each patient's SVM was validated using data collected during the mRVTRS. Compared to the expert consensus, the tic detection accuracy was 85.63 ± 15.28% during the mRVTRS, while overall movement classification accuracy was 94.23 ± 5.97%. Conclusions: These results demonstrate that wearable sensors can capture physiological differences between tic and voluntary movements and are comparable to expert consensus. Significance: Ultimately, wearables could individualize and improve care for people with TS, provide a robust and objective measure of tics, and allow data collection in real-world settings.