{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,15]],"date-time":"2026-10-15T13:45:13Z","timestamp":1792071913112,"version":"4.4.0"},"reference-count":51,"publisher":"Wiley","issue":"9","license":[{"start":{"date-parts":[[2026,9,7]],"date-time":"2026-09-07T00:00:00Z","timestamp":1788739200000},"content-version":"vor","delay-in-days":6,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"},{"start":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T00:00:00Z","timestamp":1788220800000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2026,9]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Magnetic resonance (MR) images are essential for clinical diagnosis, yet their quality is often compromised by noise and motion artifacts, potentially obscuring anatomical details and reducing diagnostic accuracy. Existing deep learning approaches frequently address denoising and motion artifact suppression as separate tasks, overlook pixel correlations and varying noise levels, require manual parameter tuning, and lack uncertainty evaluation, thus limiting their practical utility in clinical\u00a0settings.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>This study aims to develop a Dual\u2010branch MRI Denoising and Motion Artifact Suppression with Uncertainty Estimation and Segmentation\u2010based Evaluation (DbUSNet). The model is designed to adaptively handle varying noise and motion artifact levels, preserve complex structural details, and provide uncertainty estimates for the processed outputs. This may offer insights into downstream clinical tasks such as image\u00a0segmentation.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>The DbUSNet integrates two specialized branches: A Heterogeneous Window Transformer Network (HWTNet) for adaptive denoising using both local and global feature modeling, and an improved U\u2010Net based Motion Artifact Suppression Network (ASNet). Each branch is coupled with an estimation subnetwork\u2014Noise Estimation Subnetwork (NES) and Artifact Estimation Subnetwork (AES)\u2014to dynamically assess and adapt to input specific degradation levels. The outputs of both branches are fused via an attention\u2010based fusion module. Uncertainty is quantified during inference using Monte Carlo (MC) dropout. The model was trained and tested on brain MR scans from Brats2021 dataset (2000 samples), OpenNeuro dataset (22 samples), and MR\u2010ART dataset (148 samples), incorporating synthetic Rician noise and motion artifacts. Downstream segmentation performance was evaluated on the Medical Segmentation Decathlon (MSD) brain dataset (1666 samples) and a prostate MR dataset (203 samples).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>On the Brats2021, OpenNeuro, and MR\u2010ART testsets, DbUSNet achieved PSNR of 28.23 dB, 27.13 dB, and 27.83 dB, SSIM of 0.93, 0.91, and 0.92, NMSE of 0.08, 0.08, and 0.08, respectively. For the downstream segmentation task, the denoised and motion artifact suppressed outputs led to average Dice score improvements of 3.10%, 3.59%, 2.46%, and 2.15% for T1, T1ce, FLAIR, and T2 modalities in the MSD brain dataset, and 2.16% and 2.24% for T2 and ADC sequences in the prostate\u00a0dataset.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The DbUSNet framework jointly performs adaptive denoising and motion artifact suppression within a unified pipeline and enables pixel\u2010wise uncertainty quantification. Evaluations on controlled brain MRI datasets show consistent image quality gains and improved downstream segmentation performance, though the limited size and healthy\u2010only composition of real motion test cohorts restrict wide clinical extrapolation. This work delivers a feasible technical prototype for MR image enhancement; larger, diverse clinical patient datasets will be required to further validate its clinical\u00a0utility.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.70662","type":"journal-article","created":{"date-parts":[[2026,9,7]],"date-time":"2026-09-07T08:50:58Z","timestamp":1788771058000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Dual\u2010branch MRI denoising and motion artifact suppression with uncertainty estimation and segmentation\u2010based evaluation"],"prefix":"10.1002","volume":"53","author":[{"given":"Mingfu","family":"Jiang","sequence":"first","affiliation":[{"name":"Faculty of Applied Sciences Macao Polytechnic University Macao China"},{"name":"College of Information Engineering Xinyang Agriculture and Forestry University Xinyang Henan China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chi Kin","family":"Lam","sequence":"additional","affiliation":[{"name":"Faculty of Applied Sciences Macao Polytechnic University Macao China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhuoneng","family":"Zhang","sequence":"additional","affiliation":[{"name":"Faculty of Applied Sciences Macao Polytechnic University Macao