English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Machine learning practice is often impacted by confounders. Confounding canbe particularly severe in remote digital health studies where the participantsself-select to enter the study. While many different confounding adjustmentapproaches have been proposed in the literature, most of these methods rely onmodeling assumptions, and it is unclear how robust they are to violations ofthese assumptions. This realization has recently motivated the development ofrestricted permutation methods to quantify the influence of observedconfounders on the predictive performance of a machine learning models andevaluate if confounding adjustment methods are working as expected. In thispaper we show, nonetheless, that restricted permutations can generate biasedestimates of the contribution of the confounders to the predictive performanceof a learner, and we propose an alternative approach to tackle this problem. Byviewing a classification task from a causality perspective, we are able toleverage conditional independence tests between predictions and test set labelsand confounders in order to detect confounding on the predictive performance ofa classifier. We illustrate the application of our causality-based approach todata collected from mHealth study in Parkinson's disease.

Causality-based tests to detect the influence of confounders on mobile health diagnostic applications: a comparison with restricted permutations