Brain and behavior · 2026
Machine Learning-Based Classification of Conduct Disorder Comorbidity in Youth With ADHD.
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.
Çağır Büşra, Kurban Rifat, Özmen Sevgi
Аннотация
PURPOSE: To identify clinical profiles associated with conduct disorder (CD) among adolescents with attention-deficit/hyperactivity disorder (ADHD) using integrated statistical and machine learning approaches.
METHODS: A total of 138 adolescents aged 11-18 years (100 ADHD; 38 ADHD+CD) were assessed using multidimensional sociodemographic and psychometric measures. Ninety-nine variables were analyzed. Feature selection and classification analyses were performed using χ2 and mRMR methods combined with supervised machine learning models, including logistic regression, support vector machine (SVM), decision tree, random forest, gradient boosting, and artificial neural networks. Model performance was evaluated using accuracy, recall, specificity, and the area under the receiver operating characteristic curve (AUC).
RESULTS: Low maternal education, disciplinary punishment, school absenteeism, medication-free periods, peer smoking and delinquency, family criminal history, physical abuse, and later ADHD diagnosis (≥12 years) consistently emerged as key factors associated with CD. In the exploratory comparison, the Top-20 SVM had the highest observed accuracy and F1-score among the evaluated configurations (accuracy = 0.87; AUC = 0.90; F1-score = 0.71), with high specificity and moderate sensitivity; no single model-selection metric was prespecified.
CONCLUSION: These findings demonstrate that machine learning can identify clinical profiles associated with CD comorbidity in adolescents with ADHD. Such approaches may complement comprehensive clinical assessment but should not be interpreted as tools for predicting future CD development.
METHODS: A total of 138 adolescents aged 11-18 years (100 ADHD; 38 ADHD+CD) were assessed using multidimensional sociodemographic and psychometric measures. Ninety-nine variables were analyzed. Feature selection and classification analyses were performed using χ2 and mRMR methods combined with supervised machine learning models, including logistic regression, support vector machine (SVM), decision tree, random forest, gradient boosting, and artificial neural networks. Model performance was evaluated using accuracy, recall, specificity, and the area under the receiver operating characteristic curve (AUC).
RESULTS: Low maternal education, disciplinary punishment, school absenteeism, medication-free periods, peer smoking and delinquency, family criminal history, physical abuse, and later ADHD diagnosis (≥12 years) consistently emerged as key factors associated with CD. In the exploratory comparison, the Top-20 SVM had the highest observed accuracy and F1-score among the evaluated configurations (accuracy = 0.87; AUC = 0.90; F1-score = 0.71), with high specificity and moderate sensitivity; no single model-selection metric was prespecified.
CONCLUSION: These findings demonstrate that machine learning can identify clinical profiles associated with CD comorbidity in adolescents with ADHD. Such approaches may complement comprehensive clinical assessment but should not be interpreted as tools for predicting future CD development.