Related Experiment Video
Updated: Jul 18, 2026
Deep-Learning Based Multi-Joint Synchronous Tracking for Objective Quantification of Hindlimb Locomotor Kinematics in Rats
Published on: April 3, 2026
Support vector machines and other pattern recognition approaches to the diagnosis of cerebral palsy gait
Joarder Kamruzzaman1, Rezaul K Begg
1School of Information Technology, Monash University, Gippsland Campus, Churchill, Vic 3842, Australia. Joarder.Kamruzzaman@infotech.monash.edu.au
Insights
Support Vector Machines (SVM) accurately identify cerebral palsy (CP) gait in children using stride length and cadence. Normalizing gait parameters significantly improves classification accuracy for better diagnosis and treatment evaluation.
Area of Science:
- Biomedical Engineering
- Gait Analysis
- Machine Learning in Medicine
Background:
- Cerebral palsy (CP) gait identification is crucial for diagnosis and treatment assessment.
- Automated gait analysis offers objective evaluation methods.
Purpose of the Study:
- To explore Support Vector Machines (SVM) for automated CP gait classification in children.
- To evaluate the effectiveness of basic temporal-spatial gait parameters for CP detection.
Main Methods:
- Utilized SVM classifier with stride length and cadence as input features.
- Applied a tenfold cross-validation scheme on a dataset of healthy children and those with spastic diplegia CP.
- Investigated the impact of normalizing gait parameters by leg length and age.
Main Results:
- Initial SVM classification achieved 83.33% accuracy.
- Normalized gait parameters significantly improved accuracy to 96.80%, outperforming linear discriminant analysis and multilayer-perceptron.
- Polynomial and radial basis SVM kernels demonstrated superior performance.
Conclusions:
- SVM provides a highly accurate and efficient method for identifying CP gait in children.
- Normalized gait parameters enhance classification accuracy, aiding in diagnosis and treatment monitoring.
- This approach is valuable for evaluating rehabilitation techniques and treatment outcomes in CP.
Abstract:
Accurate identification of cerebral palsy (CP) gait is important for diagnosis as well as for proper evaluation of the treatment outcomes. This paper explores the use of support vector machines (SVM) for automated detection and classification of children with CP using two basic temporal-spatial gait parameters (stride length and cadence) as input features. Application of the SVM method to a children's dataset (68 normal healthy and 88 with spastic diplegia form of CP) and testing on tenfold cross-validation scheme demonstrated that an SVM classifier was able to classify the children groups with an overall accuracy of 83.33% [sensitivity 82.95%, specificity 83.82%, area under the receiver operating curve (AUC-ROC = 0.88)]. Classification accuracy improved significantly when the gait parameters were normalized by the individual leg length and age, leading to an overall accuracy of 96.80% (sensitivity 94.32%, specificity 100%, AUC-ROC area = 0.9924). This accuracy result was, respectively, 3.21% and 1.93% higher when compared to an linear discriminant analysis and an multilayer-perceptron-based classifier. SVM classifier also attains considerably higher ROC area than the other two classifiers. Among the four SVM kernel functions (linear, polynomial, radial basis, and analysis of variance spline) studied, the polynomial and radial basis kernel performed comparably and outperformed the others. Classifier's performance as functions of regularization and kernel parameters was also investigated. The enhanced classification accuracy of the SVM using only two easily obtainable basic gait parameters makes it attractive for identifying CP children as well as for evaluating the effectiveness of various treatment methods and rehabilitation techniques.
