{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T05:43:11Z","timestamp":1790833391352,"version":"4.1.0"},"reference-count":52,"publisher":"Wiley","issue":"9","license":[{"start":{"date-parts":[[2019,7,26]],"date-time":"2019-07-26T00:00:00Z","timestamp":1564099200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"funder":[{"DOI":"10.13039\/100007210","name":"Varian Medical Systems","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100007210","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":[[2019,9]]},"abstract":"<jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>The improved soft tissue contrast of magnetic resonance imaging\u00a0(MRI) compared to computed tomography (CT) makes it a useful imaging modality for radiotherapy treatment planning. Even when MR images are acquired for treatment planning, the standard clinical practice currently also requires a CT for dose calculation and x\u2010ray\u2013based patient positioning. This increases workloads, introduces uncertainty due to the required inter\u2010modality image registrations, and involves unnecessary irradiation. While it would be beneficial to use exclusively MR images, a method needs to be employed to estimate a synthetic CT (sCT) for generating electron density maps and patient positioning reference images. We investigated 2D and 3D convolutional neural networks (CNNs) to generate a male pelvic sCT using a T1\u2010weighted MR image and compare their performance.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>A retrospective study was performed using CTs and T1\u2010weighted MR images of 20 prostate cancer patients. CTs were deformably registered to MR images to create CT\u2010MR pairs for training networks. The proposed 2D CNN, which contained 27 convolutional layers, was modified from the state\u2010of\u2010the\u2010art\u00a02D CNN to save computational memory and prepare for building the 3D CNN. The proposed 2D and 3D models were trained from scratch to map intensities of T1\u2010weighted MR images to CT Hounsfield Unit (HU) values. Each sCT was generated in a fivefold cross\u2010validation framework and compared with the corresponding deformed CT (dCT) using voxel\u2010wise mean absolute error (MAE). The sCT geometric accuracy was evaluated by comparing bone regions, defined by thresholding at 150\u00a0HU in the dCTs and the sCTs, using dice similarity coefficient (DSC), recall, and precision. To evaluate sCT patient positioning accuracy, bone regions in dCTs and sCTs were rigidly registered to the corresponding cone\u2010beam CTs. The resulting paired Euler transformation vectors were compared by calculating translation vector distances and absolute differences of Euler angles. Statistical tests were performed to evaluate the differences among the proposed models and Han\u2019s model.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Generating a pelvic sCT required approximately 5.5\u00a0s using the proposed models. The average MAEs within the body contour were 40.5\u00a0\u00b1\u00a05.4\u00a0HU (mean\u00a0\u00b1\u00a0SD) and 37.6\u00a0\u00b1\u00a05.1\u00a0HU for the 2D and 3D CNNs, respectively. The average DSC, recall, and precision for the bone region (thresholding the CT at 150\u00a0HU) were 0.81\u00a0\u00b1\u00a00.04, 0.85\u00a0\u00b1\u00a00.04, and 0.77\u00a0\u00b1\u00a00.09 for the 2D CNN, and 0.82\u00a0\u00b1\u00a00.04, 0.84\u00a0\u00b1\u00a00.04, and 0.80\u00a0\u00b1\u00a00.08 for the 3D CNN, respectively. For both models, mean translation vector distances are less than 0.6\u00a0mm with mean absolute differences of Euler angles less than 0.5\u00b0.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>The 2D and 3D CNNs generated accurate pelvic sCTs for the 20 patients using T1\u2010weighted MR images. Statistical tests indicated that the proposed 3D model was able to generate sCTs with smaller MAE and higher bone region precision compared to 2D models. Results of patient alignment tests suggested that sCTs generated by the proposed CNNs can provide accurate patient positioning. The accuracy of the dose calculation using generated sCTs will be tested and compared for the proposed models in the future.