To develop an interpretable multimodal framework for detecting tic events in children with tic disorders by translating model decisions into structured, time-aligned evidence from synchronized video and physiological signals.
Approach:
Key Findings:
TIC-XNet achieved a window-level AUC of 0.915 ± 0.019 on the pooled shared test set.
It demonstrated higher event-level recall and precision, fewer missed events, and lower post-buffering prediction latency compared to comparator models.
The outputs showed higher decision fidelity, greater stability under perturbation, and closer temporal alignment with expert-annotated tic onsets.
Subject-level translated numerical signals were associated with tic severity.
Interpretation:
Evidence translation can support more interpretable multimodal detection of tic events in children with tic disorders while maintaining strong predictive performance.
Conclusion:
The study indicates that TIC-XNet provides a robust framework for detecting tic events with enhanced interpretability.
