Accurate sex prediction of cisgender and transgender individuals without brain size bias

Sex differences in brain MRI using deep learning toward fairer healthcare outcomes

This study leverages deep learning to analyze sex differences in brain MRI data, aiming to further advance fairness in medical imaging. We employed 3D T1-weighted Magnetic Resonance images from four diverse datasets: Calgary-Campinas-359, OASIS-3, Alzheimer's Disease Neuroimaging Initiative, and Cambridge Center for Aging and Neuroscience, ensuring a balanced representation of sexes and a broad demographic scope. Our methodology focused on minimal preprocessing to preserve the integrity of brain structures, utilizing a Convolutional Neural Network model for sex classification. The model achieved an accuracy of 87% on the test set without employing total intracranial volume (TIV) adjustment techniques. We observed that while the model exhibited biases at extreme brain sizes, it performed with less bias when the TIV distributions overlapped more. Saliency maps were used to identify brain regions significant in sex differentiation, revealing that certain supratentorial and infratentorial regions were important for predictions. Furthermore, our interdisciplinary team, comprising machine learning specialists and a radiologist, ensured diverse perspectives in validating the results. The detailed investigation of sex differences in brain MRI in this study, highlighted by the sex differences map, offers valuable insights into sex-specific aspects of medical imaging and could aid in developing sex-based bias mitigation strategies, contributing to the future development of fair AI algorithms. Awareness of the brain's differences between sexes enables more equitable AI predictions, promoting fairness in healthcare outcomes. Our code and saliency maps are available at https://github.com/mahsadibaji/sex-differences-brain-dl.

Sociodemographic bias in clinical machine learning models: a scoping review of algorithmic bias instances and mechanisms

BACKGROUND AND OBJECTIVES

Clinical machine learning (ML) technologies can sometimes be biased and their use could exacerbate health disparities. The extent to which bias is present, the groups who most frequently experience bias, and the mechanism through which bias is introduced in clinical ML applications is not well described. The objective of this study was to examine instances of bias in clinical ML models. We identified the sociodemographic subgroups PROGRESS that experienced bias and the reported mechanisms of bias introduction.

METHODS

We searched MEDLINE, EMBASE, PsycINFO, and Web of Science for all studies that evaluated bias on sociodemographic factors within ML algorithms created for the purpose of facilitating clinical care. The scoping review was conducted according to the Joanna Briggs Institute guide and reported using the PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) extension for scoping reviews.

RESULTS

We identified 6448 articles, of which 760 reported on a clinical ML model and 91 (12.0%) completed a bias evaluation and met all inclusion criteria. Most studies evaluated a single sociodemographic factor (n = 56, 61.5%). The most frequently evaluated sociodemographic factor was race (n = 59, 64.8%), followed by sex/gender (n = 41, 45.1%), and age (n = 24, 26.4%), with one study (1.1%) evaluating intersectional factors. Of all studies, 74.7% (n = 68) reported that bias was present, 18.7% (n = 17) reported bias was not present, and 6.6% (n = 6) did not state whether bias was present. When present, 87% of studies reported bias against groups with socioeconomic disadvantage.

CONCLUSION

Most ML algorithms that were evaluated for bias demonstrated bias on sociodemographic factors. Furthermore, most bias evaluations concentrated on race, sex/gender, and age, while other sociodemographic factors and their intersection were infrequently assessed. Given potential health equity implications, bias assessments should be completed for all clinical ML models.

Modeling brain sex in the limbic system as phenotype for female-prevalent mental disorders

BACKGROUND

Sex differences exist in the prevalence and clinical manifestation of several mental disorders, suggesting that sex-specific brain phenotypes may play key roles. Previous research used machine learning models to classify sex from imaging data of the whole brain and studied the association of class probabilities with mental health, potentially overlooking regional specific characteristics.

METHODS

We here investigated if a regionally constrained model of brain volumetric imaging data may provide estimates that are more sensitive to mental health than whole brain-based estimates. Given its known role in emotional processing and mood disorders, we focused on the limbic system. Using two different cohorts of healthy subjects, the Human Connectome Project and the Queensland Twin IMaging, we investigated sex differences and heritability of brain volumes of limbic structures compared to non-limbic structures, and subsequently applied regionally constrained machine learning models trained solely on limbic or non-limbic features. To investigate the biological underpinnings of such models, we assessed the heritability of the obtained sex class probability estimates, and we investigated the association with major depression diagnosis in an independent clinical sample. All analyses were performed both with and without controlling for estimated total intracranial volume (eTIV).

