{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,19]],"date-time":"2026-10-19T16:01:15Z","timestamp":1792425675616,"version":"4.5.0"},"reference-count":50,"publisher":"Wiley","issue":"11","license":[{"start":{"date-parts":[[2023,9,22]],"date-time":"2023-09-22T00:00:00Z","timestamp":1695340800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2023,11]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Purpose<\/jats:title><jats:p>Deep learning\u2010based networks have become increasingly popular in the field of medical image segmentation. The purpose of this research was to develop and optimize a new architecture for automatic segmentation of the prostate gland and normal organs in the pelvic, thoracic, and upper gastro\u2010intestinal (GI) regions.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>We developed an architecture which combines a <jats:styled-content>s<\/jats:styled-content>hifted\u2010<jats:styled-content>win<\/jats:styled-content>dow (Swin) transformer with a convolutional U\u2010Net. The network includes a parallel encoder, a cross\u2010fusion block, and a CNN\u2010based decoder to extract local and global information and merge related features on the same scale. A skip connection is applied between the cross\u2010fusion block and decoder to integrate low\u2010level semantic features. Attention gates (AGs) are integrated within the CNN to suppress features in image background regions. Our network is termed \u201cSwinAttUNet.\u201d We optimized the architecture for automatic image segmentation. Training datasets consisted of planning\u2010CT datasets from 300 prostate cancer patients from an institutional database and 100 CT datasets from a publicly available dataset (CT\u2010ORG). Images were linearly interpolated and resampled to a spatial resolution of (1.0\u00a0\u00d7\u00a01.0\u00d7\u00a01.5) mm<jats:sup>3<\/jats:sup>. A volume patch (192\u00a0\u00d7\u00a0192\u00a0\u00d7\u00a096) was used for training and inference, and the dataset was split into training (75%), validation (10%), and test (15%) cohorts. Data augmentation transforms were applied consisting of random flip, rotation, and intensity scaling. The loss function comprised Dice and cross\u2010entropy equally weighted and summed. We evaluated Dice coefficients (DSC), 95th percentile Hausdorff Distances (HD95), and Average Surface Distances (ASD) between results of our network and ground truth data.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>SwinAttUNet, DSC values were 86.54\u00a0\u00b1\u00a01.21, 94.15\u00a0\u00b1\u00a01.17, and 87.15\u00a0\u00b1\u00a01.68% and HD95 values were 5.06\u00a0\u00b1\u00a01.42, 3.16\u00a0\u00b1\u00a00.93, and 5.54\u00a0\u00b1\u00a01.63\u00a0mm for the prostate, bladder, and rectum, respectively. Respective ASD values were 1.45\u00a0\u00b1\u00a00.57, 0.82\u00a0\u00b1\u00a00.12, and 1.42\u00a0\u00b1\u00a00.38\u00a0mm. For the lung, liver, kidneys and pelvic bones, respective DSC values were: 97.90\u00a0\u00b1\u00a00.80, 96.16\u00a0\u00b1\u00a00.76, 93.74\u00a0\u00b1\u00a02.25, and 89.31\u00a0\u00b1\u00a03.87%. Respective HD95 values were: 5.13\u00a0\u00b1\u00a04.11, 2.73\u00a0\u00b1\u00a01.19, 2.29\u00a0\u00b1\u00a01.47, and 5.31\u00a0\u00b1\u00a01.25\u00a0mm. Respective ASD values were: 1.88\u00a0\u00b1\u00a01.45, 1.78\u00a0\u00b1\u00a01.21, 0.71\u00a0\u00b1\u00a00.43, and 1.21\u00a0\u00b1\u00a01.11\u00a0mm. Our network outperformed several existing deep learning approaches using only attention\u2010based convolutional or Transformer\u2010based feature strategies, as detailed in the results section.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusions<\/jats:title><jats:p>We have demonstrated that our new architecture combining Transformer\u2010 and convolution\u2010based features is able to better learn the local and global context for automatic segmentation of multi\u2010organ, CT\u2010based anatomy.<\/jats:p><\/jats:sec>","DOI":"10.1002\/mp.16750","type":"journal-article","created":{"date-parts":[[2023,9,22]],"date-time":"2023-09-22T19:08:33Z","timestamp":1695409713000},"page":"6990-7002","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":16,"title":["A new architecture combining convolutional and transformer\u2010based networks for automatic 3D multi\u2010organ segmentation on CT images"],"prefix":"10.1002","volume":"50","author":[{"given":"Chengyin","family":"Li","sequence":"first","affiliation":[{"name":"College of Engineering \u2010 Dept. of Computer Science Wayne State University  Detroit Michigan USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hassan","family":"Bagher\u2010Ebadian","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Henry Ford Cancer Institute  Detroit Michigan USA"},{"name":"Department of Radiology Michigan State University  East Lansing Michigan USA"},{"name":"Department of Osteopathic Medicine