English
 
User Manual Privacy Policy Disclaimer Contact us
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
  Domain-specific cues improve robustness of deep learning-based segmentation of CT volumes

Kloenne, M., Niehaus, S., Lampe, L., Merola, A., Reinelt, J., Roeder, I., et al. (2020). Domain-specific cues improve robustness of deep learning-based segmentation of CT volumes. Scientific Reports, 10(1): 10712. doi:10.1038/s41598-020-67544-y.

Item is

Basic

show hide
Item Permalink: http://hdl.handle.net/21.11116/0000-0006-D55D-2 Version Permalink: http://hdl.handle.net/21.11116/0000-0006-D55E-1
Genre: Journal Article

Files

show Files
hide Files
:
Kloenne_2020.pdf (Publisher version), 2MB
Name:
Kloenne_2020.pdf
Description:
-
Visibility:
Public
MIME-Type / Checksum:
application/pdf / [MD5]
Technical Metadata:
Copyright Date:
-
Copyright Info:
-
License:
-

Locators

show

Creators

show
hide
 Creators:
Kloenne, Marie1, 2, Author
Niehaus, Sebastian1, 3, Author
Lampe, Leonie1, Author
Merola, Alberto1, Author
Reinelt, Janis1, Author
Roeder, Ingo3, 4, Author
Scherf, Nico3, 5, Author              
Affiliations:
1AICURA medical GmbH, Berlin, Germany, ou_persistent22              
2Faculty of Technology, University of Bielefeld, Germany, ou_persistent22              
3Institute for Medical Informatics and Biometry, University Hospital Carl Gustav Carus, Dresden, Germany, ou_persistent22              
4National Center of Tumor Diseases (NCT), Dresden, Germany, ou_persistent22              
5Department Neurophysics (Weiskopf), MPI for Human Cognitive and Brain Sciences, Max Planck Society, ou_2205649              

Content

show
hide
Free keywords: Image processing; Machine learning; Medical imaging; Three-dimensional imaging; Tomography
 Abstract: Machine learning has considerably improved medical image analysis in the past years. Although data-driven approaches are intrinsically adaptive and thus, generic, they often do not perform the same way on data from different imaging modalities. In particular computed tomography (CT) data poses many challenges to medical image segmentation based on convolutional neural networks (CNNs), mostly due to the broad dynamic range of intensities and the varying number of recorded slices of CT volumes. In this paper, we address these issues with a framework that adds domain-specific data preprocessing and augmentation to state-of-the-art CNN architectures. Our major focus is to stabilise the prediction performance over samples as a mandatory requirement for use in automated and semi-automated workflows in the clinical environment. To validate the architecture-independent effects of our approach we compare a neural architecture based on dilated convolutions for parallel multi-scale processing (a modified Mixed-Scale Dense Network: MS-D Net) to traditional scaling operations (a modified U-Net). Finally, we show that an ensemble model combines the strengths across different individual methods. Our framework is simple to implement into existing deep learning pipelines for CT analysis. It performs well on a range of tasks such as liver and kidney segmentation, without significant differences in prediction performance on strongly differing volume sizes and varying slice thickness. Thus our framework is an essential step towards performing robust segmentation of unknown real-world samples.

Details

show
hide
Language(s): eng - English
 Dates: 2020-02-052020-06-042020-07-01
 Publication Status: Published online
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: DOI: 10.1038/s41598-020-67544-y
PMID: 32612129
PMC: PMC7329868
 Degree: -

Event

show

Legal Case

show

Project information

show

Source 1

show
hide
Title: Scientific Reports
  Abbreviation : Sci. Rep.
Source Genre: Journal
 Creator(s):
Affiliations:
Publ. Info: London, UK : Nature Publishing Group
Pages: - Volume / Issue: 10 (1) Sequence Number: 10712 Start / End Page: - Identifier: ISSN: 2045-2322
CoNE: https://pure.mpg.de/cone/journals/resource/2045-2322