Related Experiment Video
Updated: Sep 18, 2025
06:54
Clinical-oriented Three-dimensional Gait Analysis Method for Evaluating Gait Disorder
Published on: March 4, 2018
14.3K
Scoping Review of Machine Learning Techniques in Marker-Based Clinical Gait Analysis.
Kevin N Dibbern1,2,3, Maddalena G Krzak2,3, Alejandro Olivas3
1Department of Pediatrics, University of Nebraska Medical Center, Omaha, NE 68198, USA.
Bioengineering (Basel, Switzerland)
|June 26, 2025
Summary
Machine learning (ML) enhances 3D gait analysis (3DGA) for clinical insights. While deep learning excels with large datasets, traditional ML methods offer robustness and explainability, especially with limited data.
Area of Science:
- Biomedical Engineering
- Data Science
- Clinical Biomechanics
Background:
- Novel machine learning (ML) techniques are increasingly used in quantitative marker-based 3D gait analysis (3DGA).
- These advancements show potential for enhancing the interpretation of clinical gait analysis.
- There is a need to characterize the current literature on ML applications in 3DGA for clinical utility.
Purpose of the Study:
- To conduct a scoping review of the literature on machine learning applications in marker-based 3D gait analysis.
- To identify how ML is used to provide clinical insights for improved patient care.
- To characterize the state-of-the-art in ML for clinical gait analysis.
Main Methods:
- A comprehensive scoping review was performed using PubMed and Web of Science databases.
- Search terms were derived from relevant articles and refined by experts in gait analysis and machine learning.
- Study inclusion was determined through adjudication by three independent reviewers.
Main Results:
- The review identified 105 relevant papers from an initial search of 4324 articles.
- Commonly applied ML techniques include support vector machines, neural networks (NNs), and logistic regression.
- The most frequent clinical conditions analyzed were cerebral palsy, Parkinson's disease, and post-stroke patients.
Conclusions:
- Machine learning is broadly applied in gait analysis literature.
- Deep learning methods show success with large datasets, while traditional ML techniques are robust for smaller datasets and offer better explainability.
- Explainable AI (XAI) can enhance model interpretability but is not yet widely adopted in this field.
