Poster No:
1106
Submission Type:
Abstract Submission
Authors:
Josh Nguyen1, Dominic Dwyer1, Scott Tagliaferri1, Simon Hartmann1, Scott Clark1, Isabelle Scott1, Johanna Wigman2, Ashleigh Lin3, Andrew Thompson4, Cassandra Wannan4, Caroline Gao4, Stephen Wood4, Paul Amminger4, Alison Yung5, Nikolaos Koutsouleris6, Jessica Hartmann7, Hok Pan Yuen8, Christopher Davey9, Angelica Ronald10, Patrick McGorry1, Christel Middeldorp11, Barnaby Nelson1, Lianne Schmaal1
Institutions:
1Centre for Youth Mental Health, The University of Melbourne, Melbourne, Australia, 2University Medical Centre Groningen, Groningen, 9713GZ, Netherlands, 3School of Population and Global Health, The University of Western Australia, Perth, WA, 4Centre for Youth Mental Health, The University of Melbourne, Parkville, Victoria, 5Deakin University, Victoria, Australia, 6University of Munich, Munich, Germany, 7Department of Psychiatry, Chonnam National University Medical School, Gwangju, Korea, 8Centre for Youth Mental Health, The University of Melbourne, Melbourne, Victoria, 9Department of Psychiatry, The University of Melbourne, Melbourne, VIC, 10Department of Psychology, University of Surrey, Guildford, Surrey, GU2 7XH, UK, 11Department of Child and Adolescent Psychiatry and Psychology, Amsterdam, Netherlands
First Author:
Josh Nguyen
Centre for Youth Mental Health, The University of Melbourne
Melbourne, Australia
Co-Author(s):
Dominic Dwyer
Centre for Youth Mental Health, The University of Melbourne
Melbourne, Australia
Scott Tagliaferri
Centre for Youth Mental Health, The University of Melbourne
Melbourne, Australia
Simon Hartmann
Centre for Youth Mental Health, The University of Melbourne
Melbourne, Australia
Scott Clark
Centre for Youth Mental Health, The University of Melbourne
Melbourne, Australia
Isabelle Scott
Centre for Youth Mental Health, The University of Melbourne
Melbourne, Australia
Johanna Wigman
University Medical Centre Groningen
Groningen, 9713GZ, Netherlands
Ashleigh Lin
School of Population and Global Health, The University of Western Australia
Perth, WA
Andrew Thompson
Centre for Youth Mental Health, The University of Melbourne
Parkville, Victoria
Cassandra Wannan
Centre for Youth Mental Health, The University of Melbourne
Parkville, Victoria
Caroline Gao
Centre for Youth Mental Health, The University of Melbourne
Parkville, Victoria
Stephen Wood
Centre for Youth Mental Health, The University of Melbourne
Parkville, Victoria
Paul Amminger
Centre for Youth Mental Health, The University of Melbourne
Parkville, Victoria
Jessica Hartmann
Department of Psychiatry, Chonnam National University Medical School
Gwangju, Korea
Hok Pan Yuen
Centre for Youth Mental Health, The University of Melbourne
Melbourne, Victoria
Christopher Davey
Department of Psychiatry, The University of Melbourne
Melbourne, VIC
Angelica Ronald
Department of Psychology, University of Surrey
Guildford, Surrey, GU2 7XH, UK
Patrick McGorry
Centre for Youth Mental Health, The University of Melbourne
Melbourne, Australia
Christel Middeldorp
Department of Child and Adolescent Psychiatry and Psychology
Amsterdam, Netherlands
Barnaby Nelson
Centre for Youth Mental Health, The University of Melbourne
Melbourne, Australia
Lianne Schmaal
Centre for Youth Mental Health, The University of Melbourne
Melbourne, Australia
Introduction:
Suicide is among the leading causes of death in adolescents (World Health Organization 2021, n.d.). Despite the high prevalence of suicide ideations (17%), only a third of those with suicidal ideation will attempt suicide (Nock et al., 2013). The key challenge is to predict which suicidal ideators will eventually attempt suicide. To date, few empirical longitudinal studies have addressed this critical question in suicide research. Most prior research has been cross-sectional, focused primarily on the adult population, and is limited to a narrow set of clinical features, often overlooking potential biological predictors, including structural and functional MRI features (Franklin et al., 2017). Our study is the first to develop a machine learning model for predicting the first onset of suicide attempt over 4-year follow-ups among adolescent ideators using 185 sociodemographic, clinical, neurocognitive, functional, and structural neuroimaging features.
