Dataset Information

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

ABSTRACT: The increasing use of machine learning approaches on neuroimaging data comes with the important concern of confounding variables which might lead to biased predictions and in turn spurious conclusions about the relationship between the features and the target. A prominent example is the brain size difference between women and men. This difference in total intracranial volume (TIV) can cause bias when employing machine learning approaches for the investigation of sex differences in brain morphology. A TIV-biased model will not capture qualitative sex differences in brain organization but rather learn to classify an individual's sex based on brain size differences, thus leading to spurious and misleading conclusions, for example when comparing brain morphology between cisgender- and transgender individuals. In this study, TIV bias in sex classification models applied to cis- and transgender individuals was systematically investigated by controlling for TIV either through featurewise confound removal or by matching the training samples for TIV. Our results provide strong evidence that models not biased by TIV can classify the sex of both cis- and transgender individuals with high accuracy, highlighting the importance of appropriate modeling to avoid bias in automated decision making.

SUBMITTER: Wiersch L

PROVIDER: S-EPMC10449927 | biostudies-literature | 2023 Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

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

Wiersch Lisa L Hamdan Sami S Hoffstaedter Felix F Votinov Mikhail M Habel Ute U Clemens Benjamin B Derntl Birgit B Eickhoff Simon B SB Patil Kaustubh R KR Weis Susanne S

Scientific reports 20230824 1

The increasing use of machine learning approaches on neuroimaging data comes with the important concern of confounding variables which might lead to biased predictions and in turn spurious conclusions about the relationship between the features and the target. A prominent example is the brain size difference between women and men. This difference in total intracranial volume (TIV) can cause bias when employing machine learning approaches for the investigation of sex differences in brain morpholo ...[more]

PMID: 37620339

Similar Datasets

Project description:Abstract Background: While there is growing awareness of the need to support the physical and mental wellbeing of transgender people, some may be reluctant to seek help from healthcare professionals. Little is understood about the mechanisms that influence help-seeking behavior in this group. Aims: This study aimed to compare transgender and cisgender participants in their likelihood to seek help for both physical and mental health conditions, and to explore whether this help-seeking behavior is predicted by a range of sociodemographic and psychological variables. Methods: 123 participants living in Ireland (cisgender= 67; transgender= 56) completed a questionnaire which included demographic questions, as well as measures of optimism (LOT-R), self-esteem (RSES), psychological distress (GHQ-12), attitudes towards seeking psychological help (ATSPPH-SF), and attitudes towards seeking help for a physical health problem (Attitudes Towards Seeking Medical Help Scale- Action/Intervention subscale). Associations between predictor variables and mental and physical health seeking were explored using correlation analysis and stepwise regressions. Results: Transgender participants were less likely to seek help for a physical health issue than cisgender participants, but did not differ in mental health help-seeking behaviors. Results suggest that this may be due to differences in optimism, self-esteem and psychological distress. Transgender participants had significantly lower optimism and self-esteem, which were two factors linked to poorer physical health seeking behaviors. Optimism also emerged as a significant predictor in mental health seeking behaviors. Discussion: The lack of a significant difference for mental health help-seeking between the transgender and cisgender participants is encouraging, as it suggests that there is less stigma surrounding mental illness than expected, however findings also contradict previous findings suggesting that physical health is less stigmatized. This could be due to stigma relating to gender-specific healthcare and suggests that healthcare professionals should acknowledge the specific healthcare needs and concerns among transgender individuals.

Dataset Information

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

Publications

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

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets