{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T13:59:34Z","timestamp":1791381574633,"version":"4.3.3"},"reference-count":38,"publisher":"Wiley","issue":"8","license":[{"start":{"date-parts":[[2025,8,13]],"date-time":"2025-08-13T00:00:00Z","timestamp":1755043200000},"content-version":"vor","delay-in-days":12,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"},{"start":{"date-parts":[[2025,8,1]],"date-time":"2025-08-01T00:00:00Z","timestamp":1754006400000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/501100002858","name":"China Postdoctoral Science Foundation","doi-asserted-by":"publisher","award":["2022M713218"],"award-info":[{"award-number":["2022M713218"]}],"id":[{"id":"10.13039\/501100002858","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["U22A20352"],"award-info":[{"award-number":["U22A20352"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003819","name":"Natural Science Foundation of Hubei Province","doi-asserted-by":"publisher","award":["2024AFB284"],"award-info":[{"award-number":["2024AFB284"]}],"id":[{"id":"10.13039\/501100003819","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2025,8]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>The existing deep learning (DL) segmentation methods of COVID\u201019 lesions in chest CT scans require tedious manual labels for training. This is labor\u2010intensive and time\u2010consuming for radiologists, and also limits the application of DL methods in clinical practice.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>To develop an unsupervised DL segmentation method that does not need the participation of radiologists to provide manual labels for training.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      Two hundred chest 3D CT scans of 48 COVID\u201019 patients (including 48 baseline CT scans and 152 follow\u2010up CT scans) were used as training set and another 65 chest 3D CT scans were used as test set. A novel self\u2010learning segmentation method was proposed to train a DL segmentation model based on the intrinsic physiology information of the patients instead of the external professional knowledge of the radiologists. The proposed method comprised two modules: a self\u2010annotation module and a training module. In self\u2010annotation module, the progression of COVID\u201019 pneumonia and the change of lung volumes between a pair of baseline CT and follow\u2010up CT were combined to automatically annotate the abnormal lung region (referred as\n                      <jats:italic>self\u2010label<\/jats:italic>\n                      ) in the follow\u2010up CT. Then in training module, the follow\u2010up CT scans and the corresponding self\u2010labels were used to train a DL segmentation model (referred as\n                      <jats:italic>Model\u2010S<\/jats:italic>\n                      ). The COVID\u201019 lesions segmented by Model\u2010S in the test set were quantitatively compared with the ground truth. To further evaluate the proposed method, the manual labels of 152 follow\u2010up CT scans were delineated by radiologists and compared with the corresponding self\u2010labels. Then another DL segmentation model (referred as\n                      <jats:italic>Model\u2010M<\/jats:italic>\n                      ) was trained based on these manual labels. The segmentation results between Model\u2010S and Model\u2010M in the test set were also compared.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      As for evaluating the self\u2010annotation results, the Dice similarity coefficient (DSC) between self\u2010labels and manual labels in follow\u2010up CT scans was 93.0%\u00a0\u00b1\u00a010.5%. The lesion volumes of self\u2010labels and manual labels had a high Spearman correlation (\n                      <jats:italic>r<\/jats:italic>\n                      \u00a0=\u00a00.99;\n                      <jats:italic>p<\/jats:italic>\n                      \u00a0&lt;\u00a00.001). As for evaluating the self\u2010learning segmentation results, the DSC between the lesions segmented by Model\u2010S and ground truth, and between the lesions segmented by Model\u2010M and ground truth were 82.1%\u00a0\u00b1\u00a06.0% and 83.4%\u00a0\u00b1\u00a05.2%, respectively. The lesion volumes segmented by Model\u2010S and the ground truth had a high Spearman correlation (\n                      <jats:italic>r<\/jats:italic>\n                      \u00a0=\u00a00.99;\n                      <jats:italic>p<\/jats:italic>\n                      \u00a0&lt;\u00a00.001). Besides, the DSC between the lesions segmented by Model\u2010S and segmented by Model\u2010M was 87.5%\u00a0\u00b1\u00a04.0%. The lesion volumes segmented by Model\u2010S and Model\u2010M also had a high Spearman correlation (\n                      <jats:italic>r<\/jats:italic>\n                      \u00a0=\u00a00.98;\n                      <jats:italic>p<\/jats:italic>\n                      \u00a0&lt;\u00a00.001).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>The proposed unsupervised segmentation method can automatically segment COVID\u201019 lesions accurately in chest 3D CT scans without requiring radiologists to provide any manual labels for training in the establishment process of the method.