Methods:
Five-wave data from the multisite, longitudinal Adolescent Brain Cognitive Development (ABCD) study were used. Using the 10-fold nested cross-validation, our model was trained on 70% of the sample across 15 sites and externally validated in 7 holdout sites using the top 10 contributing features. Racial/ethnic and gender biases in model performance were investigated. High-resolution T1-weighted images and task-based fMRI data were obtained at each ABCD site using 3T MRI systems. Based on a two-decade review of 131 neuroimaging studies (Schmaal et al., 2020), we selected 21 regions exclusively linked with suicide attempt or the ideation-attempt transition. FreeSurfer (version 5.3.0) was used to derive cortical thickness, cortical and subcortical volumes, and surface area for these regions. Additionally, we included neural activations (n=77 parcellations) evoked by two fMRI tasks: Monetary Incentive Delay task (anticipation of reward/loss versus neutral conditions; Knutson et al., 2000) and Stop Signal Task (correct stop versus correct go condition; Logan, 1994).
Results:
Among 828 adolescents (101 suicide attempts, 727 suicidal ideation only; mean [SD] age: 118.8 [7.57] months; 42% female), the top 10 risk factors included the presence of non-suicidal self-injury, access to means, generalized anxiety disorder, a lack of history of mental health treatment, low parental monitoring, reduced activation in the left caudate during correct stop versus correct go trials, reduced cortical surface area in the right caudal middle frontal region, victim of bullying, increased depressive symptoms and impulsivity (negative urgency). The final elastic net model using these top predictors were validated in 7 holdout sites and yielded moderate performance with area under the curve (AUC[SD] = 0.70[0.04], p < .001, negative predictive validity = 0.94 [0.02], specificity = 0.70[0.02], sensitivity = 0.58[0.03], PPV = 0.25[0.02]). Model calibration was good with an average error between predicted and observed probabilities of 0.03.
Conclusions:
This study presents the first machine learning model to utilize multimodal features for predicting the transition from suicidal ideation to attempt in adolescents. Our model outperforms current clinical judgment, which has been shown to perform only slightly better than chance. In addition to clinical and neurocognitive risk factors, we identified new brain signatures that indicate impaired inhibitory control and executive functions that predict the suicide ideation-attempt transition. These findings suggest that suicide risk assessments should adopt a comprehensive, biopsychosocial approach, integrating both psychological and neurobiological factors. Therefore, suicide risk assessments should consider adopting a comprehensive, biopsychosocial approach. Future research should evaluate the cost-benefit of integrating such machine learning models into clinical care to enhance the early detection and prevention of suicide risk.
Disorders of the Nervous System:
Psychiatric (eg. Depression, Anxiety, Schizophrenia) 2
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI)
Classification and Predictive Modeling 1
Connectivity (eg. functional, effective, structural)
Novel Imaging Acquisition Methods:
Multi-Modal Imaging
Keywords:
Computational Neuroscience
1|2Indicates the priority used for review
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.
I accept
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:
I do not want to participate in the reproducibility challenge.
Please indicate below if your study was a "resting state" or "task-activation” study.
Task-activation
Other
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Patients
Was this research conducted in the United States?
Yes
Are you Internal Review Board (IRB) certified?
Please note: Failure to have IRB, if applicable will lead to automatic rejection of abstract.
Yes, I have IRB or AUCC approval
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.
Yes
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.
Not applicable
Please indicate which methods were used in your research:
Functional MRI
Structural MRI
Computational modeling
For human MRI, what field strength scanner do you use?
3.0T
Which processing packages did you use for your study?
Free Surfer
Provide references using APA citation style.
1. Jaroszewski, A. C. (2017). Risk factors for suicidal thoughts and behaviors: A meta-analysis of 50 years of research. Psychological Bulletin, 143(2), 187–232.
2. Knutson, B. (2000). FMRI Visualization of Brain Activity during a Monetary Incentive Delay Task. NeuroImage, 12(1), 20–27.
3. Logan, G. D. (1994). On the ability to inhibit thought and action: A users’ guide to the stop signal paradigm. In Inhibitory processes in attention, memory, and language (pp. 189–239). Academic Press.
4. Nock, M. K., Green (2013). Prevalence, Correlates, and Treatment of Lifetime Suicidal Behavior Among Adolescents: Results From the National Comorbidity Survey Replication Adolescent Supplement. JAMA Psychiatry, 70(3), 300–310.
5. Schmaal, L. (2020). Imaging suicidal thoughts and behaviors: A comprehensive review of 2 decades of neuroimaging studies. Molecular Psychiatry, 25(2), 408–427.
6. World Health Organization 2021. (n.d.). Suicide worldwide in 2019: :Global Health Estimates.
No