{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,9]],"date-time":"2026-06-09T03:21:48Z","timestamp":1780975308886,"version":"3.54.1"},"reference-count":50,"publisher":"Wiley","issue":"2","license":[{"start":{"date-parts":[[2021,9,4]],"date-time":"2021-09-04T00:00:00Z","timestamp":1630713600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100010677","name":"H2020 Health","doi-asserted-by":"publisher","award":["E! 12326 ILLUMINUS"],"award-info":[{"award-number":["E! 12326 ILLUMINUS"]}],"id":[{"id":"10.13039\/100010677","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001711","name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung","doi-asserted-by":"publisher","award":["SNRF 320030_176052"],"award-info":[{"award-number":["SNRF 320030_176052"]}],"id":[{"id":"10.13039\/501100001711","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Magnetic Resonance in Med"],"published-print":{"date-parts":[[2022,2]]},"abstract":"<jats:sec><jats:title>Purpose<\/jats:title><jats:p>We compare the performance of three commonly used MRI\u2010guided attenuation correction approaches in torso PET\/MRI, namely segmentation\u2010, atlas\u2010, and deep learning\u2010based algorithms.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>Twenty\u2010five co\u2010registered torso <jats:sup>18<\/jats:sup>F\u2010FDG PET\/CT and PET\/MR images were enrolled. PET attenuation maps were generated from in\u2010phase Dixon MRI using a three\u2010tissue class segmentation\u2010based approach (soft\u2010tissue, lung, and background air), voxel\u2010wise weighting atlas\u2010based approach, and a residual convolutional neural network. The bias in standardized uptake value (SUV) was calculated for each approach considering CT\u2010based attenuation corrected PET images as reference. In addition to the overall performance assessment of these approaches, the primary focus of this work was on recognizing the origins of potential outliers, notably body truncation, metal\u2010artifacts, abnormal anatomy, and small malignant lesions in the lungs.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>The deep learning approach outperformed both atlas\u2010 and segmentation\u2010based methods resulting in less than 4% SUV bias across 25 patients compared to the segmentation\u2010based method with up to 20% SUV bias in bony structures and the atlas\u2010based method with 9% bias in the lung. The deep learning\u2010based method exhibited superior performance. Yet, in case of sever truncation and metallic\u2010artifacts in the input MRI, this approach was outperformed by the atlas\u2010based method, exhibiting suboptimal performance in the affected regions. Conversely, for abnormal anatomies, such as a patient presenting with one lung or small malignant lesion in the lung, the deep learning algorithm exhibited promising performance compared to other methods.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusion<\/jats:title><jats:p>The deep learning\u2010based method provides promising outcome for synthetic CT generation from MRI. However, metal\u2010artifact and body truncation should be specifically addressed.<\/jats:p><\/jats:sec>","DOI":"10.1002\/mrm.29003","type":"journal-article","created":{"date-parts":[[2021,9,4]],"date-time":"2021-09-04T13:16:55Z","timestamp":1630761415000},"page":"686-701","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":20,"title":["MRI\u2010guided attenuation correction in torso PET\/MRI: Assessment of segmentation\u2010, atlas\u2010, and deep learning\u2010based approaches in the presence of outliers"],"prefix":"10.1002","volume":"87","author":[{"given":"Hossein","family":"Arabi","sequence":"first","affiliation":[{"name":"Division of Nuclear Medicine and Molecular Imaging Geneva University Hospital Geneva Switzerland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7559-5297","authenticated-orcid":false,"given":"Habib","family":"Zaidi","sequence":"additional","affiliation":[{"name":"Division of Nuclear Medicine and Molecular Imaging Geneva University Hospital Geneva Switzerland"},{"name":"Geneva University Neurocenter Geneva University Geneva Switzerland"},{"name":"Department of Nuclear Medicine and Molecular Imaging University of Groningen University Medical Center Groningen Groningen Netherlands"},{"name":"Department of Nuclear Medicine University of Southern Denmark Odense Denmark"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2021,9,4]]},"reference":[{"key":"e_1_2_7_2_1","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2015.2482225"},{"key":"e_1_2_7_3_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4941014"},{"key":"e_1_2_7_4_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11307-014-0777-5"},{"key":"e_1_2_7_5_1","doi-asserted-by":"publisher","DOI":"10.2967\/jnumed.108.054726"},{"key":"e_1_2_7_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2017.09.006"},{"key":"e_1_2_7_7_1","doi-asserted-by":"publisher","DOI":"10.2967\/jnumed.117.204156"},{"key":"e_1_2_7_8_1","doi-asserted-by":"publisher","DOI":"10.1186\/2197-7364-2-S1-A31"},{"key":"e_1_2_7_9_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2017170700"},{"key":"e_1_2_7_10_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