Related Experiment Video
Updated: Aug 5, 2026
14:52
Computerized Dynamic Posturography for Postural Control Assessment in Patients with Intermittent Claudication
Published on: December 11, 2013
Using Machine Learning-Based Classification of Postural Stability in Cervicogenic Headache Patients: Predictors and
Mohamed Abdelaziz Emam1,2,3, Magda Ramadan4, Andras Attila Horvath3,5
1Basic Sciences Department, Faculty of Physical Therapy, Kafr El Sheikh University, Kafr El Sheikh 33511, Egypt.
Life (Basel, Switzerland)
|July 28, 2026
Summary
Machine learning accurately classified postural stability in cervicogenic headache (CEH) patients. Pain intensity and sensorimotor factors significantly influence balance, guiding personalized physiotherapy.
Area of Science:
- Neurology
- Biomedical Engineering
- Physical Therapy
Background:
- Cervicogenic headache (CEH) is linked to cervical spine dysfunction, impacting postural control and sensorimotor integration.
- Current assessments for CEH often overlook balance impairments, necessitating advanced analytical methods.
- Machine learning (ML) can integrate complex clinical data to understand postural stability in CEH.
Purpose of the Study:
- To apply ML algorithms to identify patterns in postural stability among CEH patients.
- To determine key clinical and sensorimotor factors influencing balance regulation in CEH.
- To explore the potential of ML for precision physiotherapy in CEH management.
Main Methods:
- A secondary analysis utilized baseline data from 68 CEH participants diagnosed per ICHD-3 criteria.
- Postural stability was categorized (High, Moderate, Low) using quantitative scores.
- Five ML algorithms (Gradient Boosting, Random Forest, XGBoost, SVM, Logistic Regression) were trained and validated using 10-fold cross-validation.
Main Results:
- Gradient Boosting achieved the highest accuracy (0.857) and F1 score (0.857), with a cross-validated accuracy of 0.802.
- Center-of-pressure sway and pain intensity were the strongest predictors of stability.
- Cervical flexion range of motion and gaze accuracy also contributed, while demographics had minimal impact.
Conclusions:
- ML models effectively differentiate postural stability levels in individuals with CEH.
- Pain and sensorimotor control are critical for balance regulation in CEH.
- Predictive analytics using ML can inform tailored rehabilitation strategies for precision physiotherapy.