TIC-XNet: a structured evidence translation framework for interpretable multimodal pediatric tic event detection with improved temporal alignment and fidelity - Scorecard - MDSpire
Advertisement
TIC-XNet: a structured evidence translation framework for interpretable multimodal pediatric tic event detection with improved temporal alignment and fidelity
Clinical Scorecard: TIC-XNet: An Interpretable Framework for Enhanced Detection of Pediatric Tic Events Through Structured Evidence and Improved Temporal Alignment
At a Glance
Category
Detail
Condition
Pediatric Tic Disorders
Key Mechanisms
Multimodal analysis of synchronized video, heart rate, and electrodermal activity signals.
Target Population
Children with clinically diagnosed tic disorders.
Care Setting
Clinical assessments and home-based observations.
Key Highlights
TIC-XNet achieved a window-level AUC of 0.915 ± 0.019.
Higher event-level recall and precision compared to comparator models.
Fewer missed events and lower post-buffering prediction latency.
Translated outputs showed higher decision fidelity and stability.
Subject-level signals were associated with tic severity.
Guideline-Based Recommendations
Diagnosis
Current clinical evaluation relies on physician observation and rating instruments like the Yale Global Tic Severity Scale (YGTSS).
Management
Objective assessment through automated tic detection methods is recommended.
Monitoring & Follow-up
Continuous monitoring approaches are of considerable clinical value.
Risks
Short in-clinic assessments may fail to reflect the temporal dynamics and contextual variability of symptoms.
Patient & Prescribing Data
Children with tic disorders.
Automated tic assessment methods are being explored to improve diagnosis and monitoring.
Clinical Best Practices
Integrate multimodal behavioral and physiological signals for robust tic detection.
Utilize explainable artificial intelligence to enhance clinical trust and decision-making.
