Dataset Open Access

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

Dominik Müller; Iñaki Soto Rey; Frank Kramer


JSON Export

{
  "files": [
    {
      "links": {
        "self": "https://zenodo.org/api/files/a30f8f5e-8471-4b20-8339-3d928f3e9ac4/evaluation.zip"
      }, 
      "checksum": "md5:74c65427c595986c41d74ed5b1ee9bfc", 
      "bucket": "a30f8f5e-8471-4b20-8339-3d928f3e9ac4", 
      "key": "evaluation.zip", 
      "type": "zip", 
      "size": 662048089
    }, 
    {
      "links": {
        "self": "https://zenodo.org/api/files/a30f8f5e-8471-4b20-8339-3d928f3e9ac4/models.zip"
      }, 
      "checksum": "md5:20ac6f5186d7e98edbd54bc547d50ad6", 
      "bucket": "a30f8f5e-8471-4b20-8339-3d928f3e9ac4", 
      "key": "models.zip", 
      "type": "zip", 
      "size": 2524272040
    }, 
    {
      "links": {
        "self": "https://zenodo.org/api/files/a30f8f5e-8471-4b20-8339-3d928f3e9ac4/predictions.zip"
      }, 
      "checksum": "md5:3fb40f3f9c05f224ec22421a7b9ad450", 
      "bucket": "a30f8f5e-8471-4b20-8339-3d928f3e9ac4", 
      "key": "predictions.zip", 
      "type": "zip", 
      "size": 17639032
    }, 
    {
      "links": {
        "self": "https://zenodo.org/api/files/a30f8f5e-8471-4b20-8339-3d928f3e9ac4/supplementary.zip"
      }, 
      "checksum": "md5:2a67cc8a134b7f89996e1321e4465402", 
      "bucket": "a30f8f5e-8471-4b20-8339-3d928f3e9ac4", 
      "key": "supplementary.zip", 
      "type": "zip", 
      "size": 4260382
    }
  ], 
  "owners": [
    108522
  ], 
  "doi": "10.5281/zenodo.4875738", 
  "stats": {
    "version_unique_downloads": 416.0, 
    "unique_views": 922.0, 
    "views": 1157.0, 
    "version_views": 4405.0, 
    "unique_downloads": 33.0, 
    "version_unique_views": 3712.0, 
    "volume": 34214756931.0, 
    "version_downloads": 810.0, 
    "downloads": 45.0, 
    "version_volume": 1025016594976.0
  }, 
  "links": {
    "doi": "https://doi.org/10.5281/zenodo.4875738", 
    "conceptdoi": "https://doi.org/10.5281/zenodo.3902292", 
    "bucket": "https://zenodo.org/api/files/a30f8f5e-8471-4b20-8339-3d928f3e9ac4", 
    "conceptbadge": "https://zenodo.org/badge/doi/10.5281/zenodo.3902292.svg", 
    "html": "https://zenodo.org/record/4875738", 
    "latest_html": "https://zenodo.org/record/4875738", 
    "badge": "https://zenodo.org/badge/doi/10.5281/zenodo.4875738.svg", 
    "latest": "https://zenodo.org/api/records/4875738"
  }, 
  "conceptdoi": "10.5281/zenodo.3902292", 
  "created": "2021-05-30T16:21:33.952654+00:00", 
  "updated": "2021-05-31T13:48:12.730489+00:00", 
  "conceptrecid": "3902292", 
  "revision": 2, 
  "id": 4875738, 
  "metadata": {
    "access_right_category": "success", 
    "doi": "10.5281/zenodo.4875738", 
    "description": "<p><strong>Background:</strong> 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.</p>\n\n<p><strong>Methods:</strong> 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.</p>\n\n<p><strong>Results:</strong> 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.</p>\n\n<p><strong>Conclusions:</strong> 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.</p>\n\n<p>The code and model are available under the following link:<br>\nhttps://github.com/frankkramer-lab/covid19.MIScnn</p>", 
    "license": {
      "id": "CC-BY-4.0"
    }, 
    "title": "Robust Chest CT Image Segmentation of COVID-19 Lung Infection based on limited data", 
    "notes": "Code: https://github.com/frankkramer-lab/covid19.MIScnn", 
    "relations": {
      "version": [
        {
          "count": 3, 
          "index": 2, 
          "parent": {
            "pid_type": "recid", 
            "pid_value": "3902292"
          }, 
          "is_last": true, 
          "last_child": {
            "pid_type": "recid", 
            "pid_value": "4875738"
          }
        }
      ]
    }, 
    "communities": [
      {
        "id": "covid-19"
      }, 
      {
        "id": "zenodo"
      }
    ], 
    "version": "2.0", 
    "keywords": [
      "COVID-19", 
      "segmentation", 
      "computed tomography", 
      "deep learning", 
      "artificial intelligence", 
      "clinical decision support", 
      "medical image analysis"
    ], 
    "publication_date": "2021-05-30", 
    "creators": [
      {
        "orcid": "0000-0003-0838-9885", 
        "affiliation": "IT-Infrastructure for Translational Medical Research, Faculty of Applied Computer Science, Faculty of Medicine, University of Augsburg, Germany", 
        "name": "Dominik M\u00fcller"
      }, 
      {
        "orcid": "0000-0003-3061-5818", 
        "affiliation": "IT-Infrastructure for Translational Medical Research, Faculty of Applied Computer Science, Faculty of Medicine, University of Augsburg, Germany", 
        "name": "I\u00f1aki Soto Rey"
      }, 
      {
        "orcid": "0000-0002-2857-7122", 
        "affiliation": "IT-Infrastructure for Translational Medical Research, Faculty of Applied Computer Science, Faculty of Medicine, University of Augsburg, Germany", 
        "name": "Frank Kramer"
      }
    ], 
    "access_right": "open", 
    "resource_type": {
      "type": "dataset", 
      "title": "Dataset"
    }, 
    "related_identifiers": [
      {
        "scheme": "url", 
        "identifier": "https://github.com/frankkramer-lab/covid19.MIScnn", 
        "relation": "isSupplementTo", 
        "resource_type": "software"
      }, 
      {
        "scheme": "doi", 
        "identifier": "10.5281/zenodo.3902292", 
        "relation": "isVersionOf"
      }
    ]
  }
}
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