{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T02:34:47Z","timestamp":1790130887760,"version":"4.0.1"},"reference-count":41,"publisher":"Wiley","issue":"7","license":[{"start":{"date-parts":[[2021,6,22]],"date-time":"2021-06-22T00:00:00Z","timestamp":1624320000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["81974276"],"award-info":[{"award-number":["81974276"]}],"id":[{"id":"10.13039\/501100001809","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":[[2021,7]]},"abstract":"<jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>Diffuse large B\u2010cell lymphoma (DLBCL) is an aggressive type of lymphoma with high mortality and poor prognosis that especially has a high incidence in Asia. Accurate segmentation of DLBCL lesions is crucial for clinical radiation therapy. However, manual delineation of DLBCL lesions is tedious and time\u2010consuming. Automatic segmentation provides an alternative solution but is difficult for diffuse lesions without the sufficient utilization of multimodality information. Our work is the first study focusing on positron emission tomography and computed tomography (PET\u2010CT) feature fusion for the DLBCL segmentation issue. We aim to improve the fusion performance of complementary information contained in PET\u2010CT imaging with a hybrid learning module in the supervised convolutional neural network.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>First, two encoder branches extract single\u2010modality features, respectively. Next, the hybrid learning component utilizes them to generate spatial fusion maps which can quantify the contribution of complementary information. Such feature fusion maps are then concatenated with specific\u2010modality (i.e., PET and CT) feature maps to obtain a representation of the final fused feature maps in different scales. Finally, the reconstruction part of our network creates a prediction map of DLBCL lesions by integrating and up\u2010sampling the final fused feature maps from encoder blocks in different scales.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The ability of our method was evaluated to detect foreground and segment lesions in three independent body regions (nasopharynx, chest, and abdomen) of a set of 45 PET\u2010CT scans. Extensive ablation experiments compared our method to four baseline techniques for multimodality fusion (input\u2010level (IL) fusion, multichannel (MC) strategy, multibranch (MB) strategy, and quantitative weighting (QW) fusion). The results showed that our method achieved a high detection accuracy (99.63% in the nasopharynx, 99.51% in the chest, and 99.21% in the abdomen) and had the superiority in segmentation performance with the mean dice similarity coefficient (DSC) of 73.03% and the modified Hausdorff distance (MHD) of 4.39\u00a0mm, when compared with the baselines (DSC: IL: 53.08%, MC: 63.59%, MB: 69.98%, and QW: 72.19%; MHD: IL: 12.16\u00a0mm, MC: 6.46\u00a0mm, MB: 4.83\u00a0mm, and QW: 4.89\u00a0mm).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>\n                      A promising segmentation method has been proposed for the challenging DLBCL lesions in PET\u2010CT images, which improves the understanding of complementary information by feature fusion and may guide clinical radiotherapy. The statistically significant analysis based on\n                      <jats:italic>P<\/jats:italic>\n                      \u2010value calculation has indicated a degree of significant difference between our proposed method and other baselines (almost metrics:\n                      <jats:italic>P<\/jats:italic>\n                      \u00a0&lt;\u00a00.05). This is a preliminary research using a small sample size, and we will collect data continually to achieve the larger verification study.\n                    <\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.14847","type":"journal-article","created":{"date-parts":[[2021,3,18]],"date-time":"2021-03-18T16:28:55Z","timestamp":1616084935000},"page":"3665-3678","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":32,"title":["Diffuse large B\u2010cell lymphoma segmentation in PET\u2010CT images via hybrid learning for feature fusion"],"prefix":"10.1002","volume":"48","author":[{"given":"Cheng","family":"Yuan","sequence":"first","affiliation":[{"name":"School of Biomedical Engineering Shanghai Jiao Tong University Shanghai 200040 China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Miao","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine Ruijin Hospital Shanghai Jiao Tong University School of Medicine Shanghai 200025 