{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T15:00:49Z","timestamp":1777906849313,"version":"3.51.4"},"reference-count":51,"publisher":"Wiley","issue":"3","license":[{"start":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T00:00:00Z","timestamp":1773878400000},"content-version":"vor","delay-in-days":18,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"},{"start":{"date-parts":[[2026,3,1]],"date-time":"2026-03-01T00:00:00Z","timestamp":1772323200000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/100020595","name":"National Science and Technology Council","doi-asserted-by":"publisher","award":["NSTC114\u20102221\u2010E\u2010002\u2010093"],"award-info":[{"award-number":["NSTC114\u20102221\u2010E\u2010002\u2010093"]}],"id":[{"id":"10.13039\/100020595","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":[[2026,3]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Intravoxel incoherent motion (IVIM) imaging, a diffusion magnetic resonance imaging technique, is commonly used to quantify tissue perfusion and diffusion. Traditional pixel\u2010by\u2010pixel fitting methods, however, often suffer from high noise, causing unreliable parameter estimates.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose and Methods<\/jats:title>\n                    <jats:p>To address this issue, a novel unsupervised learning\u2010based framework combining a one\u2010dimensional (1D) convolutional neural network (CNN) with a bidirectional long short\u2010term memory (BiLSTM) network and a multi\u2010head attention mechanism (MHAM) was proposed. Several techniques were proposed to reduce the effect of random weight initialization, noisy input data, and overfitting\/underfitting on the estimation of IVIM parameters. The performance of the proposed method was evaluated using both simulated and experimental data, and the results were compared with those obtained using the deep neural network (DNN) method and the Bayesian\u2010Markov random fields (MRF) method.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Simulation results showed that the proposed method achieved lower root mean square error values than the other two methods, indicating more reliable IVIM parameter estimates. The only exception was at a signal\u2010to\u2010noise ratio of 100, where it performed similarly to the Bayesian\u2010MRF method. For the abdominal datasets, the proposed method yielded IVIM parameter estimates that closely matched the literature\u2010reported values and avoided the overestimation of pseudo\u2010diffusion coefficients (\n                      <jats:italic>D<\/jats:italic>\n                      <jats:sup>*<\/jats:sup>\n                      ) observed in the other two methods. For the brain dataset, the perfusion fractions and diffusion coefficients obtained from all three methods were consistent with the literature\u2010reported ranges; however, only the DNN method tended to overestimate\n                      <jats:italic>D<\/jats:italic>\n                      <jats:sup>*<\/jats:sup>\n                      .\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>These findings suggest that the proposed CNN\u2010BiLSTM\u2010MHAM model is a promising approach for IVIM parameter estimation.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.70396","type":"journal-article","created":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T16:01:37Z","timestamp":1773936097000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Unsupervised 1D CNN \u2010bidirectional long short\u2010term memory model with multi\u2010head attention for generating intravoxel incoherent motion maps"],"prefix":"10.1002","volume":"53","author":[{"given":"Zhong\u2010Yi","family":"Li","sequence":"first","affiliation":[{"name":"Institute of Medical Device and Imaging College of Medicine National Taiwan University  Taipei City Taiwan"},{"name":"Department of Radiology The Second Affiliated Hospital Zhejiang University School of Medicine  Hangzhou Zhejiang China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hsuan\u2010Ming","family":"Huang","sequence":"additional","affiliation":[{"name":"Institute of Medical Device and Imaging College of Medicine National Taiwan University  Taipei City Taiwan"},{"name":"Program for Precision Health and Intelligent Medicine, Graduate School of Advanced Technology National Taiwan University  Taipei City Taiwan"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"311","published-online":{"date-parts":[[2026,3,19]]},"reference":[{"key":"e_1_2_9_2_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.26373"},{"key":"e_1_2_9_3_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiology.168.2.3393671"},{"key":"e_1_2_9_4_1","first-page":"119","volume-title":"Investigations on the Theory of the Brownian Movement","author":"Einstein