Related Experiment Video
Updated: Jun 6, 2025
Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable
Published on: May 17, 2024
Diagnosis of Autism in Children Based on their Gait Pattern and Movement Signs Using the Kinect Sensor
Shabnam Akhoondi Yazdi1, Amin Janghorbani1, Ali Maleki2
1Departement of Biotechnology, Faculty of New Sciences and Technologies, Semnan University, Semnan, Iran.
Insights
This study identifies autistic children using gait analysis from Kinect sensor data. Medical knowledge-based features achieved 87% accuracy, outperforming statistical features for early autism detection.
Area of Science:
- Biomedical Engineering
- Developmental Neuroscience
- Clinical Diagnostics
Background:
- Autism spectrum disorder (ASD) is a developmental condition affecting social interaction and communication.
- Early diagnosis of ASD is critical for intervention and reducing long-term effects.
- Autism presents as a movement disorder, characterized by atypical gait patterns and motor control issues.
Purpose of the Study:
- To identify autistic children using gait pattern analysis from Kinect sensor data.
- To compare the effectiveness of statistical features versus medical knowledge-based features for autism detection.
- To evaluate machine learning classifiers for classifying autistic children based on gait data.
Main Methods:
- Collected gait data (joint positions, angles) from 50 autistic and 50 typically developing children using a Kinect sensor.
- Extracted two sets of features: statistical gait parameters and features derived from known autistic behaviors.
- Applied statistical tests for feature selection and classified data using Naïve Bayes, SVM, k-NN, and ensemble methods.
Main Results:
- Medical knowledge-based features yielded the highest accuracy (87%) with an ensemble classifier, demonstrating 86% sensitivity and 88% specificity.
- Statistical features achieved 84% accuracy with Naïve Bayes, showing 86% sensitivity and 82% specificity.
- The 16-feature vector based on medical knowledge proved superior to the 42 statistical features.
Conclusions:
- Gait analysis using Kinect sensor data and medical knowledge-based features is a promising method for identifying autistic children.
- This approach offers a non-invasive, objective tool for early autism detection.
- The study highlights the potential of leveraging specific behavioral insights for improved diagnostic accuracy in ASD.
Background:
Autism spectrum disorders are a type of developmental disorder that primarily disrupt social interactions and communications. Autism has no treatment, but early diagnosis of it is crucial to reduce these effects. The incidence of autism is represented in repetitive patterns of children's motion. When walking, these children tighten their muscles and cannot control and maintain their body position. Autism is not only a mental health disorder but also a movement disorder.
Method:
This study aims to identify autistic children based on data recorded from their gait patterns using a Kinect sensor. The database used in this study comprises walking information, such as joint positions and angles between joints, of 50 autistic and 50 healthy children. Two groups of features were extracted from the Kinect data in this study. The first one was statistical features of joints' position and angles between joints. The second group was the features based on medical knowledge about autistic children's behaviors. Then, extracted features were evaluated through statistical tests, and optimal features were selected. Finally, these selected features were classified by naïve Bayes, support vector machine, k-nearest neighbors, and ensemble classifier.
Results:
The highest classification accuracy for medical knowledge-based features was 87% with 86% sensitivity and 88% specificity using an ensemble classifier; for statistical features, 84% of accuracy was obtained with 86% sensitivity and 82% specificity using naïve Bayes.
Conclusion:
The dimension of the resulted feature vector based on autistic children's medical knowledge was 16, with an accuracy of 87%, showing the superiority of these features compared to 42 statistical features.
