{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,15]],"date-time":"2026-05-15T02:48:27Z","timestamp":1778813307003,"version":"3.51.4"},"reference-count":30,"publisher":"Termedia Sp. z.o.o.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Pol J Radiol"],"abstract":"<jats:sec><jats:title>Purpose<\/jats:title><jats:p>To evaluate the feasibility of using a deep learning (DL) model to generate fat-suppression images and detect abnormalities on knee magnetic resonance imaging (MRI) through the fat-suppression image-subtraction method.<\/jats:p><\/jats:sec><jats:sec><jats:title>Material and methods<\/jats:title><jats:p>A total of 45 knee MRI studies in patients with knee disorders and 12 knee MRI studies in healthy volunteers were enrolled. The DL model was developed using 2-dimensional convolutional neural networks for generating fat-suppression images and subtracting generated fat-suppression images without any abnormal findings from those with normal\/abnormal findings and detecting\/classifying abnormalities on knee MRI. The image qualities of the generated fat-suppression images and subtraction-images were assessed. The accuracy, average precision, average recall, F-measure, sensitivity, and area under the receiver operator characteristic curve (AUROC) of DL or each abnormality were calculated.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>A total of 2472 image datasets, each consisting of one slice of original T1WI, original intermediate-weighted images, generated fat-suppression (FS)-intermediate-weighted images without any abnormal findings, generated FS-intermediate-weighted images with normal\/abnormal findings, and subtraction images between the generated FS-intermediate-weighted images at the same cross-section, were created. The generated fat-suppression images were of adequate image quality. Of the 2472 subtraction-images, 2203 (89.1%) were judged to be of adequate image quality. The accuracies for overall abnormalities, anterior cruciate ligament, bone marrow, cartilage, meniscus, and others were 89.5-95.1%. The average precision, average recall, and F-measure were 73.4-90.6%, 77.5-89.4%, and 78.4-89.4%, respectively. The sensitivity was 57.4-90.5%. The AUROCs were 0.910-0.979.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusions<\/jats:title><jats:p>The DL model was able to generate fat-suppression images of sufficient quality to detect abnormalities on knee MRI through the fat-suppression image-subtraction method.<\/jats:p><\/jats:sec>","DOI":"10.5114\/pjr.2023.133660","type":"journal-article","created":{"date-parts":[[2023,12,19]],"date-time":"2023-12-19T10:20:19Z","timestamp":1702981219000},"page":"562-573","source":"Crossref","is-referenced-by-count":5,"title":["Feasibility of the fat-suppression image-subtraction method using deep learning for abnormality detection on knee MRI"],"prefix":"10.5114","volume":"88","author":[{"given":"Shusuke","family":"Kasuya","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tsutomu","family":"Inaoka","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Akihiko","family":"Wada","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tomoya","family":"Nakatsuka","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Koichi","family":"Nakagawa","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hitoshi","family":"Terada","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"3061","published-online":{"date-parts":[[2023,12,8]]},"reference":[{"key":"ref-1157117","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics12020537"},{"key":"ref-1157118","doi-asserted-by":"crossref","first-page":"315","DOI":"10.1007\/s00256-021-03830-8","article-title":"Artificial intelligence for MRI diagnosis of joints: a scoping review of the current state-of-the art of deep learningbased approaches","volume":"51","author":"Fritz B","year":"2022","unstructured":"Fritz B, Fritz J. Artificial intelligence for MRI diagnosis of joints: a scoping review of the current state-of-the art of deep learningbased approaches. Skeletal Radiol 2022; 51: 315-329.","journal-title":"Skeletal Radiol"},{"key":"ref-1157119","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.2020190207"},{"key":"ref-1157120","doi-asserted-by":"crossref","first-page":"400","DOI":"10.1002\/jmri.26246","article-title":"3D convolutional neural networks for detection