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

Automated disease detection in neuroimaging holds promise to improve thediagnostic ability of radiologists, but routinely collected clinical datafrequently contains technical and demographic confounding factors that causedata to both differ between sites and be systematically associated with thedisease of interest, thus negatively affecting the robustness of diagnosticmodels. There is a critical need for diagnostic deep learning models that cantrain on such imbalanced datasets without being influenced by these confounds.In this work, we introduce a novel deep learning architecture, MUCRAN(Multi-Confound Regression Adversarial Network), to train a deep learning modelon clinical brain MRI while regressing demographic and technical confoundingfactors. We trained MUCRAN using 17,076 clinical T1 Axial brain MRIs collectedfrom Massachusetts General Hospital before 2019 and demonstrated that MUCRANcould successfully regress major confounding factors in the vast clinical data.We also applied a method for quantifying uncertainty across an ensemble ofthese models to automatically exclude out-of-distribution data in the ADdetection. By combining MUCRAN and the uncertainty quantification method, weshowed consistent and significant increases in the AD detection accuracy fornewly collected MGH data (post-2019) and for data from other hospitals. MUCRANoffers a generalizable approach for heterogenous clinical data fordeep-learning-based automatic disease detection.

Adversarial confound regression and uncertainty measurements to classify heterogeneous clinical MRI in Mass General Brigham