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.13672","type":"journal-article","created":{"date-parts":[[2019,6,20]],"date-time":"2019-06-20T14:39:37Z","timestamp":1561041577000},"page":"3788-3798","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":95,"title":["Deep learning approaches using 2D and 3D convolutional neural networks for generating male pelvic synthetic computed tomography from magnetic resonance imaging"],"prefix":"10.1002","volume":"46","author":[{"given":"Jie","family":"Fu","sequence":"first","affiliation":[{"name":"David Geffen School of Medicine University of California, Los Angeles 10833 Le Conte Ave Los Angeles 90095 CA USA"},{"name":"Department of Radiation Oncology University of California, Los Angeles 200 Suite B265, Medical Plaza Driveway Los Angeles 90095 CA USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yingli","family":"Yang","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology University of California, Los Angeles 200 Suite B265, Medical Plaza Driveway Los Angeles 90095 CA USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kamal","family":"Singhrao","sequence":"additional","affiliation":[{"name":"David Geffen School of Medicine University of California, Los Angeles 10833 Le Conte Ave Los Angeles 90095 CA USA"},{"name":"Department of Radiation Oncology University of California, Los Angeles 200 Suite B265, Medical Plaza Driveway Los Angeles 90095 CA USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dan","family":"Ruan","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology University of California, Los Angeles 200 Suite B265, Medical Plaza Driveway Los Angeles 90095 CA USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fang\u2010I","family":"Chu","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology University of California, Los Angeles 200 Suite B265, Medical Plaza Driveway Los Angeles 90095 CA USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daniel A.","family":"Low","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology University of California, Los Angeles 200 Suite B265, Medical Plaza Driveway Los Angeles 90095 CA USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"John H.","family":"Lewis","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology University of California, Los Angeles 200 Suite B265, Medical Plaza Driveway Los Angeles 90095 CA USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2019,7,26]]},"reference":[{"key":"e_1_2_8_2_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.semradonc.2014.02.003"},{"key":"e_1_2_8_3_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejmp.2017.02.017"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.1177\/153303460700600210"},{"key":"e_1_2_8_5_1","doi-asserted-by":"publisher","DOI":"10.1007\/s12094-012-0992-y"},{"key":"e_1_2_8_6_1","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.0501532102"},{"key":"e_1_2_8_7_1","doi-asserted-by":"publisher","DOI":"10.1038\/ncponc1073"},{"key":"e_1_2_8_8_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2009.01.065"},{"key":"e_1_2_8_9_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13014-016-0747-y"},{"key":"e_1_2_8_10_1","doi-asserted-by":"publisher","DOI":"10.1088\/0031-9155\/60\/2\/825"},{"key":"e_1_2_8_11_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2011.11.056"},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2015.08.045"},{"key":"e_1_2_8_13_1","doi-asserted-by":"publisher","DOI":"10.1186\/1748-717X-9-16"},{"key":"e_1_2_8_14_1","doi-asserted-by":"publisher","DOI":"10.1120\/jacmp.v15i5.4851"},{"key":"e_1_2_8_15_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13014-014-0309-0"},{"key":"e_1_2_8_16_1","doi-asserted-by":"publisher","DOI":"10.2967\/jnumed.111.092577"},{"key":"e_1_2_8_17_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4842575"},{"key":"e_1_2_8_18_1","doi-asserted-by":"publisher","DOI":"10.3109\/0284186X.2012.692883"},{"key":"e_1_2_8_19_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.3578928"},{"key":"e_1_2_8_20_1","doi-asserted-by":"publisher","DOI":"10.1088\/0031-9155\/60\/7\/2869"},{"key":"e_1_2_8_21_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4938266"},{"key":"e_1_2_8_22_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2015.07.001"},{"key":"e_1_2_8_23_1","doi-asserted-by":"publisher","DOI":"10.1088\/0031-9155\/58\/23\/8419"},{"key":"e_1_2_8_24_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.3696376"},{"key":"e_1_2_8_25_1","doi-asserted-by":"publisher","DOI":"10.1038\/nature14539"},{"key":"e_1_2_8_26_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46976-8_18"},{"key":"e_1_2_8_27_1","doi-asserted-by":"crossref","unstructured":"NieD TrulloR LianJ et al. Medical image synthesis with context\u2010aware generative adversarial networks. 