and severity staging of meniscus and PFJ cartilage morphological degenerative changes in osteoarthritis and anterior cruciate ligament subjects","volume":"49","author":"Pedoia V","year":"2019","unstructured":"Pedoia V, Norman B, Mehany SN, et al. 3D convolutional neural networks for detection and severity staging of meniscus and PFJ cartilage morphological degenerative changes in osteoarthritis and anterior cruciate ligament subjects. J Magn Reson Imaging 2019; 49: 400-410.","journal-title":"J Magn Reson Imaging"},{"key":"ref-1157121","doi-asserted-by":"crossref","first-page":"160","DOI":"10.1148\/radiol.2018172986","article-title":"Deep learning approach for evaluating knee MR images: achieving high diagnostic performance for cartilage lesion detection","volume":"289","author":"Liu F","year":"2018","unstructured":"Liu F, Zhou Z, Samsonov A, et al. Deep learning approach for evaluating knee MR images: achieving high diagnostic performance for cartilage lesion detection. Radiology 2018; 289: 160-169.","journal-title":"Radiology"},{"key":"ref-1157122","doi-asserted-by":"crossref","first-page":"980","DOI":"10.1007\/s10278-019-00193-4","article-title":"Deep learning for detection of complete anterior cruciate ligament tear","volume":"32","author":"Chang PD","year":"2019","unstructured":"Chang PD, Wong TT, Rasiej MJ. Deep learning for detection of complete anterior cruciate ligament tear. J Digit Imaging 2019; 32: 980-986.","journal-title":"J Digit Imaging"},{"key":"ref-1157123","first-page":"499","article-title":"Deep convolutional neural network-based diagnosis of anterior cruciate ligament tears","volume":"55","author":"Germann C","year":"2020","unstructured":"Germann C, Marbach G, Civardi F, et al. Deep convolutional neural network-based diagnosis of anterior cruciate ligament tears. Invest Radiol 2020; 55: 499-506.","journal-title":"Invest Radiol"},{"key":"ref-1157124","doi-asserted-by":"crossref","first-page":"235","DOI":"10.1016\/j.diii.2019.03.002","article-title":"Automatic knee meniscus tear detection and orientation classification with Mask-RCNN","volume":"100","author":"Couteaux V","year":"2019","unstructured":"Couteaux V, Si-Mohamed S, Nempont O, et al. Automatic knee meniscus tear detection and orientation classification with Mask-RCNN. Diag Interv Imaging 2019; 100: 235-242.","journal-title":"Diag Interv Imaging"},{"key":"ref-1157125","doi-asserted-by":"crossref","first-page":"64","DOI":"10.1016\/j.ejmp.2021.02.010","article-title":"Meniscal lesion detection and characterization in adult knee MRI: a deep learning model approach with external validation","volume":"83","author":"Rizk B","year":"2021","unstructured":"Rizk B, Brat H, Zille P, et al. Meniscal lesion detection and characterization in adult knee MRI: a deep learning model approach with external validation. Physica Medica 2021; 83: 64-71.","journal-title":"Physica Medica"},{"key":"ref-1157126","doi-asserted-by":"crossref","first-page":"2388","DOI":"10.1109\/JBHI.2021.3081355","article-title":"Interpretable and lightweight 3-D deep learning model for automated ACL diagnosis","volume":"25","author":"Jeon YS","year":"2021","unstructured":"Jeon YS, Yoshino K, Hagiwara S, et al. Interpretable and lightweight 3-D deep learning model for automated ACL diagnosis. IEEE J Biomed Health Inform 2021; 25: 2388-2397.","journal-title":"IEEE J Biomed Health Inform"},{"key":"ref-1157127","doi-asserted-by":"crossref","first-page":"1745","DOI":"10.1002\/jmri.27266","article-title":"Deep learning approach for anterior cruciate ligament lesion detection: evaluation of diagnostic performance using arthroscopy as the reference standard","volume":"52","author":"Zhang L","year":"2020","unstructured":"Zhang L, Li M, Zhou Y, et al. Deep learning approach for anterior cruciate ligament lesion detection: evaluation of diagnostic performance using arthroscopy as the reference standard. J Magn Reson Imaging 2020; 52: 1745-1752.","journal-title":"J Magn Reson Imaging"},{"key":"ref-1157128","doi-asserted-by":"crossref","first-page":"1207","DOI":"10.1007\/s00256-020-03410-2","article-title":"Deep convolutional neural network-based detection of meniscus tears: comparison with radiologists and surgery as standard of reference","volume":"49","author":"Fritz B","year":"2020","unstructured":"Fritz B, Marbach G, Civardi F, et al. Deep convolutional