2016.02.002"},{"key":"e_1_2_7_11_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00259-016-3422-5"},{"key":"e_1_2_7_12_1","doi-asserted-by":"publisher","DOI":"10.1186\/s41824-020-00086-8"},{"key":"e_1_2_7_13_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13187"},{"key":"e_1_2_7_14_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14075"},{"key":"e_1_2_7_15_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13617"},{"key":"e_1_2_7_16_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0223141"},{"key":"e_1_2_7_17_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00259-019-04380-x"},{"key":"e_1_2_7_18_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14418"},{"key":"e_1_2_7_19_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ab0606"},{"key":"e_1_2_7_20_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00259-020-04852-5"},{"key":"e_1_2_7_21_1","doi-asserted-by":"publisher","DOI":"10.1002\/hbm.25039"},{"key":"e_1_2_7_22_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ab4eb7"},{"key":"e_1_2_7_23_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2020.101718"},{"key":"e_1_2_7_24_1","doi-asserted-by":"publisher","DOI":"10.2967\/jnumed.118.219493"},{"key":"e_1_2_7_25_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ab41af"},{"key":"e_1_2_7_26_1","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-bioeng-082420-020343"},{"key":"e_1_2_7_27_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4963809"},{"key":"e_1_2_7_28_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ab25bc"},{"key":"e_1_2_7_29_1","doi-asserted-by":"publisher","DOI":"10.1088\/0031-9155\/56\/10\/013"},{"key":"e_1_2_7_30_1","doi-asserted-by":"publisher","DOI":"10.1120\/jacmp.v14i4.4163"},{"key":"e_1_2_7_31_1","first-page":"348","volume-title":"On the Compactness, Efficiency, and Representation of 3D Convolutional Networks: Brain Parcellation as a Pretext Task","author":"Li W","year":"2017"},{"key":"e_1_2_7_32_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cmpb.2018.01.025"},{"key":"e_1_2_7_33_1","article-title":"Comparison of different deep learning architectures for synthetic CT generation from MR images","author":"Bahrami A","journal-title":"arXiv preprint arXiv:210506315 2021"},{"key":"e_1_2_7_34_1","article-title":"Deep learning\u2010based synthetic CT generation from MR images: comparison of generative adversarial and residual neural networks","author":"Gholamiankhah F","year":"2021","journal-title":"arXiv preprint arXiv:2103.01609"},{"key":"e_1_2_7_35_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2021.102079"},{"key":"e_1_2_7_36_1","doi-asserted-by":"publisher","DOI":"10.1097\/RLU.0000000000003585"},{"key":"e_1_2_7_37_1","doi-asserted-by":"publisher","DOI":"10.52547\/ijrr.19.2.7"},{"key":"e_1_2_7_38_1","article-title":"A disciplined approach to neural network hyper\u2010parameters: part 1\u2013learning rate, batch size, momentum, and weight decay","author":"Smith LN","journal-title":"arXiv preprint arXiv:180309820 2018"},{"key":"e_1_2_7_39_1","doi-asserted-by":"publisher","DOI":"10.1088\/0031-9155\/61\/17\/6531"},{"key":"e_1_2_7_40_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2016.01.060"},{"key":"e_1_2_7_41_1","doi-asserted-by":"publisher","DOI":"10.1120\/jacmp.v14i4.4163"},{"key":"e_1_2_7_42_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00259-010-1603-1"},{"key":"e_1_2_7_43_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.2174132"},{"key":"e_1_2_7_44_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00330-021-07709-z"},{"key":"e_1_2_7_45_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.3484090"},{"key":"e_1_2_7_46_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13550-020-00644-y"},{"key":"e_1_2_7_47_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2016.11.003"},{"key":"e_1_2_7_48_1","doi-asserted-by":"publisher","DOI":"10.2967\/jnumed.110.078949"},{"key":"e_1_2_7_49_1","doi-asserted-by":"publisher","DOI":"10.1109\/TNS.2013.2278624"},{"key":"e_1_2_7_50_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/abb02c"},{"key":"e_1_2_7_51_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.12449"}],"container-title":["Magnetic Resonance in Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mrm.29003","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mrm.29003","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mrm.29003","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,8,24]],"date-time":"2023-08-24T09:22:31Z","timestamp":1692868951000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/mrm.29003"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,9,4]]},"references-count":50,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2022,2]]}},"alternative-id":["10.1002\/mrm.29003"],"URL":"https:\/\/doi.org\/10.1002\/mrm.29003","archive":["Portico"],"relation":{},"ISSN":["0740-3194","1522-2594"],"issn-type":[{"value":"0740-3194","type":"print"},{"value":"1522-2594","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,9,4]]},"assertion":[{"value":"2021-06-04","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-08-21","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-09-04","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}