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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Performance metrics for medical image segmentation models are used to measurethe agreement between the reference annotation and the predicted segmentation.Usually, overlap metrics, such as the Dice, are used as a metric to evaluatethe performance of these models in order for results to be comparable. However,there is a mismatch between the distributions of cases and difficulty level ofsegmentation tasks in public data sets compared to clinical practice. Commonmetrics fail to measure the impact of this mismatch, especially for clinicaldata sets that include low signal pathologies, a difficult segmentation task,and uncertain, small, or empty reference annotations. This limitation mayresult in ineffective research of machine learning practitioners in designingand optimizing models. Dimensions of evaluating clinical value includeconsideration of the uncertainty of reference annotations, independence fromreference annotation volume size, and evaluation of classification of emptyreference annotations. We study how uncertain, small, and empty referenceannotations influence the value of metrics for medical image segmentation on anin-house data set regardless of the model. We examine metrics behavior on thepredictions of a standard deep learning framework in order to identify metricswith clinical value. We compare to a public benchmark data set (BraTS 2019)with a high-signal pathology and certain, larger, and no empty referenceannotations. We may show machine learning practitioners, how uncertain, small,or empty reference annotations require a rethinking of the evaluation andoptimizing procedures. The evaluation code was released to encourage furtheranalysis of this topic.https://github.com/SophieOstmeier/UncertainSmallEmpty.git

USE-Evaluator: Performance Metrics for Medical Image Segmentation Models with Uncertain, Small or Empty Reference Annotations