Multimodal MRI Marker Captures the Relationship Between Cognition and Mental Health in UK Biobank
Poster No:
1102
Submission Type:
Abstract Submission
Authors:
Irina Buianova1, Mateus Silvestrin2, Jeremiah Deng3, Narun Pat1
Institutions:
1Department of Psychology, University of Otago, Dunedin, New Zealand, 2Federal University of the São Francisco Valley, Petrolina, Brazil, 3School of Computing, University of Otago, Dunedin, New Zealand
First Author:
Co-Author(s):
Introduction:
Cognitive deficits are common in psychiatric illnesses (Suddell et al., 2023). Research suggests this interrelation stems from shared neurobiological substrates (Etkin et al., 2013). Nevertheless, we still lack robust neural indicators that capture this relationship. Brain Magnetic Resonance Imaging (MRI) and machine learning studies have demonstrated that MRI data can be used to derive neural indicators that predict cognition with reasonable performance (Pat et al., 2022). Yet, different MRI modalities capture distinct aspects of the brain and various MRI quantification techniques result in different neuroimaging phenotypes. Given the heterogeneity of neuroimaging phenotypes from different MRI modalities, it remains unclear which neuroimaging phenotypes yield neural indicators of cognition that best explain the covariation between cognition and mental health, and whether combining these phenotypes enhances the explanation of the cognition-mental health relationship.
Methods:
We address the two research questions using the largest population-level dataset, the UK Biobank cohort (n > 14 000). First, we leveraged 12 cognitive performance scores to derive a general cognition factor or the g-factor. Next, we created a predictive model to evaluate the covariation between the g-factor and 133 mental health indices. To obtain neural indicators of cognition, we built predictive models based on 72 individual neuroimaging phenotypes and phenotypes combined within and across three MRI modalities – diffusion-weighted MRI (dwMRI), resting-state functional MRI (rsMRI), and structural MRI (sMRI). Finally, we conducted commonality analyses to quantify how much of the cognition-mental health relationship is captured by each neural indicator. For commonality analyses, we created a series of linear regression models to compute the proportion of variance of the observed g-factor (response variable) explained uniquely or jointly by the g-factor predicted from mental health and MRI (explanatory variables). We quantified the contribution of each neuroimaging phenotype to the relationship between cognition and mental health as a percentage ratio between the variance that MRI shares with mental health in explaining the g-factor, i.e., the common variance, and the variance that mental health explains regardless of MRI, i.e., the total variance (Fig. 1).
Results:
Mental health predicted individual differences in cognition with a medium-size performance (r = 0.3). The predictive performance of neuroimaging phenotypes was strongly related to their ability to capture the link between cognition and mental health. Overall, neural indicators of cognition derived from 72 neuroimaging phenotypes explained 2–26% of the cognition-mental health covariation. The highest proportion of the cognition-mental health relationship captured by neuroimaging phenotypes from dwMRI was via the number of streamlines connecting grey matter regions (19%), from rsMRI via functional connectivity between 55 large-scale networks (26%), and from sMRI via volumetric characteristics of subcortical structures (22%). Combining neuroimaging phenotypes within each MRI modality allowed us to capture 26–32% of the cognition-mental health relationship. Combining information from all 72 neuroimaging phenotypes across three MRI modalities enhanced the explanation to 48% (Fig. 2).
Conclusions:
We present an integrated approach to derive multimodal neural markers of cognition that can be transdiagnostically linked to psychopathology and show that the improvement in the predictive ability of MRI-based neural indicators of cognition extends beyond the prediction of cognition itself, allowing us to capture more covariation between cognition and mental health. By demonstrating that MRI-based neural indicators of cognition capture the same variance that mental health shares with cognition, we support the utility of such indicators for understanding the etiology of psychopathology.
Disorders of the Nervous System:
Psychiatric (eg. Depression, Anxiety, Schizophrenia)
Higher Cognitive Functions:
Higher Cognitive Functions Other 2
Lifespan Development:
Aging
Modeling and Analysis Methods:
Classification and Predictive Modeling 1
Multivariate Approaches
Keywords:
ADULTS
Affective Disorders
Cognition
Computational Neuroscience
Data analysis
Machine Learning
Modeling
MRI
Psychiatric Disorders
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.
2. Pat, N., Wang, Y., Anney, R., Riglin, L., Thapar, A., & Stringaris, A. (2022). Longitudinally stable, brain-based predictive models mediate the relationships between childhood cognition and socio-demographic, psychological and genetic factors. Human Brain Mapping, 43(18), 5520–5542.
3. Suddell, S., Mahedy, L., Skirrow, C., Penton-Voak, I. S., Munafò, M. R., & Wootton, R. E. (2023). Cognitive functioning in anxiety and depression: Results from the ALSPAC cohort. Royal Society Open Science, 10(8), 221161.