ABSTRACT
OBJECTIVES: To investigate the efficacy and required indicators of Children Neuropsychological and Behavioral Scale-Revision 2016 (CNBS-R2016) in the differential diagnosis of autism spectrum disorder (ASD) and global developmental delay (GDD). METHODS: A total of 277 children with ASD and 415 children with GDD, aged 18-48 months, were enrolled as subjects. CNBS-R2016 was used to assess the developmental levels of six domains, i.e., gross motor, fine motor, adaptive ability, language, social behavior, and warning behavior, and a total of 13 indicators on intelligence age and developmental quotient (DQ) were obtained as the input features. Five commonly used machine learning classifiers were used for training to calculate the classification accuracy, sensitivity, and specificity of each classifier. RESULTS: DQ of warning behavior was selected as the first feature in all five classifiers, and the use of this indicator alone had a classification accuracy of 78.90%. When the DQ of warning behavior was used in combination with the intelligence age of warning behavior, gross motor, and language, it had the highest classification accuracy of 86.71%. CONCLUSIONS: Machine learning combined with CNBS-R2016 can effectively distinguish children with ASD from those with GDD. The DQ of warning behavior plays an important role in machine learning, and its combination with other features can improve classification accuracy, providing a basis for the efficient and accurate differential diagnosis of ASD and GDD in clinical practice.
Subject(s)
Autism Spectrum Disorder , Child , Humans , Autism Spectrum Disorder/diagnosis , Autism Spectrum Disorder/psychology , Diagnosis, Differential , Machine Learning , Social BehaviorABSTRACT
OBJECTIVES@#To investigate the efficacy and required indicators of Children Neuropsychological and Behavioral Scale-Revision 2016 (CNBS-R2016) in the differential diagnosis of autism spectrum disorder (ASD) and global developmental delay (GDD).@*METHODS@#A total of 277 children with ASD and 415 children with GDD, aged 18-48 months, were enrolled as subjects. CNBS-R2016 was used to assess the developmental levels of six domains, i.e., gross motor, fine motor, adaptive ability, language, social behavior, and warning behavior, and a total of 13 indicators on intelligence age and developmental quotient (DQ) were obtained as the input features. Five commonly used machine learning classifiers were used for training to calculate the classification accuracy, sensitivity, and specificity of each classifier.@*RESULTS@#DQ of warning behavior was selected as the first feature in all five classifiers, and the use of this indicator alone had a classification accuracy of 78.90%. When the DQ of warning behavior was used in combination with the intelligence age of warning behavior, gross motor, and language, it had the highest classification accuracy of 86.71%.@*CONCLUSIONS@#Machine learning combined with CNBS-R2016 can effectively distinguish children with ASD from those with GDD. The DQ of warning behavior plays an important role in machine learning, and its combination with other features can improve classification accuracy, providing a basis for the efficient and accurate differential diagnosis of ASD and GDD in clinical practice.
Subject(s)
Child , Humans , Autism Spectrum Disorder/psychology , Diagnosis, Differential , Machine Learning , Social BehaviorABSTRACT
Background: The Children Neuropsychological and Behavioral Scale-Revision 2016 (CNBS-R2016) is a widely used developmental assessment tool for children aged 0-6 years in China. The communication warning behavior subscale of CNBS-R2016 is used to assess the symptoms of autism spectrum disorder (ASD), and its value of >30 points indicates ASD based on CNBS-R2016. However, we observed that children with relatively lower values were also diagnosed with ASD later on in clinical practice. Thus, this study aimed to identify the suitable cutoff value for ASD screening recommended by the communication warning behavior of CNBS-R2016. Materials and methods: A total of 90 typically developing (TD) children and 316 children with developmental disorders such as ASD, developmental language disorder (DLD), and global developmental delay (GDD; 130 in the ASD group, 100 in the DLD group, and 86 in the GDD group) were enrolled in this study. All subjects were evaluated based on the CNBS-R2016. The newly recommended cutoff value of communication warning behavior for screening ASD was analyzed with receiver operating curves. Results: Children in the ASD group presented with lower developmental levels than TD, DLD, and GDD groups in overall developmental quotient assessed by CNBS-R2016. We compared the consistency between the scores of communication warning behavior subscale and Autism Behavior Checklist (ABC), Childhood Autism Rating Scale (CARS), Autism Diagnostic Observation Schedule, second edition (ADOS-2), and clinical diagnosis for the classification of ASD at a value of 30 based on the previously and newly recommended cutoff value of 12 by the CNBS-R2016. The Kappa values between the communication warning behavior and ABC, CARS, ADOS-2, and clinical diagnosis were 0.494, 0.476, 0.137, and 0.529, respectively, with an agreement rate of 76.90%, 76.26%, 52.03%, and 82.27%, respectively, when the cutoff point was 30. The corresponding Kappa values were 0.891, 0.816, 0.613, and 0.844, respectively, and the corresponding agreement rate was 94.62%, 90.82%, 90.54%, and 93.10%, respectively, when the cutoff point was 12. Conclusion: The communication warning behavior subscale of CNBS-R2016 is important for screening ASD. When the communication warning behavior score is 12 points or greater, considerable attention and further comprehensive diagnostic evaluation for ASD are required to achieve the early detection and diagnosis of ASD in children.
