Multiband functional dysconnectivity patterns for exploratory identification of ASD–ADHD comorbidity
Journal of King Saud University Computer and Information Sciences · 2026-11 · 38 (9)
DOI: 10.1007/s44443-026-01345-7 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of King Saud University Computer and Information Sciences, online erschienen am 02.11.2026. Verfasst von Hongbiao Ye und 7 weiteren Autor:innen. Die Arbeit nennt 62 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Department of Science and Technology of Jilin Province.
Abstract (Originalsprache)
Autism spectrum disorder (ASD) and attention-deficit/hyperactivity disorder (ADHD) frequently co-occur and exhibit overlapping behavioral and neural characteristics, making differential diagnosis and mechanistic interpretation challenging. To address this issue, we propose MBF-Net, an interpretable graph learning framework that integrates multiband resting-state functional magnetic resonance imaging (rs-fMRI) features with phase-locking value (PLV)-based functional connectivity to characterize abnormal brain network organization in ASD, ADHD, and their comorbidity. By jointly modeling frequency-specific neural activity, temporal dynamics, and inter-regional phase synchronization, MBF-Net enables both accurate classification and biologically meaningful interpretation. Experiments on multi-site datasets from ADHD-200, ABIDE I, and ABIDE II achieved accuracies of 95.21% ± 1.12% (95% CI 93.82–96.60) for ADHD versus healthy controls and 87.52% ± 2.61% (95% CI 84.28–90.76) for ASD versus healthy controls under five repeated stratified 10-fold cross-validations, with subject-level paired tests confirming significance against the strongest baseline. For the more challenging exploratory three-class classification task, MBF-Net achieved 73.69% ± 2.17% accuracy, 69.32% ± 2.06% macro-F1, and 81.32% ± 0.96% macro-AUC under five repeated stratified 10-fold cross-validation, with the highest mean ACC and macro-AUC among compared methods. However, because this cohort combines ADHD-only subjects from a single ADHD-200 site with ASD-only and comorbid subjects from ABIDE II, part of the three-class accuracy may reflect cross-dataset composition. Consistently, a supplementary within-ABIDE II two-class analysis (ASD-only vs. comorbid ASD–ADHD), which removes this shortcut, yielded lower accuracy (67.8% pooled cross-validation; 64.0% leave-one-site-out), indicating that the present evidence supports exploratory characterization of ASD–ADHD comorbidity more strongly than site-independent three-way differential diagnosis.
Förderung
- Department of Science and Technology of Jilin Province
Zitieren
APA
Ye, H., Han, Q., Niu, L., Dong, X., Dong, H., Xue, Y., Zhao, J., & Kuang, Z. (2026). Multiband functional dysconnectivity patterns for exploratory identification of ASD–ADHD comorbidity. Journal of King Saud University Computer and Information Sciences, 38(9). https://doi.org/10.1007/s44443-026-01345-7
Vancouver
Ye H, Han Q, Niu L, Dong X, Dong H, Xue Y, et al. Multiband functional dysconnectivity patterns for exploratory identification of ASD–ADHD comorbidity. Journal of King Saud University Computer and Information Sciences. 2026;38(9). doi:10.1007/s44443-026-01345-7
BibTeX
@article{Ye2026,
author = {Ye, Hongbiao and Han, Qiulei and Niu, Liangjie and Dong, Xu and Dong, Hanyu and Xue, Yang and Zhao, Jian and Kuang, Zhejun},
title = {Multiband functional dysconnectivity patterns for exploratory identification of ASD–ADHD comorbidity},
journal = {Journal of King Saud University Computer and Information Sciences},
year = {2026},
volume = {38},
number = {9},
doi = {10.1007/s44443-026-01345-7}
}
EUVIMED