TY - JOUR
AU - 
AB - Received December 4, 2021, accepted December 18, 2021, date of publication January 6, 2022, date of current version January 13, 2022. Digital Object Identifier 10.1109/ACCESS.2022.3141021 Robust Segmentation Models Using an Uncertainty Slice Sampling-Based Annotation Workflow 1,2 1 1,3 GRZEGORZ CHLEBUS , ANDREA SCHENK , HORST K. HAHN , 1,2 1,4 BRAM VAN GINNEKEN , AND HANS MEINE Fraunhofer Institute for Digital Medicine MEVIS, 28359 Bremen, Germany Diagnostic Image Analysis Group, Department of Medical Imaging, Radboud University Medical Center, 6525 Nijmegen, The Netherlands Department of Computer Science and Electrical Engineering, Jacobs University, 28759 Bremen, Germany Medical Image Computing Group, University of Bremen, 28359 Bremen, Germany Corresponding author: Grzegorz Chlebus (grzegorz.chlebus@mevis.fraunhofer.de) This work was supported by the Fraunhofer-Gesellschaft. ABSTRACT Semantic segmentation neural networks require pixel-level annotations in large quantities to achieve a good performance. In the medical domain, such annotations are expensive because they are time-consuming and require expert knowledge. Active learning optimizes the annotation effort by devising strategies to select cases for labeling that are the most informative to the model. In this work, we propose an uncertainty slice sampling (USS) strategy for the semantic segmentation of 3D medical volumes that selects 2D image slices for annotation and we compare 
TI - Robust Segmentation Models Using an Uncertainty Slice Sampling-Based Annotation Workflow
JF - IEEE Access
DO - 10.1109/access.2022.3141021
DA - 2022-01-01
UR - https://www.deepdyve.com/lp/unpaywall/robust-segmentation-models-using-an-uncertainty-slice-sampling-based-D94ciiU60s
DP - DeepDyve
ER -