Michigan State University East Lansing Michigan USA"},{"name":"Department of Physics Oakland University  Rochester Michigan USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rafi Ibn","family":"Sultan","sequence":"additional","affiliation":[{"name":"College of Engineering \u2010 Dept. of Computer Science Wayne State University  Detroit Michigan USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mohamed","family":"Elshaikh","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Henry Ford Cancer Institute  Detroit Michigan USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Benjamin","family":"Movsas","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Henry Ford Cancer Institute  Detroit Michigan USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dongxiao","family":"Zhu","sequence":"additional","affiliation":[{"name":"College of Engineering \u2010 Dept. of Computer Science Wayne State University  Detroit Michigan USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Indrin J.","family":"Chetty","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Henry Ford Cancer Institute  Detroit Michigan USA"},{"name":"Department of Radiation Oncology Cedars Sinai Medical Center  Los Angeles CA USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2023,9,22]]},"reference":[{"key":"e_1_2_8_2_1","doi-asserted-by":"publisher","DOI":"10.3322\/caac.21552"},{"key":"e_1_2_8_3_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2015.07.126"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-021-00416-5"},{"key":"e_1_2_8_5_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.16188"},{"key":"e_1_2_8_6_1","doi-asserted-by":"crossref","unstructured":"El NaqaI DasS The role of machine and deep learning in modern medical physics.2020.","DOI":"10.1002\/mp.14088"},{"key":"e_1_2_8_7_1","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2018.00110"},{"key":"e_1_2_8_8_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.15982"},{"key":"e_1_2_8_9_1","doi-asserted-by":"publisher","DOI":"10.1053\/j.semradonc.2003.10.003"},{"key":"e_1_2_8_10_1","doi-asserted-by":"publisher","DOI":"10.1117\/1.JMI.7.1.012705"},{"key":"e_1_2_8_11_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14755"},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2022.854349"},{"key":"e_1_2_8_13_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.12890"},{"key":"e_1_2_8_14_1","doi-asserted-by":"crossref","unstructured":"GlockerB PaulyO KonukogluE CriminisiA.Joint classification\u2010regression forests for spatially structured multi\u2010object segmentation.Paper presented at: Computer Vision\u2013ECCV 2012: 12th European Conference on Computer Vision Florence Italy October 7\u201313 2012 Proceedings Part IV 122012.","DOI":"10.1007\/978-3-642-33765-9_62"},{"key":"e_1_2_8_15_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cviu.2014.04.015"},{"key":"e_1_2_8_16_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2016.2519264"},{"key":"e_1_2_8_17_1","unstructured":"RonnebergerO FischerP BroxT.U\u2010net: convolutional networks for biomedical image segmentation.Paper presented at: Medical Image Computing and Computer\u2010Assisted Intervention\u2013MICCAI 2015: 18th International Conference Munich Germany October 5\u20139 2015 Proceedings Part III 182015."},{"key":"e_1_2_8_18_1","doi-asserted-by":"crossref","unstructured":"MilletariF NavabN AhmadiS\u2010A.V\u2010net: fully convolutional neural networks for volumetric medical image segmentation.Paper presented at: 2016 fourth international conference on 3D vision (3DV) Stanford CA USA 2016:565\u2010571.","DOI":"10.1109\/3DV.2016.79"},{"key":"e_1_2_8_19_1","doi-asserted-by":"publisher","DOI":"10.1088\/2057-1976\/aad100"},{"key":"e_1_2_8_20_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-020-01008-z"},{"key":"e_1_2_8_21_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.semradonc.2019.02.001"},{"key":"e_1_2_8_22_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14140"},{"key":"e_1_2_8_23_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2019.2948320"},{"key":"e_1_2_8_24_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-022-21206-3"},{"key":"e_1_2_8_25_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-021-20966-2"},{"key":"e_1_2_8_26_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00066-020-01607-x"},{"key":"e_1_2_8_27_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2019.2907072"},{"key":"e_1_2_8_28_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/aaf11c"},{"key":"e_1_2_8_29_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ac95f7"},{"key":"e_1_2_8_30_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ab2818"},{"key":"e_1_2_8_31_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2020.3025517"},{"key":"e_1_2_8_32_1","article-title":"Understanding the effective receptive field in deep convolutional neural networks","volume":"29","author":"Luo W","year":"2016","journal-title":"Adv Neural Inf Process Syst"},{"key":"e_1_2_8_33_1","unstructured":"DosovitskiyA BeyerL KolesnikovA et\u00a0al.An image is worth 16\u00d716 words: Transformers for image recognition at scale. arXiv preprint arXiv:201011929.2020."