Deep Learning Pose Estimation for Multi-Label Recognition of Combined Hyperkinetic Movement Disorders

TL;DR

This paper introduces a deep learning-based pose estimation framework that analyzes clinical videos to objectively identify and differentiate multiple hyperkinetic movement disorders, addressing the limitations of subjective clinical assessments.

Contribution

It presents a novel pose-based machine learning approach that extracts detailed kinematic features from routine videos to improve diagnosis of overlapping hyperkinetic movement disorders.

Findings

Effective extraction of meaningful kinematic features

Potential for improved diagnostic accuracy

Addresses inter-rater variability in clinical assessments

Abstract

Hyperkinetic movement disorders (HMDs) such as dystonia, tremor, chorea, myoclonus, and tics are disabling motor manifestations across childhood and adulthood. Their fluctuating, intermittent, and frequently co-occurring expressions hinder clinical recognition and longitudinal monitoring, which remain largely subjective and vulnerable to inter-rater variability. Objective and scalable methods to distinguish overlapping HMD phenotypes from routine clinical videos are still lacking. Here, we developed a pose-based machine-learning framework that converts standard outpatient videos into anatomically meaningful keypoint time series and computes kinematic descriptors spanning statistical, temporal, spectral, and higher-order irregularity-complexity features.