Dataset Open Access

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

Dominik Müller; Iñaki Soto Rey; Frank Kramer


DataCite XML Export

<?xml version='1.0' encoding='utf-8'?>
<resource xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://datacite.org/schema/kernel-4" xsi:schemaLocation="http://datacite.org/schema/kernel-4 http://schema.datacite.org/meta/kernel-4.1/metadata.xsd">
  <identifier identifierType="DOI">10.5281/zenodo.4875738</identifier>
  <creators>
    <creator>
      <creatorName>Dominik Müller</creatorName>
      <nameIdentifier nameIdentifierScheme="ORCID" schemeURI="http://orcid.org/">0000-0003-0838-9885</nameIdentifier>
      <affiliation>IT-Infrastructure for Translational Medical Research, Faculty of Applied Computer Science, Faculty of Medicine, University of Augsburg, Germany</affiliation>
    </creator>
    <creator>
      <creatorName>Iñaki Soto Rey</creatorName>
      <nameIdentifier nameIdentifierScheme="ORCID" schemeURI="http://orcid.org/">0000-0003-3061-5818</nameIdentifier>
      <affiliation>IT-Infrastructure for Translational Medical Research, Faculty of Applied Computer Science, Faculty of Medicine, University of Augsburg, Germany</affiliation>
    </creator>
    <creator>
      <creatorName>Frank Kramer</creatorName>
      <nameIdentifier nameIdentifierScheme="ORCID" schemeURI="http://orcid.org/">0000-0002-2857-7122</nameIdentifier>
      <affiliation>IT-Infrastructure for Translational Medical Research, Faculty of Applied Computer Science, Faculty of Medicine, University of Augsburg, Germany</affiliation>
    </creator>
  </creators>
  <titles>
    <title>Robust Chest CT Image Segmentation of COVID-19 Lung Infection based on limited data</title>
  </titles>
  <publisher>Zenodo</publisher>
  <publicationYear>2021</publicationYear>
  <subjects>
    <subject>COVID-19</subject>
    <subject>segmentation</subject>
    <subject>computed tomography</subject>
    <subject>deep learning</subject>
    <subject>artificial intelligence</subject>
    <subject>clinical decision support</subject>
    <subject>medical image analysis</subject>
  </subjects>
  <dates>
    <date dateType="Issued">2021-05-30</date>
  </dates>
  <resourceType resourceTypeGeneral="Dataset"/>
  <alternateIdentifiers>
    <alternateIdentifier alternateIdentifierType="url">https://zenodo.org/record/4875738</alternateIdentifier>
  </alternateIdentifiers>
  <relatedIdentifiers>
    <relatedIdentifier relatedIdentifierType="URL" relationType="IsSupplementTo" resourceTypeGeneral="Software">https://github.com/frankkramer-lab/covid19.MIScnn</relatedIdentifier>
    <relatedIdentifier relatedIdentifierType="DOI" relationType="IsVersionOf">10.5281/zenodo.3902292</relatedIdentifier>
    <relatedIdentifier relatedIdentifierType="URL" relationType="IsPartOf">https://zenodo.org/communities/covid-19</relatedIdentifier>
    <relatedIdentifier relatedIdentifierType="URL" relationType="IsPartOf">https://zenodo.org/communities/zenodo</relatedIdentifier>
  </relatedIdentifiers>
  <version>2.0</version>
  <rightsList>
    <rights rightsURI="https://creativecommons.org/licenses/by/4.0/legalcode">Creative Commons Attribution 4.0 International</rights>
    <rights rightsURI="info:eu-repo/semantics/openAccess">Open Access</rights>
  </rightsList>
  <descriptions>
    <description descriptionType="Abstract">&lt;p&gt;&lt;strong&gt;Background:&lt;/strong&gt; 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.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Methods:&lt;/strong&gt; 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.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Results:&lt;/strong&gt; 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.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Conclusions:&lt;/strong&gt; 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.&lt;/p&gt;

&lt;p&gt;The code and model are available under the following link:&lt;br&gt;
https://github.com/frankkramer-lab/covid19.MIScnn&lt;/p&gt;</description>
    <description descriptionType="Other">Code: https://github.com/frankkramer-lab/covid19.MIScnn</description>
  </descriptions>
</resource>
3,684
724
views
downloads
All versions This version
Views 3,684687
Downloads 72427
Data volume 927.9 GB19.4 GB
Unique views 3,166565
Unique downloads 35321

Share

Cite as