Graph Neural Networks for Analyzing Trauma-Related Brain Structure in Children and Adolescents: A Pilot Study
This study explores the potential of graph neural networks (GNNs) in analyzing brain networks of children and adolescents exposed to trauma, addressing limitations in traditional neuroimaging approaches. MRI-based brain data from trauma-exposed and control groups were modeled as whole-brain networks...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/15/1/277 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841549414568558592 |
|---|---|
| author | Harim Jeong Minjoo Kang Shanon McLeay R. J. R. Blair Unsun Chung Soonjo Hwang |
| author_facet | Harim Jeong Minjoo Kang Shanon McLeay R. J. R. Blair Unsun Chung Soonjo Hwang |
| author_sort | Harim Jeong |
| collection | DOAJ |
| description | This study explores the potential of graph neural networks (GNNs) in analyzing brain networks of children and adolescents exposed to trauma, addressing limitations in traditional neuroimaging approaches. MRI-based brain data from trauma-exposed and control groups were modeled as whole-brain networks using regions-of-interest (ROIs), with GNNs applied to capture complex, non-linear connectivity patterns. Results revealed that the trauma-exposed group exhibited simplified network structures with higher importance in regions associated with cognitive and emotional regulation, such as the posterior cerebellum. In contrast, the control group demonstrated richer connectivity patterns, emphasizing regions related to motor and visual processing, such as the Right Lingual Gyrus. Compared to traditional <i>t</i>-test results highlighting regional density differences, the GNN approach uncovered deeper, network-level insights into the relationships between brain regions. These findings demonstrate the utility of GNNs in advancing neuroimaging research, offering new perspectives on trauma’s impact on brain connectivity and paving the way for future applications in understanding neural mechanisms and interventions. |
| format | Article |
| id | doaj-art-b012715708d34dffb50bba4183cd9ec9 |
| institution | Kabale University |
| issn | 2076-3417 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj-art-b012715708d34dffb50bba4183cd9ec92025-01-10T13:15:00ZengMDPI AGApplied Sciences2076-34172024-12-0115127710.3390/app15010277Graph Neural Networks for Analyzing Trauma-Related Brain Structure in Children and Adolescents: A Pilot StudyHarim Jeong0Minjoo Kang1Shanon McLeay2R. J. R. Blair3Unsun Chung4Soonjo Hwang5Psychiatry Department, University of Nebraska Medical Center, Omaha, NE 68198, USAPsychiatry Department, University of Nebraska Medical Center, Omaha, NE 68198, USAPsychiatry Department, University of Nebraska Medical Center, Omaha, NE 68198, USAChild and Adolescent Mental Health Centre, Mental Health Services, Capital Region of Denmark, 2900 Copenhagen, DenmarkKyoungpook National University Hospital, Daegu 41905, Republic of KoreaPsychiatry Department, University of Nebraska Medical Center, Omaha, NE 68198, USAThis study explores the potential of graph neural networks (GNNs) in analyzing brain networks of children and adolescents exposed to trauma, addressing limitations in traditional neuroimaging approaches. MRI-based brain data from trauma-exposed and control groups were modeled as whole-brain networks using regions-of-interest (ROIs), with GNNs applied to capture complex, non-linear connectivity patterns. Results revealed that the trauma-exposed group exhibited simplified network structures with higher importance in regions associated with cognitive and emotional regulation, such as the posterior cerebellum. In contrast, the control group demonstrated richer connectivity patterns, emphasizing regions related to motor and visual processing, such as the Right Lingual Gyrus. Compared to traditional <i>t</i>-test results highlighting regional density differences, the GNN approach uncovered deeper, network-level insights into the relationships between brain regions. These findings demonstrate the utility of GNNs in advancing neuroimaging research, offering new perspectives on trauma’s impact on brain connectivity and paving the way for future applications in understanding neural mechanisms and interventions.https://www.mdpi.com/2076-3417/15/1/277graph neural networks (GNNs)brain structurechildhood traumaMRI data analysisneuroimagingpilot study |
| spellingShingle | Harim Jeong Minjoo Kang Shanon McLeay R. J. R. Blair Unsun Chung Soonjo Hwang Graph Neural Networks for Analyzing Trauma-Related Brain Structure in Children and Adolescents: A Pilot Study Applied Sciences graph neural networks (GNNs) brain structure childhood trauma MRI data analysis neuroimaging pilot study |
| title | Graph Neural Networks for Analyzing Trauma-Related Brain Structure in Children and Adolescents: A Pilot Study |
| title_full | Graph Neural Networks for Analyzing Trauma-Related Brain Structure in Children and Adolescents: A Pilot Study |
| title_fullStr | Graph Neural Networks for Analyzing Trauma-Related Brain Structure in Children and Adolescents: A Pilot Study |
| title_full_unstemmed | Graph Neural Networks for Analyzing Trauma-Related Brain Structure in Children and Adolescents: A Pilot Study |
| title_short | Graph Neural Networks for Analyzing Trauma-Related Brain Structure in Children and Adolescents: A Pilot Study |
| title_sort | graph neural networks for analyzing trauma related brain structure in children and adolescents a pilot study |
| topic | graph neural networks (GNNs) brain structure childhood trauma MRI data analysis neuroimaging pilot study |
| url | https://www.mdpi.com/2076-3417/15/1/277 |
| work_keys_str_mv | AT harimjeong graphneuralnetworksforanalyzingtraumarelatedbrainstructureinchildrenandadolescentsapilotstudy AT minjookang graphneuralnetworksforanalyzingtraumarelatedbrainstructureinchildrenandadolescentsapilotstudy AT shanonmcleay graphneuralnetworksforanalyzingtraumarelatedbrainstructureinchildrenandadolescentsapilotstudy AT rjrblair graphneuralnetworksforanalyzingtraumarelatedbrainstructureinchildrenandadolescentsapilotstudy AT unsunchung graphneuralnetworksforanalyzingtraumarelatedbrainstructureinchildrenandadolescentsapilotstudy AT soonjohwang graphneuralnetworksforanalyzingtraumarelatedbrainstructureinchildrenandadolescentsapilotstudy |