RESULTS

Limbic structures show greater sex differences and are more heritable compared to non-limbic structures in both analyses, with and without eTIV control. Consequently, machine learning models performed well at classifying sex based solely on limbic structures and achieved performance as high as those on non-limbic or whole brain data, despite the much smaller number of features in the limbic system. The resulting class probabilities were heritable, suggesting potentially meaningful underlying biological information. Applied to an independent population with major depressive disorder, we found that depression is associated with male-female class probabilities, with largest effects obtained using the limbic model. This association was significant for models not controlling for eTIV whereas in those controlling for eTIV the associations did not pass significance correction.

CONCLUSIONS

Overall, our results highlight the potential utility of regionally constrained models of brain sex to better understand the link between sex differences in the brain and mental disorders.

Sex classification from functional brain connectivity: Generalization to multiple datasets

Machine learning (ML) approaches are increasingly being applied to neuroimaging data. Studies in neuroscience typically have to rely on a limited set of training data which may impair the generalizability of ML models. However, it is still unclear which kind of training sample is best suited to optimize generalization performance. In the present study, we systematically investigated the generalization performance of sex classification models trained on the parcelwise connectivity profile of either single samples or compound samples of two different sizes. Generalization performance was quantified in terms of mean across-sample classification accuracy and spatial consistency of accurately classifying parcels. Our results indicate that the generalization performance of parcelwise classifiers (pwCs) trained on single dataset samples is dependent on the specific test samples. Certain datasets seem to "match" in the sense that classifiers trained on a sample from one dataset achieved a high accuracy when tested on the respected other one and vice versa. The pwCs trained on the compound samples demonstrated overall highest generalization performance for all test samples, including one derived from a dataset not included in building the training samples. Thus, our results indicate that both a large sample size and a heterogeneous data composition of a training sample have a central role in achieving generalizable results.

Sex and Gender in Population Neuroscience

To understand psychiatric and neurological disorders and the structural and functional properties of the human brain, it is essential to consider the roles of sex and gender. In this chapter, I first define sex and gender and describe studies of sex differences in non-human animals. In humans, I describe the sex differences in behavioral and clinical phenotypes and neuroimaging-derived phenotypes, including whole-brain measures, regional subcortical and cortical measures, and structural and functional connectivity. Although structural whole-brain sex differences are large, regional effects (adjusting for whole-brain volumes) are typically much smaller and often fail to replicate. Nevertheless, while an individual neuroimaging feature may have a small effect size, aggregating them in a "maleness/femaleness" score or machine learning multivariate paradigm may prove to be predictive and informative of sex- and gender-related traits. Finally, I conclude by summarizing emerging investigations of gender norms and gender identity and provide methodological recommendations to incorporate sex and gender in population neuroscience research.

Univariate and multivariate sex differences and similarities in gray matter volume within essential language-processing areas

BACKGROUND

Sex differences in language-related abilities have been reported. It is generally assumed that these differences stem from a different organization of language in the brains of females and males. However, research in this area has been relatively scarce, methodologically heterogeneous and has yielded conflicting results.

METHODS

Univariate and multivariate sex differences and similarities in gray matter volume (GMVOL) within 18 essential language-processing brain areas were assessed in a sex-balanced sample (N = 588) of right-handed young adults. Univariate analyses involved location, spread, and shape comparisons of the females' and males' distributions and were conducted with several robust statistical methods able to quantify the size of sex differences and similarities in a complementary way. Multivariate sex differences and similarities were estimated by the same methods in the continuous scores provided by two distinct multivariate procedures (logistic regression and a multivariate analog of the Wilcoxon-Mann-Whitney test). Additional analyses were addressed to compare the outcomes of these two multivariate analytical strategies and described their structure (that is, the relative contribution of each brain area to the multivariate effects).

RESULTS

When not adjusted for total intracranial volume (TIV) variation, "large" univariate sex differences (males > females) were found in all 18 brain areas considered. In contrast, "small" differences (females > males) in just two of these brain areas were found when controlling for TIV. The two multivariate methods tested provided very similar results. Multivariate sex differences surpassed univariate differences, yielding "large" differences indicative of larger volumes in males when calculated from raw GMVOL estimates. Conversely, when calculated from TIV-adjusted GMVOL, multivariate differences were "medium" and indicative of larger volumes in females. Despite their distinct size and direction, multivariate sex differences in raw and TIV-adjusted GMVOL shared a similar structure and allowed us to identify the components of the SENT_CORE network which more likely contribute to the observed effects.

CONCLUSIONS

Our results confirm and extend previous findings about univariate sex differences in language-processing areas, offering unprecedented evidence at the multivariate level. We also observed that the size and direction of these differences vary quite substantially depending on whether they are estimated from raw or TIV-adjusted GMVOL measurements.