Peer-reviewed article
Application of Machine Learning Techniques to Detect the Children with Autism Spectrum Disorder
Mengyi Liao, Hengyao Duan, Guangshuai Wang ·
Summary
Written by AI, not yet AICS's view · claude-haiku-4-5-20251001This study developed a machine learning system to detect autism spectrum disorder (ASD) in children by combining physiological data from brain activity (EEG) with behavioral observations of eye movement and facial expression. The system achieved 87.5% accuracy in distinguishing children with ASD from typically developing children. The approach aims to make early detection faster and less costly than current methods that rely on expert assessment. The research found that brain activity data was most useful for detection, but combining all three data types improved results.
Key points
- Machine learning classifier achieved 87.50% accuracy combining EEG, eye fixation, and facial expression data
- EEG data showed the strongest discriminative power for detecting ASD
- Physiological and behavioral data have complementary characteristics that improve classification when combined
- The approach could reduce costs and improve efficiency compared to expert-dependent assessment methods
- Feature extraction techniques were developed for eye fixation, facial expression, and EEG data
Limitations
The document does not report sample size, participant demographics, or validation on independent test sets. It provides no information on the age range of children studied, geographic diversity, or whether the system was tested in real-world clinical settings.
Abstract
Early detection of autism spectrum disorder (ASD) is highly beneficial to the health sustainability of children. Existing detection methods depend on the assessment of experts, which are subjective and costly. In this study, we proposed a machine learning approach that fuses physiological data (electroencephalography, EEG) and behavioral data (eye fixation and facial expression) to detect children with ASD. Its implementation can improve detection efficiency and reduce costs. First, we used an innovative approach to extract features of eye fixation, facial expression, and EEG data. Then, a hybrid fusion approach based on a weighted naive Bayes algorithm was presented for multimodal data fusion with a classification accuracy of 87.50%. Results suggest that the machine learning classification approach in this study is effective for the early detection of ASD. Confusion matrices and graphs demonstrate that eye fixation, facial expression, and EEG have different discriminative powers for the detection of ASD and typically developing children, and EEG may be the most discriminative information. The physiological and behavioral data have important complementary characteristics. Thus, the machine learning approach proposed in this study, which combines the complementary information, can significantly improve classification accuracy.
Findings the Hub has extracted
Each claim below is labelled by its type of design and links back to this report.
- Intervention
- Machine learning system using weighted naive Bayes algorithm to fuse EEG, eye fixation, and facial expression data for ASD detection
- Population
- Children with autism spectrum disorder and typically developing children (specific sample size not stated)
Key finding
Classification accuracy of 87.50% achieved with hybrid fusion approach