How fair is your graph? Exploring fairness concerns in neuroimaging studies

Fernanda Ribeiro, Valentina Shumovskaia, Thomas Davies, Ira Ktena
Proceedings of the 7th Machine Learning for Healthcare Conference, PMLR 182:459-478, 2022.

Abstract

Recent work on neuroimaging has demonstrated significant benefits of using population graphs to capture non-imaging information in the prediction of neurodegenerative and neurodevelopmental disorders. These non-imaging attributes may not only contain demographic information about the individuals, e.g. age or sex, but also the acquisition site, as imaging protocols and hardware might significantly differ across sites in large-scale studies. The effect of the latter is particularly prevalent in functional connectomics studies, where it remains unclear how to sufficiently homogenise fMRI signals across the different sites. In addition, recent studies have highlighted the need to investigate potential biases in the classifiers devised using large-scale datasets, which might be imbalanced in terms of one or more sensitive attributes. This can be exacerbated when employing these attributes in a population graph to explicitly introduce inductive biases to the machine learning model and lead to disparate predictive performance across sub-populations. This study scrutinises such a system and aims to uncover potential biases of a semi-supervised classifier that relies on a population graph. We further explore the effect of the graph structure and stratification strategies, as well as methods to mitigate such biases and produce fairer predictions across the population.

Cite this Paper

BibTeX
@InProceedings{pmlr-v182-ribeiro22a, title = {How fair is your graph? Exploring fairness concerns in neuroimaging studies}, author = {Ribeiro, Fernanda and Shumovskaia, Valentina and Davies, Thomas and Ktena, Ira}, booktitle = {Proceedings of the 7th Machine Learning for Healthcare Conference}, pages = {459--478}, year = {2022}, editor = {Lipton, Zachary and Ranganath, Rajesh and Sendak, Mark and Sjoding, Michael and Yeung, Serena}, volume = {182}, series = {Proceedings of Machine Learning Research}, month = {05--06 Aug}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v182/ribeiro22a/ribeiro22a.pdf}, url = {https://proceedings.mlr.press/v182/ribeiro22a.html}, abstract = {Recent work on neuroimaging has demonstrated significant benefits of using population graphs to capture non-imaging information in the prediction of neurodegenerative and neurodevelopmental disorders. These non-imaging attributes may not only contain demographic information about the individuals, e.g. age or sex, but also the acquisition site, as imaging protocols and hardware might significantly differ across sites in large-scale studies. The effect of the latter is particularly prevalent in functional connectomics studies, where it remains unclear how to sufficiently homogenise fMRI signals across the different sites. In addition, recent studies have highlighted the need to investigate potential biases in the classifiers devised using large-scale datasets, which might be imbalanced in terms of one or more sensitive attributes. This can be exacerbated when employing these attributes in a population graph to explicitly introduce inductive biases to the machine learning model and lead to disparate predictive performance across sub-populations. This study scrutinises such a system and aims to uncover potential biases of a semi-supervised classifier that relies on a population graph. We further explore the effect of the graph structure and stratification strategies, as well as methods to mitigate such biases and produce fairer predictions across the population.} }
Endnote
%0 Conference Paper %T How fair is your graph? Exploring fairness concerns in neuroimaging studies %A Fernanda Ribeiro %A Valentina Shumovskaia %A Thomas Davies %A Ira Ktena %B Proceedings of the 7th Machine Learning for Healthcare Conference %C Proceedings of Machine Learning Research %D 2022 %E Zachary Lipton %E Rajesh Ranganath %E Mark Sendak %E Michael Sjoding %E Serena Yeung %F pmlr-v182-ribeiro22a %I PMLR %P 459--478 %U https://proceedings.mlr.press/v182/ribeiro22a.html %V 182 %X Recent work on neuroimaging has demonstrated significant benefits of using population graphs to capture non-imaging information in the prediction of neurodegenerative and neurodevelopmental disorders. These non-imaging attributes may not only contain demographic information about the individuals, e.g. age or sex, but also the acquisition site, as imaging protocols and hardware might significantly differ across sites in large-scale studies. The effect of the latter is particularly prevalent in functional connectomics studies, where it remains unclear how to sufficiently homogenise fMRI signals across the different sites. In addition, recent studies have highlighted the need to investigate potential biases in the classifiers devised using large-scale datasets, which might be imbalanced in terms of one or more sensitive attributes. This can be exacerbated when employing these attributes in a population graph to explicitly introduce inductive biases to the machine learning model and lead to disparate predictive performance across sub-populations. This study scrutinises such a system and aims to uncover potential biases of a semi-supervised classifier that relies on a population graph. We further explore the effect of the graph structure and stratification strategies, as well as methods to mitigate such biases and produce fairer predictions across the population.
APA
Ribeiro, F., Shumovskaia, V., Davies, T. & Ktena, I.. (2022). How fair is your graph? Exploring fairness concerns in neuroimaging studies. Proceedings of the 7th Machine Learning for Healthcare Conference, in Proceedings of Machine Learning Research 182:459-478 Available from https://proceedings.mlr.press/v182/ribeiro22a.html.

Related Material