China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Poorya","family":"MohammadiNasab","sequence":"additional","affiliation":[{"name":"Center for Medical Physics and Biomedical Engineering Medical University Vienna Vienna Austria"},{"name":"Research Center for Clinical AI\u2010Research in Omics and Medical Data Science, Department of Medicine Danube Private University Krems Austria"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hing\u2010Chiu","family":"Chang","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering Chinese University of Hong Kong Hong Kong China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tong","family":"Tong","sequence":"additional","affiliation":[{"name":"College of Physics and Information Engineering Fuzhou University Fuzhou Fujian China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hualong","family":"She","sequence":"additional","affiliation":[{"name":"The Third People's Hospital of Chenzhou City Chenzhou Hunan China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhen","family":"Yuan","sequence":"additional","affiliation":[{"name":"Faculty of Health Sciences University of Macau Macao China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tao","family":"Tan","sequence":"additional","affiliation":[{"name":"Faculty of Applied Sciences Macao Polytechnic University Macao China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,9,7]]},"reference":[{"key":"e_1_2_8_2_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2022.103736"},{"key":"e_1_2_8_3_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2015.2427157"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10334-020-00881-1"},{"key":"e_1_2_8_5_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.mri.2020.05.002"},{"key":"e_1_2_8_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2022.119411"},{"key":"e_1_2_8_7_1","unstructured":"ZhangL WuM WangL SteffensDC PotterGG LiuM.Iterative learning for joint image denoising and motion artifact correction of 3D brain MRI.arXiv preprint posted online March 13 2024. doi:10.48550\/arXiv.2403.08162"},{"key":"e_1_2_8_8_1","doi-asserted-by":"publisher","DOI":"10.1162\/imag_a_00374"},{"key":"e_1_2_8_9_1","unstructured":"ZhuZ RehmanA CaoX et\u00a0al.Advancing low\u2010field MRI with a universal denoising imaging transformer: towards fast and high\u2010quality imaging.arXiv preprint posted online April 30 2024. doi:10.48550\/arXiv.2404.19167"},{"key":"e_1_2_8_10_1","unstructured":"ManorH MichaeliT.On the posterior distribution in denoising: Application to Uncertainty Quantification. In:International Conference on Learning Representations.ICLR;2024."},{"key":"e_1_2_8_11_1","doi-asserted-by":"crossref","unstructured":"ChenG ZhuF HengPA.An efficient statistical method for image noise level estimation. In:International Conference on Learning Representations.ICLR;2015:477\u2013485.","DOI":"10.1109\/ICCV.2015.62"},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2008.2001399"},{"key":"e_1_2_8_13_1","doi-asserted-by":"crossref","unstructured":"GuoS YanZ ZhangK ZuoW ZhangL.Toward convolutional blind denoising of real photographs. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition.IEEE;2019:1712\u20131722.","DOI":"10.1109\/CVPR.2019.00181"},{"key":"e_1_2_8_14_1","doi-asserted-by":"crossref","unstructured":"RonnebergerO FischerP BroxT.U\u2010net: Convolutional networks for biomedical image segmentation. In:Medical Image Computing and Computer\u2010Assisted Intervention\u2010MICCAI.Springer;2015:234\u2013241.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"e_1_2_8_15_1","unstructured":"BaidU GhodasaraS MohanS et\u00a0al.The RSNA\u2010ASNR\u2010MICCAI BraTS 2021 benchmark on brain tumor segmentation and radiogenomic classification.arXiv preprint posted online July 5 2021. doi:10.48550\/arXiv.2107.02314"},{"key":"e_1_2_8_16_1","volume-title":"Datasets with and without Deliberate Head Movements for Evaluating the Performance of Markerless Prospective Motion Correction and Selective Reacquisition in a General Clinical Protocol for Brain MRI","author":"Ganz M","year":"2023"},{"key":"e_1_2_8_17_1","doi-asserted-by":"crossref","unstructured":"EichhornH FrostR KongsgaardA et\u00a0al.Evaluating the performance of markerless prospective motion correction and selective reacquisition in a general clinical protocol for brain MRI.PsyArXiv preprint posted online November 9 2022. doi:10.31234\/osf.io\/vzh4g","DOI":"10.31234\/osf.io\/vzh4g"},{"key":"e_1_2_8_18_1","unstructured":"BakasS ReyesM IntE MenzeB.Identifying the best machine learning algorithms for brain tumor segmentation progression assessment and overall survival prediction in the BRATS challenge.arXiv preprint posted online November 18 