China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinyun","family":"Huang","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine Ruijin Hospital Shanghai Jiao Tong University School of Medicine Shanghai 200025 China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wei","family":"Xie","sequence":"additional","affiliation":[{"name":"Department of Hematology Ruijin Hospital Shanghai Jiao Tong University School of Medicine Shanghai 200025 China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaozhu","family":"Lin","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine Ruijin Hospital Shanghai Jiao Tong University School of Medicine Shanghai 200025 China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Weili","family":"Zhao","sequence":"additional","affiliation":[{"name":"Department of Hematology Ruijin Hospital Shanghai Jiao Tong University School of Medicine Shanghai 200025 China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Biao","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine Ruijin Hospital Shanghai Jiao Tong University School of Medicine Shanghai 200025 China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dahong","family":"Qian","sequence":"additional","affiliation":[{"name":"School of Biomedical Engineering Shanghai Jiao Tong University Shanghai 200040 China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2021,6,22]]},"reference":[{"key":"e_1_2_9_2_1","doi-asserted-by":"publisher","DOI":"10.3322\/caac.21357"},{"key":"e_1_2_9_3_1","first-page":"1326","article-title":"The role of pet in lymphoma","volume":"47","author":"Jhanwar YS","year":"2006","journal-title":"J Nucl Med"},{"key":"e_1_2_9_4_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00259-010-1423-3"},{"key":"e_1_2_9_5_1","doi-asserted-by":"publisher","DOI":"10.1109\/JBHI.2020.2972694"},{"key":"e_1_2_9_6_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00330-005-0088-y"},{"key":"e_1_2_9_7_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00259-014-2961-x"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.2967\/jnumed.107.044362"},{"key":"e_1_2_9_9_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2004.06.254"},{"key":"e_1_2_9_10_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00259-006-0252-x"},{"key":"e_1_2_9_11_1","doi-asserted-by":"crossref","unstructured":"DesbordesP PetitjeanC RuanS.3D automated lymphoma segmentation in PET images based on cellular automata.2014 4th International Conference on Image Processing Theory Tools and Applications (IPTA).https:\/\/doi.org\/10.1109\/ipta.2014.7001923","DOI":"10.1109\/IPTA.2014.7001923"},{"key":"e_1_2_9_12_1","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-bioeng-071516-044442"},{"key":"e_1_2_9_13_1","doi-asserted-by":"crossref","unstructured":"RothHR LuL SeffA et\u00a0al.A New 2.5D Representation for Lymph Node Detection Using Random Sets of Deep Convolutional Neural Network Observations.2014 Medical Image Computing and Computer Assisted Intervention (MICCAI).","DOI":"10.1007\/978-3-319-10404-1_65"},{"key":"e_1_2_9_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compmedimag.2016.11.008"},{"key":"e_1_2_9_15_1","doi-asserted-by":"crossref","unstructured":"HuH ShenL ZhouT DecazesP RuanS.Lymphoma segmentation in PET images based on multi\u2010view and Conv3D fusion strategy.2020 17th IEEE International Symposium on Biomedical Imaging (ISBI).","DOI":"10.1109\/ISBI45749.2020.9098595"},{"key":"e_1_2_9_16_1","doi-asserted-by":"publisher","DOI":"10.1007\/s12149-013-0719-3"},{"key":"e_1_2_9_17_1","doi-asserted-by":"publisher","DOI":"10.1109\/TRPMS.2018.2890359"},{"key":"e_1_2_9_18_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2019.2963254"},{"key":"e_1_2_9_19_1","unstructured":"NgiamJ KhoslaA KimM NamJ NgAY.Multimodal Deep Learning.2009 International Conference on Machine Learning;2009:689\u2013696.http:\/\/www.icml\u20102011.org\/papers\/399_icmlpaper.pdf"},{"key":"e_1_2_9_20_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2018.2823900"},{"key":"e_1_2_9_21_1","first-page":"225019","article-title":"Deep learning for classification of benign and malignant bone lesions in [F\u201018] NaF PET\/CT images","volume":"63","author":"Bradshaw T","year":"2018","journal-title":"J Nucl Med"},{"key":"e_1_2_9_22_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2014.12.061"},{"key":"e_1_2_9_23_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4948498"},{"key":"e_1_2_9_24_1","doi-asserted-by":"crossref","unstructured":"ZhongZ KimY ZhouL et\u00a0al.3D fully convolutional networks for co\u2010segmentation of tumors on PET\u2010CT