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

Despite their success, Machine Learning (ML) models do not generalizeeffectively to data not originating from the training distribution. To reliablyemploy ML models in real-world healthcare systems and avoid inaccuratepredictions on out-of-distribution (OOD) data, it is crucial to detect OODsamples. Numerous OOD detection approaches have been suggested in other fields- especially in computer vision - but it remains unclear whether the challengeis resolved when dealing with medical tabular data. To answer this pressingneed, we propose an extensive reproducible benchmark to compare differentmethods across a suite of tests including both near and far OODs. Our benchmarkleverages the latest versions of eICU and MIMIC-IV, two public datasetsencompassing tens of thousands of ICU patients in several hospitals. Weconsider a wide array of density-based methods and SOTA post-hoc detectorsacross diverse predictive architectures, including MLP, ResNet, andTransformer. Our findings show that i) the problem appears to be solved forfar-OODs, but remains open for near-OODs; ii) post-hoc methods alone performpoorly, but improve substantially when coupled with distance-based mechanisms;iii) the transformer architecture is far less overconfident compared to MLP andResNet.

Unmasking the Chameleons: A Benchmark for Out-of-Distribution Detection in Medical Tabular Data