A","year":"1956"},{"key":"e_1_2_9_5_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiology.161.2.3763909"},{"key":"e_1_2_9_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejrad.2013.08.006"},{"key":"e_1_2_9_7_1","doi-asserted-by":"publisher","DOI":"10.1002\/jmri.24172"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598\u2010020\u201064804\u20109"},{"key":"e_1_2_9_9_1","doi-asserted-by":"publisher","DOI":"10.1002\/jmri.24195"},{"key":"e_1_2_9_10_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2021.102144"},{"key":"e_1_2_9_11_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.1910320109"},{"key":"e_1_2_9_12_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.27036"},{"key":"e_1_2_9_13_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.1910230113"},{"key":"e_1_2_9_14_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2493080080"},{"key":"e_1_2_9_15_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.25484"},{"key":"e_1_2_9_16_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejrad.2009.06.031"},{"key":"e_1_2_9_17_1","doi-asserted-by":"publisher","DOI":"10.1137\/S1064827595289108"},{"key":"e_1_2_9_18_1","doi-asserted-by":"publisher","DOI":"10.1007\/BF01580735"},{"key":"e_1_2_9_19_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.24649"},{"key":"e_1_2_9_20_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.25765"},{"key":"e_1_2_9_21_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11222\u2010011\u20109288\u20102"},{"key":"e_1_2_9_22_1","doi-asserted-by":"publisher","DOI":"10.1002\/nbm.4774"},{"key":"e_1_2_9_23_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.28708"},{"key":"e_1_2_9_24_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.27910"},{"key":"e_1_2_9_25_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.28852"},{"key":"e_1_2_9_26_1","unstructured":"SummersC DinneenMJ Nondeterminism and instability in neural network optimization.International Conference on Machine Learning.2021:9913\u20109922."},{"key":"e_1_2_9_27_1","doi-asserted-by":"publisher","DOI":"10.1186\/s41074\u2010017\u20100030\u20107"},{"key":"e_1_2_9_28_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.imu.2023.101161"},{"key":"e_1_2_9_29_1","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"e_1_2_9_30_1","first-page":"5998","volume-title":"Advances in Neural Information Processing Systems","author":"Vaswani A","year":"2017"},{"key":"e_1_2_9_31_1","doi-asserted-by":"crossref","unstructured":"AlbawiS MohammedTA Al\u2010ZawiS.Understanding of a convolutional neural network.Proc Int Conf Eng Technol.2017:1\u20106.","DOI":"10.1109\/ICEngTechnol.2017.8308186"},{"key":"e_1_2_9_32_1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1508.04025"},{"key":"e_1_2_9_33_1","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2008.925382"},{"key":"e_1_2_9_34_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0073021"},{"key":"e_1_2_9_35_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jmr.2006.01.016"},{"key":"e_1_2_9_36_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2007.906087"},{"key":"e_1_2_9_37_1","doi-asserted-by":"publisher","DOI":"10.1023\/A:1021961616621"},{"key":"e_1_2_9_38_1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1412.6980"},{"key":"e_1_2_9_39_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.12111327"},{"key":"e_1_2_9_40_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.13122712"},{"key":"e_1_2_9_41_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4915495"},{"key":"e_1_2_9_42_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.25410"},{"key":"e_1_2_9_43_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0113846"},{"key":"e_1_2_9_44_1","doi-asserted-by":"publisher","DOI":"10.1002\/jmri.25519"},{"key":"e_1_2_9_45_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0194590"},{"key":"e_1_2_9_46_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2018.04.064"},{"key":"e_1_2_9_47_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361\u20106560\/ac9a1f"},{"key":"e_1_2_9_48_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.mri.2014.07.013"},{"key":"e_1_2_9_49_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00330\u2010015\u20103655\u2010x"},{"key":"e_1_2_9_50_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2019.116228"},{"key":"e_1_2_9_51_1","article-title":"Early stopping for deep image prior","author":"Wang H","year":"2021","journal-title":"arXiv"},{"key":"e_1_2_9_52_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11263\u2010021\u201001572\u20107"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70396","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.70396","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70396","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T13:26:32Z","timestamp":1777641992000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.70396"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3]]},"references-count":51,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2026,3]]}},"alternative-id":["10.1002\/mp.70396"],"URL":"https:\/\/doi.org\/10.1002\/mp.70396","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3]]},"assertion":[{"value":"2025-08-16","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-03-03","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-03-19","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70396"}}