},{"key":"e_1_2_8_34_1","unstructured":"ChenJ LuY YuQ et\u00a0al.Transunet: Transformers make strong encoders for medical image segmentation.arXiv preprint arXiv:210204306.2021."},{"key":"e_1_2_8_35_1","doi-asserted-by":"crossref","unstructured":"CaoH WangY ChenJ et\u00a0al.Swin\u2010unet: unet\u2010like pure transformer for medical image segmentation.Paper presented at: Computer Vision\u2013ECCV 2022 Workshops: Tel Aviv Israel October 23\u201327 2022 Proceedings Part III2023.","DOI":"10.1007\/978-3-031-25066-8_9"},{"key":"e_1_2_8_36_1","doi-asserted-by":"crossref","unstructured":"HatamizadehA TangY NathV et\u00a0al.Unetr: transformers for 3d medical image segmentation.Paper presented at: Proceedings of the IEEE\/CVF winter conference on applications of computer vision 2022.","DOI":"10.1109\/WACV51458.2022.00181"},{"key":"e_1_2_8_37_1","doi-asserted-by":"crossref","unstructured":"TangY YangD LiW et\u00a0al.Self\u2010supervised pre\u2010training of swin transformers for 3d medical image analysis.Paper presented at: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition 2022.","DOI":"10.1109\/CVPR52688.2022.02007"},{"key":"e_1_2_8_38_1","doi-asserted-by":"crossref","unstructured":"ZhangY LiuH HuQ.Transfuse: fusing transformers and cnns for medical image segmentation.Paper presented at: Medical Image Computing and Computer Assisted Intervention\u2013MICCAI 2021: 24th International Conference Strasbourg France September 27\u2013October 1 2021 Proceedings Part I 242021.","DOI":"10.1007\/978-3-030-87193-2_2"},{"key":"e_1_2_8_39_1","doi-asserted-by":"crossref","unstructured":"YooJ KimT LeeS KimSH LeeH KimTH.Enriched CNN\u2010Transformer Feature Aggregation Networks for Super\u2010Resolution.Paper presented at: Proceedings of the IEEE\/CVF Winter Conference on Applications of Computer Vision 2023.","DOI":"10.1109\/WACV56688.2023.00493"},{"key":"e_1_2_8_40_1","first-page":"59","article-title":"Transformers in medical image analysis: a review","volume":"23","author":"He K","year":"2022","journal-title":"Intelligent Medicine"},{"key":"e_1_2_8_41_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2019.03.017"},{"key":"e_1_2_8_42_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41597-020-00715-8"},{"key":"e_1_2_8_43_1","doi-asserted-by":"crossref","unstructured":"LiuZ LinY CaoY et\u00a0al.Swin transformer: hierarchical vision transformer using shifted windows.Paper presented at: Proceedings of the IEEE\/CVF international conference on computer vision 2021.","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"e_1_2_8_44_1","doi-asserted-by":"crossref","unstructured":"HeK ZhangX RenS SunJ.Identity mappings in deep residual networks.Paper presented at: Computer Vision\u2013ECCV 2016: 14th European Conference Amsterdam The Netherlands October 11\u201314 2016 Proceedings Part IV 142016.","DOI":"10.1007\/978-3-319-46493-0_38"},{"key":"e_1_2_8_45_1","unstructured":"OktayO SchlemperJ FolgocLL et\u00a0al.Attention u\u2010net: Learning where to look for the pancreas.arXiv preprint arXiv:180403999.2018."},{"key":"e_1_2_8_46_1","unstructured":"GaoY ZhouM LiuD YanZ ZhangS MetaxasDN.A data\u2010scalable transformer for medical image segmentation: architecture model efficiency and benchmark.arXiv preprint arXiv:220300131.2022."},{"key":"e_1_2_8_47_1","first-page":"1","article-title":"Mann\u2010Whitney U test","author":"McKnight PE","year":"2010","journal-title":"The Corsini encyclopedia of psychology"},{"key":"e_1_2_8_48_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2019.09.028"},{"key":"e_1_2_8_49_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ab63bb"},{"key":"e_1_2_8_50_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2021.102101"},{"key":"e_1_2_8_51_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.artmed.2021.102195"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.16750","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,7]],"date-time":"2023-11-07T17:49:44Z","timestamp":1699379384000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.16750"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,9,22]]},"references-count":50,"journal-issue":{"issue":"11","published-print":{"date-parts":[[2023,11]]}},"alternative-id":["10.1002\/mp.16750"],"URL":"https:\/\/doi.org\/10.1002\/mp.16750","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,9,22]]},"assertion":[{"value":"2023-05-09","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2023-08-13","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2023-09-22","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}