Transfer Learning of Deep Neural Networks Pretrained Using the ABCD Dataset for General Psychopathology Prediction in Korean Adolescents
Biological Psychiatry: Cognitive Neuroscience and Neuroimaging · 2025
DOI: 10.1016/j.bpsc.2025.04.005 ↗Kurz gefasst
Fachartikel in der Zeitschrift Biological Psychiatry: Cognitive Neuroscience and Neuroimaging, 2025. Verfasst von Jundong Hwang und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Brain and Cognitive Engineering, Korea University, Seoul, South Korea. Hauptthemen laut MEDLINE: Deep Learning, Major Depressive Disorder, Brain, Neural Networks, Computer. Die Arbeit nennt 57 Literaturangaben. Gefördert von: National Research Foundation of Korea, National Research Foundation, Ministry of Health and Welfare.
Abstract (Originalsprache)
BACKGROUND: In this study, we examined whether a deep neural network (DNN), trained to predict the general psychopathology factor (p factor) using functional magnetic resonance imaging (fMRI) data from adolescents in the ABCD (Adolescent Brain Cognitive Development) Study, would generalize to Korean adolescents. METHODS: We trained a scanner-generalization neural network (SGNN) to predict p factor scores from resting-state functional connectivity (RSFC) data of 6905 ABCD Study adolescents, controlling for MRI scanner-related confounds. Then, we transferred the pretrained SGNN to a DNN to predict p factor scores for 125 adolescents, including healthy individuals and individuals with major depressive disorder, using data from Seoul National University Hospital (SNUH). We compared the transferred DNN's performance with that of kernel ridge regression (KRR) and a baseline DNN. RESULTS: The transferred DNN outperformed KRR (0.17 ± 0.16; 0.60 ± 0.07) and the baseline DNN (0.17 ± 0.16; 0.69 ± 0.11), with a higher Pearson's correlation coefficient (0.29 ± 0.18) and lower mean absolute error (0.59 ± 0.09; p < .005). We identified the default mode network (DMN) and visual network (VIS) as crucial functional networks for predicting p factors across both datasets. The dorsal attention network was specific to the ABCD Study dataset, while the cingulo-opercular and ventral attention networks were specific to the SNUH dataset. CONCLUSIONS: The transferred SGNN successfully generalized to Korean adolescents. Altered RSFC in the DMN and VIS may serve as promising biomarkers for p factor prediction across diverse populations, addressing heterogeneity in demographics, diagnoses, and MRI scanner characteristics.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Brain and Cognitive Engineering, Korea University, Seoul, South Korea.
- Division of Child and Adolescent Psychiatry, Department of Psychiatry, Seoul National University Hospital, Seoul, South Korea.
- Division of Child and Adolescent Psychiatry, Department of Psychiatry, Seoul National University Hospital, Seoul, South Korea. Electronic address: kimjw412@snu.ac.kr.
- Department of Brain and Cognitive Engineering, Korea University, Seoul, South Korea; Interdisciplinary Program in Precision Public Health, Korea University, Seoul, South Korea; McGovern Institute for Brain Research, Mas…
Förderung
- National Research Foundation of Korea
- National Research Foundation
- Ministry of Health and Welfare
- Ministry of Science ICT and Future Planning
- Korea Health Industry Development Institute
- Electronics and Telecommunications Research Institute
Zitieren
APA
Hwang, J., Kang, J. E., Jeon, S., Lee, K. H., Kim, J. W., & Lee, J. H. (2025). Transfer Learning of Deep Neural Networks Pretrained Using the ABCD Dataset for General Psychopathology Prediction in Korean Adolescents. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging. https://doi.org/10.1016/j.bpsc.2025.04.005
Vancouver
Hwang J, Kang JE, Jeon S, Lee KH, Kim JW, Lee JH. Transfer Learning of Deep Neural Networks Pretrained Using the ABCD Dataset for General Psychopathology Prediction in Korean Adolescents. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging. 2025. doi:10.1016/j.bpsc.2025.04.005
BibTeX
@article{Hwang2025,
author = {Hwang, Jundong and Kang, Jae-Eon and Jeon, Soohyun and Lee, Kyung Hwa and Kim, Jae-Won and Lee, Jong-Hwan},
title = {Transfer Learning of Deep Neural Networks Pretrained Using the ABCD Dataset for General Psychopathology Prediction in Korean Adolescents},
journal = {Biological Psychiatry: Cognitive Neuroscience and Neuroimaging},
year = {2025},
doi = {10.1016/j.bpsc.2025.04.005}
}
EUVIMED