Simultaneous hyperkinetic movement disorders phenotyping: a cross-cohort pediatric transfer study using routine videos, markerless pose estimation and a tabular foundation model.
14 affiliations
- 1Service of Neurology, Department of Clinical Neurosciences, Lausanne University Hospital (CHUV) and University of Lausanne (UNIL), Lausanne, Switzerland.
- 2Institut du Neurone, Montferrier sur Lez, France.
- 3Department of Neurology, Clinique Beau Soleil, Institut Mutualiste Montpelliérain, Montpellier, France.
- 4Department of Neurosurgery, Military University Hospital of Sfax, Sfax, Tunisia.
- 5Edinburgh Medical School, University of Edinburgh, Edinburgh, United Kingdom.
- 6Movement Disorders Unit, Pediatric Neurology Department, Institut de Recerca, Hospital Sant Joan de Déu, Barcelona, Spain.
- 7European Reference Network for Rare Neurological Diseases (ERN-RND), Barcelona, Spain.
- 8U-703 Centre for Biomedical Research on Rare Diseases (CIBER-ER), Instituto de Salud Carlos III, Barcelona, Spain.
- 9Department of Neurology, CHU Montpellier, Montpellier, France.
- 10Department of Clinical Neuroscience, Umeå University, Umeå, Sweden.
- 11Defitech Center for Interventional Neurotherapies (NeuroRestore), University Hospital Lausanne and Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland.
- 12Neuro-X Institute, Ecole Polytechnique Fédérale de Lausanne, Geneva, Switzerland.
- 13Department of Neurosurgery, Lausanne University Hospital (CHUV) and University of Lausanne (UNIL), Lausanne, Switzerland.
- 14Department of Pediatrics, British Columbia Children's Hospital, Vancouver, BC, Canada.
Abstract
Background
Accurate recognition of movement disorders (MDs) phenomenology remains a demanding task in pediatric practice, compounded by mixed and evolving motor presentations, inter-examiner expertise variability, and the scarcity of structured digital tools validated across age groups. A child with combined MDs may receive different phenomenological diagnoses from different examiners, with direct consequences for etiological guidance, treatment planning, neuromodulation candidacy, and longitudinal monitoring. Video-based examination is already central to expert practice, but its systematic analysis remains frequently inaccessible because of limited human resources.Objective
To develop and externally test a video-based framework for simultaneous detection of hyperkinetic MDs phenomenologies: dystonia, tremor, myoclonus, chorea, athetosis, ballismus, stereotypies, and tics using routine clinical recordings, with external, cross-cohort transfer from adult to pediatric populations.Methods
In this proof-of-concept study, the framework combines markerless pose estimation, kinematic descriptors, and a pretrained foundation model. A shared predictive backbone was developed on 21 adults with confirmed hyperkinetic MDs and 4 healthy controls. External validation was performed on an independent cohort: a real-world pediatric sample (n = 12, monogenic combined MDs). For the external dataset, the backbone was deployed without retraining; lightweight calibration adjusted only the final subject-level decision step using a clinician-labeled small representative subset of patients.Results
After local calibration of the decision layer on the clinician-selected subset, performance improved on the held-out pediatric patients (n = 7): Hamming accuracy rose from 0.804 to 0.839 and the Jaccard index from 0.548 to 0.633 (true positives 8 to 13, false negatives 9 to 4, false positives 2 to 5, across 56 patient-phenomenology decisions). Given the small held-out sample, these estimates should be regarded as exploratory. This calibrated performance was preserved, and the Jaccard index further improved, when the evaluation was restricted to the phenomenologies with more definite clinician agreement (Hamming accuracy 0.9, Jaccard index 0.786), indicating that the gains did not rest on the least-reliable labels.Conclusion
A two-stage strategy, a reusable predictive backbone combined with lightweight, site-specific adaptation, may offer a practical and scalable route to routine-video-based digital phenotyping of MDs across age groups. A shared backbone trained on standardized adult recordings transferred phenotyping signal to an independent, real-world pediatric cohort without retraining. Local calibration of the decision layer improved performance in this small exploratory cohort, while clinical-grade reliability across the full phenotypic spectrum remains to be confirmed in larger cohorts.
Versions
Related
- Preoperative 3D morphological phenotyping of hepatocellular carcinoma via a foundation model: a non-invasive biomarker for recurrence risk stratification.
- A mechanistic classification of GNAO1 mutations links Gβ γ sequestration to hyperkinetic disease and treatment response
- A Small Dose With Large Movements: Suspected Gabapentin-Associated Mixed Hyperkinetic Movement Disorder in End-Stage Renal Disease: A Case Report.
