Gendered lifestyles as a data-driven spectrum: Associations with brain morphology and microstructure
Presented During:
Saturday, June 28, 2025: 11:30 AM - 12:45 PM
Brisbane Convention & Exhibition Centre
Room:
M3 (Mezzanine Level)
Poster No:
661
Submission Type:
Abstract Submission
Authors:
Olivier Parent1, Maria McGuinness1, Sophia Osborne1, Alice Mukora1, Manuela Costantino2, Daniela Quesada- Rodriguez1, Gabriel Devenyi3, Robert-Paul Juster4, Mahsa Dadar1, Mallar Chakravarty1
Institutions:
1McGill University, Montreal, Quebec, 2University of Chicago, Chicago, Illinois, 3Douglas Mental Health University Institute, Montreal, Quebec, 4University of Montreal, Montreal, Quebec
First Author:
Co-Author(s):
Introduction:
Understanding the influence of gender dimensions on long-term health remains elusive. Beyond gender identity, gender encompasses sociocultural expectations (gender roles), interindividual interactions (gender relations), and systemic interactions with societal structures (institutionalized gender; Tadiri et al., 2021). How each individual embodies the cultural definition of masculine and feminine gender roles varies along a continuous, non-binary spectrum. Yet, this dynamic has only been operationalized in a few studies using self-reported measures (Nielsen et al., 2021). As a result of these limitations, the impact of gendered lifestyles on brain health remains largely unexplored. Here, we developed a framework to estimate a continuous gender spectrum as expressed through lifestyle behaviors in an unbiased, data-driven manner.
Methods:
Our approach consists of using a set of variables related to lifestyle to predict the self-reported sex of participants using a machine learning algorithm, from which classification probabilities can be extracted and used as a continuous index of gendered lifestyles (IGL), akin to biological age paradigms. We used data from the UK Biobank (n=57,612; 29,636 females; 27,976 males). A list of 56 variables was curated to capture different facets of lifestyle. Subjects with >20% of missing data were excluded and the remaining missing values were imputed with XGBoost (without sex or age in the predictor set). We explored different ML algorithms and choices of hyperparameters using an autoML framework to maximize prediction accuracy (Płońska & Płoński, 2021). We then used 5-fold cross-validation to fit CatBoost models and derive out-of-fold predictions for each participant. We investigated relationships between our IGL and brain morphology and microstructure variables (Alfaro-Almagro et al., 2018) using sex-stratified models and gender-by-sex interactions.
·Fig 1
Results:
Our algorithm showed a balanced prediction for males and females with an area under the ROC curve of 0.829 (Fig 1A-B). As expected, most males and females have an IGL corresponding to their sex; however, we observed long tails in the distributions indicating the variance in gendered behaviors (Fig 1C). Important predictors of masculinity included a longer work week, more lifetime sexual partners, and taking naps, while important predictors of femininity included seeking help for mental distress, friendship satisfaction, and sleeplessness (Fig 1D). However, we note the presence of non-linear relationships from the partial dependence plots, demonstrating the flexibility of our algorithm to capture complex associations (Fig 1E). We observed that the IGL became more feminine with age in both males and females (Fig 2A). Age was regressed out of all other models. We then observed widespread associations with brain health, with a more feminine IGL being related to larger total brain, grey matter, and white matter volume (Fig 2B), larger subcortical grey matter structures (Fig 2C), higher microstructural white matter integrity (Fig 2D), and increased cortical thickness in certain regions (Fig 2E). In contrast, a more masculine IGL was related to larger ventricles and larger regional cortical surface area and grey matter volume in isolated regions. Associations reached significance more often in males but no gender-by-sex interactions were significant.
·Fig 2
Conclusions:
Using data-driven methods, we established a framework to assess the gendered lifestyles, social, and socioeconomic factors of individuals on a continuous scale that can be applied retrospectively. This revealed associations with brain morphology and microstructure, uncovering the brain health implications of gender in a specific cultural context. Of note, the UK Biobank dataset is composed of primarily White middle- to late-aged individuals (Allen et al., 2024); as such, our algorithm captured the specific gendered lifestyle dynamics within this population and is not necessarily generalizable to other cultural groups.
Emotion, Motivation and Social Neuroscience:
Social Neuroscience Other 1
Modeling and Analysis Methods:
Multivariate Approaches 2
Keywords:
Machine Learning
Morphometrics
Multivariate
White Matter
Other - Gender
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.
Nielsen, M. W., Stefanick, M. L., Peragine, D., Neilands, T. B., Ioannidis, J. P. A., Pilote, L., Prochaska, J. J., Cullen, M. R., Einstein, G., Klinge, I., LeBlanc, H., Paik, H. Y., & Schiebinger, L. (2021). Gender-related variables for health research. Biology of Sex Differences, 12(1), 23. https://doi.org/10.1186/s13293-021-00366-3
Płońska, A., & Płoński, P. (2021). MLJAR: State-of-the-art Automated Machine Learning Framework for Tabular Data. Version 0.10.3. MLJAR Sp. z o.o. https://github.com/mljar/mljar-supervised
Tadiri, C. P., Raparelli, V., Abrahamowicz, M., Kautzy-Willer, A., Kublickiene, K., Herrero, M.-T., Norris, C. M., & Pilote, L. (2021). Methods for prospectively incorporating gender into health sciences research. Journal of Clinical Epidemiology, 129, 191–197. https://doi.org/10.1016/j.jclinepi.2020.08.018