images.2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI).https:\/\/doi.org\/10.1109\/ISBI.2018.8363561","DOI":"10.1109\/ISBI.2018.8363561"},{"key":"e_1_2_9_25_1","doi-asserted-by":"crossref","unstructured":"TulderGV BruijneMD.Representation learning for cross\u2010modality classification.International MICCAI Workshop on Medical Computer Vision and Bayesian and Graphical Models for Biomedical Imaging;2017:126\u2013136.https:\/\/link.springer.com\/chapter\/10.1007\/978\u20103\u2010319\u201061188\u20104_12.","DOI":"10.1007\/978-3-319-61188-4_12"},{"key":"e_1_2_9_26_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13331"},{"key":"e_1_2_9_27_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/aaf44b"},{"key":"e_1_2_9_28_1","first-page":"1431","article-title":"Understanding the standardized uptake value, its methods, and implications for usage","volume":"45","author":"Thie JA","year":"2004","journal-title":"J Nucl Med"},{"key":"e_1_2_9_29_1","unstructured":"SimonyanK ZissermanA.Very deep convolutional networks for large\u2010scale image recognition.2014 Computer Vision and Pattern Recognition (CVPR)."},{"key":"e_1_2_9_30_1","unstructured":"IoffeS SzegedyC.Batch normalization: accelerating deep network training by reducing internal covariate shift.2015 32nd International Conference on Machine Learning (ICML);2015:448\u2013456."},{"key":"e_1_2_9_31_1","unstructured":"MaasAL HannunAY NgAY.Rectifier nonlinearities improve neural network acoustic models.ICML Workshop on Deep Learning for Audio Speech and Language;2013;30(1).http:\/\/ai.stanford.edu\/amaas\/papers\/relu_hybrid_icml2013_final.pdf."},{"key":"e_1_2_9_32_1","doi-asserted-by":"publisher","DOI":"10.1111\/1467-8659.00356"},{"key":"e_1_2_9_33_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2015.04.018"},{"key":"e_1_2_9_34_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2019.2923601"},{"key":"e_1_2_9_35_1","unstructured":"DumoulinV VisinF.A guide to convolution arithmetic for deep learning.Preprint arXiv:1603.07285;2016."},{"key":"e_1_2_9_36_1","unstructured":"AbadiM BarhamP ChenJ et\u00a0al.TensorFlow: a system for large\u2010scale machine learning.2016 12th USENIX Conference on Operating Systems Design and Implementation;2016:265\u2013283.https:\/\/arxiv.org\/abs\/1605.08695"},{"key":"e_1_2_9_37_1","unstructured":"ChetlurS WoolleyC VandermerschP et\u00a0al.cuDNN: efficient primitives for deep learning.Preprint arXiv:1410.0759;2014."},{"key":"e_1_2_9_38_1","unstructured":"SutskeverI MartensJ DahlG HintonG.On the importance of initialization and momentum in deep learning.2013 30th International Conference on Machine Learning (ICML);2013:1139\u20131147.https:\/\/doi.org\/10.1007\/s00287\u2010015\u20100911\u2010z"},{"key":"e_1_2_9_39_1","doi-asserted-by":"crossref","unstructured":"DubuissonMP JainAK.A Modified Hausdorff Distance for Object Matching.12th International Conference on Pattern Recognition;1994.https:\/\/doi.org\/10.1109\/ICPR.1994.576361","DOI":"10.1109\/ICPR.1994.576361"},{"key":"e_1_2_9_40_1","first-page":"5168","article-title":"RefineNet: multi\u2010path refinement networks for high\u2010resolution semantic segmentation","volume":"2017","author":"Lin G","year":"2017","journal-title":"Comput Vis Pattern Recogn (CVPR)"},{"key":"e_1_2_9_41_1","doi-asserted-by":"crossref","unstructured":"SzegedyC IoffeS VanhouckeV AlemiA.Inception\u2010v4 inception\u2010ResNet and the impact of residual connections on learning.2016 Computer Vision and Pattern Recognition (CVPR);2016.https:\/\/arxiv.org\/pdf\/1602.07261.pdf.","DOI":"10.1609\/aaai.v31i1.11231"},{"key":"e_1_2_9_42_1","first-page":"770","article-title":"Deep residual learning for image recognition","volume":"2016","author":"He K","year":"2016","journal-title":"Computer Vision and Pattern Recognition (CVPR)"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.14847","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.14847","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.14847","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,8,27]],"date-time":"2023-08-27T02:30:21Z","timestamp":1693103421000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.14847"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,6,22]]},"references-count":41,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2021,7]]}},"alternative-id":["10.1002\/mp.14847"],"URL":"https:\/\/doi.org\/10.1002\/mp.14847","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,6,22]]},"assertion":[{"value":"2020-10-13","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-03-10","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-06-22","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}