Image-based Quantification of Postural Deviations on Patients with Cervical Dystonia: A Machine Learning Approach Using Synthetic Training Data

arXiv cs.CV / 3/30/2026

📰 NewsSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper proposes an automated, image-based system to objectively quantify postural deviations in cervical dystonia, reducing reliance on subjective clinical rating scales like TWSTRS.
It combines a pretrained head-pose estimation approach for rotational symptoms with a deep learning model trained on ~16,000 synthetic avatar images to capture rare translational symptoms such as lateral shift.
Validation in a multicenter study uses 100 real patient images and 100 labeled synthetic avatars, comparing predicted scores against consensus ratings from 20 expert clinicians.
The system shows strong agreement for rotational components (e.g., torticollis r=0.91, laterocollis r=0.81, anteroretrocollis r=0.78) and moderate agreement for lateral shift (r=0.55), with better avatar benchmark accuracy than human raters.
Overall, the authors argue that synthetic training data can bridge limited clinical labeling and generalize to real patients, enabling more standardized monitoring and trial evaluation.

Abstract

Cervical dystonia (CD) is the most common form of dystonia, yet current assessment relies on subjective clinical rating scales, such as the Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS), which requires expertise, is subjective and faces low inter-rater reliability some items of the score. To address the lack of established objective tools for monitoring disease severity and treatment response, this study validates an automated image-based head pose and shift estimation system for patients with CD. We developed an assessment tool that combines a pretrained head-pose estimation algorithm for rotational symptoms with a deep learning model trained exclusively on ~16,000 synthetic avatar images to evaluate rare translational symptoms, specifically lateral shift. This synthetic data approach overcomes the scarcity of clinical training examples. The system's performance was validated in a multicenter study by comparing its predicted scores against the consensus ratings of 20 clinical experts using a dataset of 100 real patient images and 100 labeled synthetic avatars. The automated system demonstrated strong agreement with expert clinical ratings for rotational symptoms, achieving high correlations for torticollis (r=0.91), laterocollis (r=0.81), and anteroretrocollis (r=0.78). For lateral shift, the tool achieved a moderate correlation (r=0.55) with clinical ratings and demonstrated higher accuracy than human raters in controlled benchmark tests on avatars. By leveraging synthetic training data to bridge the clinical data gap, this model successfully generalizes to real-world patients, providing a validated, objective tool for CD postural assessment that can enable standardized clinical decision-making and trial evaluation.