- Markerless Motion Capture Reveals Movement Abnormalities in Isolated REM Sleep Behavior Disorder
- Is Markerless Motion capture accurate enough to identify kinematic risk factors associated with the ACL rupture during high-velocity change of direction? A pilot study on rugby players.
BibTeX
@article{cif2026simultaneous,
title = {{Simultaneous hyperkinetic movement disorders phenotyping: a cross-cohort pediatric transfer study using routine videos, markerless pose estimation and a tabular foundation model.}},
author = {Laura Cif and Diane Demailly and Zohra Souei and Muhammad Mushhood Ur Rehman and Juan Dario Ortigoza-Escobar and Mayté Castro-Jiménez and Cécile A Hubsch and Sophie Huby and Morgan Dornadic and Gun-Marie Hariz and Eduardo M Moraud and Jocelyne Bloch and Gabriella A Horvàth and Xavier Vasques},
journal = {Frontiers in neurology},
year = {2026},
month = oct,
doi = {10.3389/fneur.2026.1904458},
url = {https://europepmc.org/article/MED/42908833},
note = {CC BY},
}
RIS
TY - JOUR
TI - Simultaneous hyperkinetic movement disorders phenotyping: a cross-cohort pediatric transfer study using routine videos, markerless pose estimation and a tabular foundation model.
AU - Laura Cif
AU - Diane Demailly
AU - Zohra Souei
AU - Muhammad Mushhood Ur Rehman
AU - Juan Dario Ortigoza-Escobar
AU - Mayté Castro-Jiménez
AU - Cécile A Hubsch
AU - Sophie Huby
AU - Morgan Dornadic
AU - Gun-Marie Hariz
AU - Eduardo M Moraud
AU - Jocelyne Bloch
AU - Gabriella A Horvàth
AU - Xavier Vasques
PY - 2026
DA - 2026/10/07
JO - Frontiers in neurology
DO - 10.3389/fneur.2026.1904458
UR - https://europepmc.org/article/MED/42908833
AB - Background Accurate recognition of movement disorders (MDs) phenomenology remains a demanding task in pediatric practice, compounded by mixed and evolving motor presentations, inter-examiner expertise variability, and the scarcity of structured digital tools validated across age groups. A child with combined MDs may receive different phenomenological diagnoses from different examiners, with direct consequences for etiological guidance, treatment planning, neuromodulation candidacy, and longitudinal monitoring. Video-based examination is already central to expert practice, but its systematic analysis remains frequently inaccessible because of limited human resources.Objective To develop and externally test a video-based framework for simultaneous detection of hyperkinetic MDs phenomenologies: dystonia, tremor, myoclonus, chorea, athetosis, ballismus, stereotypies, and tics using routine clinical recordings, with external, cross-cohort transfer from adult to pediatric populations.Methods In this proof-of-concept study, the framework combines markerless pose estimation, kinematic descriptors, and a pretrained foundation model. A shared predictive backbone was developed on 21 adults with confirmed hyperkinetic MDs and 4 healthy controls. External validation was performed on an independent cohort: a real-world pediatric sample (n = 12, monogenic combined MDs). For the external dataset, the backbone was deployed without retraining; lightweight calibration adjusted only the final subject-level decision step using a clinician-labeled small representative subset of patients.Results After local calibration of the decision layer on the clinician-selected subset, performance improved on the held-out pediatric patients (n = 7): Hamming accuracy rose from 0.804 to 0.839 and the Jaccard index from 0.548 to 0.633 (true positives 8 to 13, false negatives 9 to 4, false positives 2 to 5, across 56 patient-phenomenology decisions). Given the small held-out sample, these estimates should be regarded as exploratory. This calibrated performance was preserved, and the Jaccard index further improved, when the evaluation was restricted to the phenomenologies with more definite clinician agreement (Hamming accuracy 0.9, Jaccard index 0.786), indicating that the gains did not rest on the least-reliable labels.Conclusion A two-stage strategy, a reusable predictive backbone combined with lightweight, site-specific adaptation, may offer a practical and scalable route to routine-video-based digital phenotyping of MDs across age groups. A shared backbone trained on standardized adult recordings transferred phenotyping signal to an independent, real-world pediatric cohort without retraining. Local calibration of the decision layer improved performance in this small exploratory cohort, while clinical-grade reliability across the full phenotypic spectrum remains to be confirmed in larger cohorts.
SN - 1664-2295
ER -