BMC psychiatry · 2026
Machine learning-based prediction of intellectual disability in children with autism spectrum disorder: using behavioral observation techniques.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Liu Xiaolin, Han Ting, Jiang Zhongquan, Hu Lifei, Li Wenhao, Song Chao
Аннотация
PURPOSE: To construct and evaluate a LightGBM prediction model for intellectual disabilities in children with Autism Spectrum Disorder (ASD).
METHODS: A total of 384 ASD children who completed the Wechsler Intelligence Test and Adaptive Behavioral Assessment System were included in the analysis. The LightGBM model was trained using behavioral observation data and underwent hyperparameter tuning and feature selection.
RESULTS: Among the ASD children, 32.9% had comorbid ID. The LightGBM model exhibited the highest sensitivity and accuracy, with values of 0.793 and 0.760, respectively. It also achieved an AUC of 0.747, with overall quality of relationships, unusual sensory interests, and gestures/postures being the top predictive features.
CONCLUSION: The LightGBM model demonstrated strong predictive performance, enabling early identification of comorbid ID in preschool children with ASD and facilitating personalized intervention strategies.
METHODS: A total of 384 ASD children who completed the Wechsler Intelligence Test and Adaptive Behavioral Assessment System were included in the analysis. The LightGBM model was trained using behavioral observation data and underwent hyperparameter tuning and feature selection.
RESULTS: Among the ASD children, 32.9% had comorbid ID. The LightGBM model exhibited the highest sensitivity and accuracy, with values of 0.793 and 0.760, respectively. It also achieved an AUC of 0.747, with overall quality of relationships, unusual sensory interests, and gestures/postures being the top predictive features.
CONCLUSION: The LightGBM model demonstrated strong predictive performance, enabling early identification of comorbid ID in preschool children with ASD and facilitating personalized intervention strategies.