Identifying diagnostic neuroimaging biomarkers for adolescent major depressive disorder
Poster No:
376
Submission Type:
Abstract Submission
Authors:
Twain Dai1, Jennifer Rodger1
Institutions:
1The University of Western Australia & Perron Institute for Neurological and Translational Sciences, Perth, Western Australia, Australia
First Author:
Twain Dai
The University of Western Australia & Perron Institute for Neurological and Translational Sciences
Perth, Western Australia, Australia
The University of Western Australia & Perron Institute for Neurological and Translational Sciences
Perth, Western Australia, Australia
Co-Author:
Jennifer Rodger
The University of Western Australia & Perron Institute for Neurological and Translational Sciences
Perth, Western Australia, Australia
The University of Western Australia & Perron Institute for Neurological and Translational Sciences
Perth, Western Australia, Australia
Introduction:
The growing incidence of adolescent major depressive disorder or depression over the past decade presents a significant challenge to public health worldwide. Despite well-defined diagnostic criteria, adolescent depression can often be misdiagnosed due to unclear pathophysiology and the subjective nature of these criteria (Beirão et al., 2020). This highlights an urgent need to identify objective and reliable biomarkers to complement traditional diagnostics, which is crucial for developing effective treatment. The integration of resting-state functional magnetic resonance imaging (rs-fMRI) and machine learning has shown promising results in identification of reliable biomarkers and improvement of diagnostic accuracy for adult depression (Kambeitz et al., 2017). Unfortunately, relevant studies on adolescent depression are lacking (Lee et al., 2019). Therefore, the present study aimed to identify diagnostic rs-fMRI biomarkers for adolescent depression.
Methods:
The behavioural and rs-fMRI data of 127 adolescents (64 adolescents with depression and 63 healthy controls) were acquired from the Boston Adolescent Neuroimaging of Depression and Anxiety dataset (Hubbard et al., 2024). For each participant, depression score was calculated using the self-report Revised Children's Anxiety and Depression Scale (Chorpita et al., 2000). Partial correlation (Smith et al., 2013) was used to compute the functional connectivity between each pair of 268 nodes defined by the Shen parcellation (Noble et al., 2017). Random forest regression using the Boruta method (Kursa and Rudnicki, 2010) was executed 500 times to determine the most stable and important features, among 35780 variables (diagnosis, sex, and 35780 pairs of functional connectivity), for depression scores. A variable was considered stable and important if it presented in more than 250 out of 500 runs of the algorithm. Following feature selection, the whole dataset was randomly split into training and testing set, with a ratio of 80:20. Support vector machine with radial basis function and repeated two-fold cross-validation was employed to build a classification model to discriminate adolescents with depression from healthy controls, based on the most important rs-fMRI features.
Results:
The depression score of adolescents with depression (mean ± standard deviation: 70 ± 15) significantly differed from that of healthy controls (37 ± 6; unpaired Cohen's d: 2.915, 95%CI: 2.357 ~ 3.403; Figure 1). Five covariates, including diagnosis and functional connectivity between four pairs of nodes, were considered the most stable and important in mapping depression severity for healthy and depressed adolescents (Figure 2). Using these four rs-fMRI features, the classification model achieved a balanced accuracy of 87.5%, with a sensitivity of 75% and specificity of 100% in the testing set.
Conclusions:
The present study demonstrates the clinical relevance of connectome-based rs-fMRI data in adolescent depression and provides valuable insights into neuroimaging biomarkers that are informative for depression diagnosis in adolescence. The classification model developed here achieves better performance compared to the literature (Sen et al., 2021), highlighting a critical role of functional connectome localized across the frontoparietal, medial frontal, default mode, motor and subcortical networks (Figure 2B) in adolescent depression. The rs-fMRI biomarkers identified here do not exactly coincide with those of adult depression (Bondi et al., 2023), suggesting that there are neurobiological differences between adult and adolescent depression. But further studies comparing these biomarkers in healthy adolescents and adults are warranted to exclude the possibility that these differences are due to brain maturation during adolescence. The current work represents an important step toward characterization of adolescent depression, which will eventually facilitate the development of individualized treatment strategies.
Disorders of the Nervous System:
Psychiatric (eg. Depression, Anxiety, Schizophrenia) 1
Modeling and Analysis Methods:
Classification and Predictive Modeling 2
Connectivity (eg. functional, effective, structural)
Multivariate Approaches
Task-Independent and Resting-State Analysis
Keywords:
FUNCTIONAL MRI
Machine Learning
Multivariate
Pediatric Disorders
Psychiatric Disorders
Other - adolescent depression; resting-state fMRI; diagnostic biomarkers
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
Bondi, E., Maggioni, E., Brambilla, P. & Delvecchio, G. 2023. A systematic review on the potential use of machine learning to classify major depressive disorder from healthy controls using resting state fMRI measures. Neuroscience and biobehavioral reviews, 144, 104972-104972.
Chorpita, B. F., Yim, L., Moffitt, C., Umemoto, L. A. & Francis, S. E. 2000. Assessment of symptoms of DSM-IV anxiety and depression in children: a revised child anxiety and depression scale. Behaviour research and therapy, 38, 835-855.
Hubbard, N. A., Bauer, C. C. C., Siless, V., Auerbach, R. P., Elam, J. S., Frosch, I. R., Henin, A., Hofmann, S. G., Hodge, M. R., Jones, R., Lenzini, P., Lo, N., Park, A. T., Pizzagalli, D. A., Vaz-Desouza, F., Gabrieli, J. D. E., Whitfield-Gabrieli, S., Yendiki, A. & Ghosh, S. S. 2024. The Human Connectome Project of adolescent anxiety and depression dataset. Scientific data, 11, 837-15.
Kambeitz, J. M. D., Cabral, C. M. A., Sacchet, M. D. S., Gotlib, I. H. P., Zahn, R. M. D., Serpa, M. H. M. D., Walter, M., Falkai, P. & Koutsouleris, N. 2017. Detecting Neuroimaging Biomarkers for Depression: A Meta-Analysis of Multivariate Pattern Recognition Studies. Biological psychiatry (1969), 82, 330-338.
Kursa, M. B. & Rudnicki, W. R. 2010. Feature Selection with the Boruta Package. Journal of statistical software, 36.
Lee, J., Pavuluri, M. N., Kim, J. H., Suh, S., Kim, I. & Lee, M.-S. 2019. Resting-state functional connectivity in medication-naïve adolescents with major depressive disorder. Psychiatry research. Neuroimaging, 288, 37-43.
Noble, S., Spann, M. N., Tokoglu, F., Shen, X., Constable, R. T. & Scheinost, D. 2017. Influences on the Test-Retest Reliability of Functional Connectivity MRI and its Relationship with Behavioral Utility. Cerebral cortex (New York, N.Y. 1991), 27, 5415-5429.
Sen, B., Cullen, K. R. & Parhi, K. K. 2021. Classification of Adolescent Major Depressive Disorder Via Static and Dynamic Connectivity. IEEE journal of biomedical and health informatics, 25, 2604-2614.
Smith, S. M., Vidaurre, D., Beckmann, C. F., Glasser, M. F., Jenkinson, M., Miller, K. L., Nichols, T. E., Robinson, E. C., Salimi-Khorshidi, G., Woolrich, M. W., Barch, D. M., Ugurbil, K. & Van Essen, D. C. 2013. Functional connectomics from resting-state fMRI. Trends Cogn Sci, 17, 666-82.