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/

Zendy Plus

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

Zendy Tools

Zendy Open

Probability calibration for deep models is highly desirable insafety-critical applications such as medical imaging. It makes outputprobabilities of deep networks interpretable, by aligning predictionprobability with the actual accuracy in test data. In image segmentation,well-calibrated probabilities allow radiologists to identify regions wheremodel-predicted segmentations are unreliable. These unreliable predictionsoften occur to out-of-domain (OOD) images that are caused by imaging artifactsor unseen imaging protocols. Unfortunately, most previous calibration methodsfor image segmentation perform sub-optimally on OOD images. To reduce thecalibration error when confronted with OOD images, we propose a novel post-hoccalibration model. Our model leverages the pixel susceptibility againstperturbations at the local level, and the shape prior information at the globallevel. The model is tested on cardiac MRI segmentation datasets that containunseen imaging artifacts and images from an unseen imaging protocol. Wedemonstrate reduced calibration errors compared with the state-of-the-artcalibration algorithm.

Improved post-hoc probability calibration for out-of-domain MRI segmentation