Dataset Open Access

Robust Chest CT Image Segmentation of COVID-19 Lung Infection based on limited data

Dominik Müller; Iñaki Soto Rey; Frank Kramer


Dublin Core Export

<?xml version='1.0' encoding='utf-8'?>
<oai_dc:dc xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/ http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
  <dc:creator>Dominik Müller</dc:creator>
  <dc:creator>Iñaki Soto Rey</dc:creator>
  <dc:creator>Frank Kramer</dc:creator>
  <dc:date>2021-05-30</dc:date>
  <dc:description>Background: The coronavirus disease 2019 (COVID-19) affects billions of lives around the world and has a significant impact on public healthcare. For quantitative assessment and disease monitoring medical imaging like computed tomography offers great potential as alternative to RT-PCR methods. For this reason, automated image segmentation is highly desired as clinical decision support. However, publicly available COVID-19 imaging data is limited which leads to overfitting of traditional approaches.

Methods: To address this problem, we propose an innovative automated segmentation pipeline for COVID-19 infected regions, which is able to handle small datasets by utilization as variant databases. Our method focuses on on-the-fly generation of unique and random image patches for training by performing several preprocessing methods and exploiting extensive data augmentation. For further reduction of the overfitting risk, we implemented a standard 3D U-Net architecture instead of new or computational complex neural network architectures.

Results: Through a k-fold cross-validation on 20 CT scans as training and validation of COVID-19, we were able to develop a highly accurate as well as robust segmentation model for lungs and COVID-19 infected regions without overfitting on limited data. We performed an in-detail analysis and discussion on the robustness of our pipeline through a sensitivity analysis based on the cross-validation and impact on model generalizability of applied preprocessing techniques. Our method achieved Dice similarity coefficients for COVID-19 infection between predicted and annotated segmentation from radiologists of 0.804 on validation and 0.661 on a separate testing set consisting of 100 patients.

Conclusions: We demonstrated that the proposed method outperforms related approaches, advances the state-of-the-art for COVID-19 segmentation and improves robust medical image analysis based on limited data.

The code and model are available under the following link:
https://github.com/frankkramer-lab/covid19.MIScnn</dc:description>
  <dc:description>Code: https://github.com/frankkramer-lab/covid19.MIScnn</dc:description>
  <dc:identifier>https://zenodo.org/record/4875738</dc:identifier>
  <dc:identifier>10.5281/zenodo.4875738</dc:identifier>
  <dc:identifier>oai:zenodo.org:4875738</dc:identifier>
  <dc:relation>url:https://github.com/frankkramer-lab/covid19.MIScnn</dc:relation>
  <dc:relation>doi:10.5281/zenodo.3902292</dc:relation>
  <dc:relation>url:https://zenodo.org/communities/covid-19</dc:relation>
  <dc:relation>url:https://zenodo.org/communities/zenodo</dc:relation>
  <dc:rights>info:eu-repo/semantics/openAccess</dc:rights>
  <dc:rights>https://creativecommons.org/licenses/by/4.0/legalcode</dc:rights>
  <dc:subject>COVID-19</dc:subject>
  <dc:subject>segmentation</dc:subject>
  <dc:subject>computed tomography</dc:subject>
  <dc:subject>deep learning</dc:subject>
  <dc:subject>artificial intelligence</dc:subject>
  <dc:subject>clinical decision support</dc:subject>
  <dc:subject>medical image analysis</dc:subject>
  <dc:title>Robust Chest CT Image Segmentation of COVID-19 Lung Infection based on limited data</dc:title>
  <dc:type>info:eu-repo/semantics/other</dc:type>
  <dc:type>dataset</dc:type>
</oai_dc:dc>
4,405
810
views
downloads
All versions This version
Views 4,4051,157
Downloads 81045
Data volume 1.0 TB34.2 GB
Unique views 3,712922
Unique downloads 41633

Share

Cite as