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

Semantic segmentation neural networks require pixel-level annotations inlarge quantities to achieve a good performance. In the medical domain, suchannotations are expensive, because they are time-consuming and require expertknowledge. Active learning optimizes the annotation effort by devisingstrategies to select cases for labeling that are most informative to the model.In this work, we propose an uncertainty slice sampling (USS) strategy forsemantic segmentation of 3D medical volumes that selects 2D image slices forannotation and compare it with various other strategies. We demonstrate theefficiency of USS on a CT liver segmentation task using multi-site data. Afterfive iterations, the training data resulting from USS consisted of 2410 slices(4% of all slices in the data pool) compared to 8121 (13%), 8641 (14%), and3730 (6%) for uncertainty volume (UVS), random volume (RVS), and random slice(RSS) sampling, respectively. Despite being trained on the smallest amount ofdata, the model based on the USS strategy evaluated on 234 test volumessignificantly outperformed models trained according to other strategies andachieved a mean Dice index of 0.964, a relative volume error of 4.2%, a meansurface distance of 1.35 mm, and a Hausdorff distance of 23.4 mm. This was onlyslightly inferior to 0.967, 3.8%, 1.18 mm, and 22.9 mm achieved by a modeltrained on all available data, but the robustness analysis using the 5thpercentile of Dice and the 95th percentile of the remaining metricsdemonstrated that USS resulted not only in the most robust model compared toother sampling schemes, but also outperformed the model trained on all dataaccording to Dice (0.946 vs. 0.945) and mean surface distance (1.92 mm vs. 2.03mm).

Robust Segmentation Models using an Uncertainty Slice Sampling Based Annotation Workflow