Automated classification of attention deficit hyperactivity disorder and conduct disorder using entropy features with ECG signals

• Published in: Computers in biology and medicine Vol. 140; p. 105120
• Main Authors: Koh, Joel.E.W., Ooi, Chui Ping, Lim-Ashworth, Nikki SJ, Vicnesh, Jahmunah, Tor, Hui Tian, Lih, Oh Shu, Tan, Ru-San, Acharya, U.Rajendra, Fung, Daniel Shuen Sheng
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.01.2022; Elsevier Limited
• Subjects: Adaptive sampling; Attention deficit hyperactivity disorder; Automation; Autonomic nervous system; Classification; Classifiers; Comorbidity; Conduct disorder; Disorders; Dynamical systems; EKG; Electrocardiography; Electroencephalography; Empirical analysis; Empirical wavelet transform; Ensemble classifiers; Entropies; Entropy; Expert systems; Feature extraction; Heart; Hyperactivity; Internal Medicine; Machine learning; Medical personnel; Mental disorders; Mental health; Other; Social skills; Software; Software development tools; Teenagers; Ten-fold validation; Time series; Variance analysis; Wavelet transforms; Entropies; Empirical wavelet transform; Attention deficit hyperactivity disorder; Ensemble classifiers; Machine learning; Conduct disorder; Ten-fold validation
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2021.105120

The most prevalent neuropsychiatric disorder among children is attention deficit hyperactivity disorder (ADHD). ADHD presents with a high prevalence of comorbid disorders such as conduct disorder (CD). The lack of definitive confirmatory diagnostic tests for ADHD and CD make diagnosis challenging. The distinction between ADHD, ADHD + CD and CD is important as the course and treatment are different. Electrocardiography (ECG) signals may become altered in behavioral disorders due to brain-heart autonomic interactions. We have developed a software tool to categorize ADHD, ADHD + CD and CD automatically on ECG signals. ECG signals from participants were decomposed using empirical wavelet transform into various modes, from which entropy features were extracted. Robust ten-fold cross-validation with adaptive synthetic sampling (ADASYN) and z-score normalization were performed at each fold. Analysis of variance (ANOVA) technique was employed to determine the variability within the three classes, and obtained the most discriminatory features. Highly significant entropy features were then fed to classifiers. Our model yielded the best classification results with the bagged tree classifier: 87.19%, 87.71% and 86.29% for accuracy, sensitivity and specificity, respectively. The proposed expert system can potentially assist mental health professionals in the stratification of the three classes, for appropriate intervention using accessible ECG signals. •Novel classification technique for ADHD using ECG signals has been proposed.•A high classification accuracy of about 88% has been yielded with the bagged tree classifier.•The developed system is robust as it has been validated by 10-fold cross validation.•The proposed expert system can potentially assist mental health professional in the stratification of ADHD, ADHD + CD and CD patients for appropriate intervention using accessible ECG signals.