2017.","DOI":"10.1007\/978-3-319-66179-7_48"},{"key":"e_1_2_8_28_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/aada6d"},{"key":"e_1_2_8_29_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.12155"},{"key":"e_1_2_8_30_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2016.2644615"},{"key":"e_1_2_8_31_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2009.02.019"},{"key":"e_1_2_8_32_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0360-3016(98)00351-4"},{"key":"e_1_2_8_33_1","doi-asserted-by":"publisher","DOI":"10.1259\/bjr\/41321492"},{"key":"e_1_2_8_34_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2010.2046908"},{"key":"e_1_2_8_35_1","doi-asserted-by":"publisher","DOI":"10.1109\/42.836373"},{"key":"e_1_2_8_36_1","doi-asserted-by":"publisher","DOI":"10.1109\/TSMC.1979.4310076"},{"key":"e_1_2_8_37_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2009.2035616"},{"key":"e_1_2_8_38_1","unstructured":"SimonyanK ZissermanA.Very deep convolutional networks for large\u2010scale image recognition. arXiv preprint arXiv:14091556.2014."},{"key":"e_1_2_8_39_1","doi-asserted-by":"crossref","unstructured":"UlyanovD VedaldiA LempitskyV.Improved texture networks: Maximizing quality and diversity in feed\u2010forward stylization and texture synthesis.2017.","DOI":"10.1109\/CVPR.2017.437"},{"key":"e_1_2_8_40_1","unstructured":"IoffeS SzegedyC.Batch normalization: Accelerating deep network training by reducing internal covariate shift.2015."},{"key":"e_1_2_8_41_1","unstructured":"IoffeS.Batch renormalization: Towards reducing minibatch dependence in batch\u2010normalized models.2017."},{"key":"e_1_2_8_42_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2016.2572683"},{"key":"e_1_2_8_43_1","doi-asserted-by":"crossref","unstructured":"HeK ZhangX RenS SunJ.Deep residual learning for image recognition.2016.","DOI":"10.1109\/CVPR.2016.90"},{"key":"e_1_2_8_44_1","unstructured":"AbadiM AgarwalA BarhamP et al.Tensorflow: Large\u2010scale machine learning on heterogeneous distributed systems. arXiv preprint arXiv:160304467.2016."},{"key":"e_1_2_8_45_1","unstructured":"KingmaD AdamBa J.A method for stochastic optimization. arXiv preprint arXiv:14126980.2014."},{"key":"e_1_2_8_46_1","doi-asserted-by":"crossref","unstructured":"HeK ZhangX RenS SunJ.Delving deep into rectifiers: Surpassing human\u2010level performance on imagenet classification.2015.","DOI":"10.1109\/ICCV.2015.123"},{"key":"e_1_2_8_47_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2014.09.015"},{"key":"e_1_2_8_48_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4958676"},{"key":"e_1_2_8_49_1","doi-asserted-by":"crossref","unstructured":"AndreasenD EdmundJM ZografosV MenzeBH Van LeemputK.Computed tomography synthesis from magnetic resonance images in the pelvis using multiple random forests and auto\u2010context features.2016.","DOI":"10.1117\/12.2216924"},{"key":"e_1_2_8_50_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4931417"},{"key":"e_1_2_8_51_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13187"},{"key":"e_1_2_8_52_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.1609271"},{"key":"e_1_2_8_53_1","doi-asserted-by":"crossref","unstructured":"\u00c7i\u00e7ek\u00d6 AbdulkadirA LienkampSS BroxT RonnebergerO.3D U\u2010Net: learning dense volumetric segmentation from sparse annotation.2016.","DOI":"10.1007\/978-3-319-46723-8_49"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.wiley.com\/onlinelibrary\/tdm\/v1\/articles\/10.1002%2Fmp.13672","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.13672","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.13672","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.13672","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,9,5]],"date-time":"2023-09-05T03:54:30Z","timestamp":1693886070000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.13672"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,7,26]]},"references-count":52,"journal-issue":{"issue":"9","published-print":{"date-parts":[[2019,9]]}},"alternative-id":["10.1002\/mp.13672"],"URL":"https:\/\/doi.org\/10.1002\/mp.13672","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2019,7,26]]},"assertion":[{"value":"2018-10-10","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2019-06-10","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2019-07-26","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}