2018. doi:10.48550\/arXiv.1811.02629"},{"key":"e_1_2_8_19_1","doi-asserted-by":"publisher","DOI":"10.1038\/sdata.2017.117"},{"key":"e_1_2_8_20_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2014.2377694"},{"key":"e_1_2_8_21_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejrad.2021.109647"},{"key":"e_1_2_8_22_1","doi-asserted-by":"publisher","DOI":"10.1002\/jmri.27585"},{"key":"e_1_2_8_23_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41597\u2010022\u201001694\u20108"},{"key":"e_1_2_8_24_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compmedimag.2025.102497"},{"key":"e_1_2_8_25_1","doi-asserted-by":"publisher","DOI":"10.1186\/1475-925X-14-2"},{"key":"e_1_2_8_26_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.1910340618"},{"key":"e_1_2_8_27_1","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2018.2839891"},{"key":"e_1_2_8_28_1","doi-asserted-by":"publisher","DOI":"10.13104\/imri.2020.24.4.196"},{"key":"e_1_2_8_29_1","first-page":"30","volume-title":"Advances in Neural Information Processing Systems","author":"Kendall A","year":"2017"},{"key":"e_1_2_8_30_1","doi-asserted-by":"publisher","DOI":"10.1109\/TSMC.2024.3429345"},{"key":"e_1_2_8_31_1","unstructured":"ChenH WangY GuoT et\u00a0al.Pre\u2010trained image processing transformer. In:IEEE\/CVF Conference on Computer Vision and Pattern Recognition2021:12299\u201312310."},{"key":"e_1_2_8_32_1","unstructured":"DosovitskiyA BeyerL KolesnikovA et\u00a0al.An image is worth 16x16 words: Transformers for image recognition at scale. In:International Conference on Learning Representations.2021."},{"key":"e_1_2_8_33_1","unstructured":"CallisSJ.Vision Transformers Explained Series. Technical Report LA\u2010UR\u201023\u201033876 Los Alamos National Laboratory (LANL);2024."},{"key":"e_1_2_8_34_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2023.110226"},{"key":"e_1_2_8_35_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2023.3273924"},{"key":"e_1_2_8_36_1","doi-asserted-by":"publisher","DOI":"10.1023\/A:1007665907178"},{"key":"e_1_2_8_37_1","doi-asserted-by":"crossref","unstructured":"KendallA GalY CipollaR.Multi\u2010task learning using uncertainty to weigh losses for scene geometry and semantics. In:Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2018:7482\u20137491.","DOI":"10.1109\/CVPR.2018.00781"},{"key":"e_1_2_8_38_1","unstructured":"GalY GhahramaniZ.Dropout as a bayesian approximation: representing model uncertainty in deep learning. In:International Conference on Machine Learning.PMLR;2016."},{"key":"e_1_2_8_39_1","doi-asserted-by":"crossref","unstructured":"BethellD GerasimouS CalinescuR.Robust uncertainty quantification using conformalised Monte Carlo prediction. In:Proceedings of the AAAI Conference on Artificial Intelligence. 38.2024:20939\u201320948.","DOI":"10.1609\/aaai.v38i19.30084"},{"key":"e_1_2_8_40_1","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2007.901238"},{"key":"e_1_2_8_41_1","unstructured":"DuffyBA ZhangW TangH et\u00a0al.Retrospective correction of motion artifact affected structural MRI images using deep learning of simulated motion. In:1st Conference on Medical Imaging with Deep Learning.2022."},{"key":"e_1_2_8_42_1","unstructured":"NapieralaMA.What is the Bonferroni correction?. In:The SAGE Encyclopedia of Communication Research Methods.SAGE Publications.2014:40\u201041.10.4135\/9781483381411.n45"},{"key":"e_1_2_8_43_1","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.e509"},{"key":"e_1_2_8_44_1","unstructured":"FardoFA ConfortoVH OliveiraFCD RodriguesPS.A formal evaluation of PSNR as quality measurement parameter for image segmentation algorithms.arXiv preprint posted online May 23 2016doi:10.48550\/arXiv.1605.07116"},{"key":"e_1_2_8_45_1","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2003.819861"},{"key":"e_1_2_8_46_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCI.2023.3299212"},{"key":"e_1_2_8_47_1","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2020.2973510"},{"key":"e_1_2_8_48_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-00296-0_5"},{"key":"e_1_2_8_49_1","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.g7327"},{"key":"e_1_2_8_50_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcm.2016.02.012"},{"key":"e_1_2_8_51_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.acra.2022.02.005"},{"key":"e_1_2_8_52_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-020-01008-z"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70662","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.70662","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70662","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,15]],"date-time":"2026-10-15T12:53:49Z","timestamp":1792068829000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.70662"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9]]},"references-count":51,"journal-issue":{"issue":"9","published-print":{"date-parts":[[2026,9]]}},"alternative-id":["10.1002\/mp.70662"],"URL":"https:\/\/doi.org\/10.1002\/mp.70662","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9]]},"assertion":[{"value":"2025-10-13","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-07-27","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-09-07","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70662"}}