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.18049","type":"journal-article","created":{"date-parts":[[2025,8,14]],"date-time":"2025-08-14T04:16:21Z","timestamp":1755144981000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Physiology based unsupervised learning for segmentation of COVID\u201019 lesions in chest 3D CT scans"],"prefix":"10.1002","volume":"52","author":[{"given":"Cheng","family":"Wang","sequence":"first","affiliation":[{"name":"School of Physics and Optoelectronic Engineering Yangtze University  Jingzhou China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhiqiang","family":"Cai","sequence":"additional","affiliation":[{"name":"The First Affiliated Hospital of Yangtze University  Jingzhou China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yao","family":"Zhang","sequence":"additional","affiliation":[{"name":"The First Affiliated Hospital of Yangtze University  Jingzhou China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shuqing","family":"Yang","sequence":"additional","affiliation":[{"name":"The First Affiliated Hospital of Yangtze University  Jingzhou China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2025,8,13]]},"reference":[{"key":"e_1_2_9_2_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2020200642"},{"key":"e_1_2_9_3_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00330-020-06865-y"},{"key":"e_1_2_9_4_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-020-17971-2"},{"key":"e_1_2_9_5_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14609"},{"key":"e_1_2_9_6_1","doi-asserted-by":"publisher","DOI":"10.7150\/thno.45985"},{"key":"e_1_2_9_7_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2020.04.045"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.1109\/RBME.2020.2987975"},{"key":"e_1_2_9_9_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00330-020-07013-2"},{"key":"e_1_2_9_10_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00330-020-07156-2"},{"key":"e_1_2_9_11_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2020.107747"},{"key":"e_1_2_9_12_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102605"},{"key":"e_1_2_9_13_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00330-020-07044-9"},{"key":"e_1_2_9_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cmpb.2021.106004"},{"key":"e_1_2_9_15_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2020.2996645"},{"key":"e_1_2_9_16_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2020.3001810"},{"key":"e_1_2_9_17_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-021-00307-0"},{"key":"e_1_2_9_18_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2021.3066161"},{"key":"e_1_2_9_19_1","doi-asserted-by":"publisher","DOI":"10.1109\/42.816070"},{"key":"e_1_2_9_20_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.597308"},{"key":"e_1_2_9_21_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2005.03.023"},{"key":"e_1_2_9_22_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2009.2035616"},{"key":"e_1_2_9_23_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.mri.2012.05.001"},{"key":"e_1_2_9_24_1","doi-asserted-by":"publisher","DOI":"10.1088\/0031-9155\/51\/4\/002"},{"key":"e_1_2_9_25_1","doi-asserted-by":"publisher","DOI":"10.1023\/A:1011444826908"},{"key":"e_1_2_9_26_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13346"},{"key":"e_1_2_9_27_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00259-021-05435-8"},{"key":"e_1_2_9_28_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2020201433"},{"key":"e_1_2_9_29_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.17562"},{"key":"e_1_2_9_30_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2020200463"},{"key":"e_1_2_9_31_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2020200370"},{"key":"e_1_2_9_32_1","doi-asserted-by":"crossref","unstructured":"KimST GoliL PaschaliM et\u00a0al.Longitudinal quantitative assessment of COVID\u201019 infection progression from chest CTs.InProceedings of international conference on medical image computing and computer\u2010assisted intervention Strasbourg France 2021. pp.273\u2010282.","DOI":"10.1007\/978-3-030-87234-2_26"},{"key":"e_1_2_9_33_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2011.2158349"},{"key":"e_1_2_9_34_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.2828378"},{"key":"e_1_2_9_35_1","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2007.909412"},{"key":"e_1_2_9_36_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2024.103385"},{"key":"e_1_2_9_37_1","doi-asserted-by":"publisher","DOI":"10.5114\/pjr.2022.119027"},{"key":"e_1_2_9_38_1","first-page":"174","article-title":"Segmentation of COVID\u201019 pneumonia lesions: a deep learning approach","volume":"34","author":"Ghomi Z","year":"2020","journal-title":"Med J Islam Repub Iran"},{"key":"e_1_2_9_39_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00330-022-09303-3"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.18049","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.18049","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.18049","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T13:38:43Z","timestamp":1791380323000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.18049"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,8]]},"references-count":38,"journal-issue":{"issue":"8","published-print":{"date-parts":[[2025,8]]}},"alternative-id":["10.1002\/mp.18049"],"URL":"https:\/\/doi.org\/10.1002\/mp.18049","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,8]]},"assertion":[{"value":"2024-10-31","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-07-12","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-08-13","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e18049"}}