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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Neuroimaging datasets keep growing in size to address increasingly complexmedical questions. However, even the largest datasets today alone are too smallfor training complex models or for finding genome wide associations. A solutionis to grow the sample size by merging data across several datasets. However,bias in datasets complicates this approach and includes additional sources ofvariation in the data instead. In this work, we combine 15 large neuroimagingdatasets to study bias. First, we detect bias by demonstrating that scans canbe correctly assigned to a dataset with 73.3% accuracy. Next, we introducemetrics to quantify the compatibility across datasets and to create embeddingsof neuroimaging sites. Finally, we incorporate the presence of bias for theselection of a training set for predicting autism. For the quantification ofthe dataset bias, we introduce two metrics: the Bhattacharyya distance betweendatasets and the age prediction error. The presented embedding of neuroimagingsites provides an interesting new visualization about the similarity ofdifferent sites. This could be used to guide the merging of data sources, whilelimiting the introduction of unwanted variation. Finally, we demonstrate aclear performance increase when incorporating dataset bias for training setselection in autism prediction. Overall, we believe that the growing amount ofneuroimaging data necessitates to incorporate data-driven methods forquantifying dataset bias in future analyses.

Detect, Quantify, and Incorporate Dataset Bias: A Neuroimaging Analysis on 12,207 Individuals