neural network-based detection of meniscus tears: comparison with radiologists and surgery as standard of reference. Skeletal Radiol 2020; 49: 1207-1217.","journal-title":"Skeletal Radiol"},{"key":"ref-1157129","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics11010105"},{"key":"ref-1157130","doi-asserted-by":"publisher","DOI":"10.3389\/fbioe.2021.747217"},{"key":"ref-1157131","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pmed.1002699"},{"key":"ref-1157132","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.2021200165"},{"key":"ref-1157133","doi-asserted-by":"crossref","first-page":"411","DOI":"10.1016\/j.ejrad.2005.12.034","article-title":"Value of fat-suppressed PD-weighted TSE-sequences for detection of anterior and posterior cruciate ligament lesions\u2013comparison to arthroscopy","volume":"58","author":"Schaefer FKW","year":"2006","unstructured":"Schaefer FKW, Schaefer PJ, Brossmann J, et al. Value of fat-suppressed PD-weighted TSE-sequences for detection of anterior and posterior cruciate ligament lesions\u2013comparison to arthroscopy. Eur J Radiol 2006; 58: 411-415.","journal-title":"Eur J Radiol"},{"key":"ref-1157134","doi-asserted-by":"crossref","first-page":"653","DOI":"10.1097\/RCT.0b013e3182251016","article-title":"Value of sagittal fatsuppressed proton-density fast-spin-echo of the knee joint as a limited protocol in evaluating internal knee derangements","volume":"35","author":"Khoury NJ","year":"2011","unstructured":"Khoury NJ, Mahfoud Z, Masrouha KZ, et al. Value of sagittal fatsuppressed proton-density fast-spin-echo of the knee joint as a limited protocol in evaluating internal knee derangements. J Comput Assist Tomogr 2011; 35: 653-661.","journal-title":"J Comput Assist Tomogr"},{"key":"ref-1157135","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-021-03150-w"},{"key":"ref-1157136","doi-asserted-by":"crossref","first-page":"159","DOI":"10.2307\/2529310","article-title":"The measurement of observer agreement for categorical data","volume":"33","author":"Landis JR","year":"1977","unstructured":"Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977; 33: 159-174.","journal-title":"Biometrics"},{"key":"ref-1157137","first-page":"357","article-title":"A deep learning system for synthetic knee magnetic resonance imaging. Is artificial intelligence-based fat-suppressed imaging feasible? Invest Radiol 2021","volume":"56","author":"Fayad LM","unstructured":"Fayad LM, Parekh VS, Luna RDC, et al. A deep learning system for synthetic knee magnetic resonance imaging. Is artificial intelligence-based fat-suppressed imaging feasible? Invest Radiol 2021 ; 56: 357-368.","journal-title":";"},{"key":"ref-1157138","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2003.00541"},{"key":"ref-1157139","first-page":"784","volume-title":"Proceedings of the Third Conference on Medical Imaging with Deep Learning","author":"Tsai CH","year":"2020","unstructured":"Tsai CH, Kiryati N, Konen E, et al. Knee injury detection using MRI with efficiently-layered network (ELNet). In: Tal A, Ismail Ben A, Marleen de B, Maxime D, Herve L, Christopher P (eds.). Proceedings of the Third Conference on Medical Imaging with Deep Learning. Proceedings of Machine Learning Research: PMLR; 2020, p. 784-794."},{"key":"ref-1157140","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.2019180091"},{"key":"ref-1157141","doi-asserted-by":"publisher","DOI":"10.1097\/MD.0000000000031510"},{"key":"ref-1157142","doi-asserted-by":"publisher","DOI":"10.1186\/s40001-022-00883-w"},{"key":"ref-1157143","doi-asserted-by":"publisher","DOI":"10.1186\/s12891-022-05468-6"},{"key":"ref-1157144","doi-asserted-by":"crossref","first-page":"486","DOI":"10.1148\/radiol.2523090028","article-title":"Knee joint: comprehensive assessment with 3D isotropic resolution fast spin-echo MR imaging\u2013diagnostic performance compared with that of conventional MR imaging at 3.0T","volume":"252","author":"Kijowski R","year":"2009","unstructured":"Kijowski R, Davis KW, Woods MA, et al. Knee joint: comprehensive assessment with 3D isotropic resolution fast spin-echo MR imaging\u2013diagnostic performance compared with that of conventional MR imaging at 3.0T. Radiology 2009; 252: 486-495.","journal-title":"Radiology"},{"key":"ref-1157145","doi-asserted-by":"crossref","first-page":"8394","DOI":"10.1007\/s00330-022-08923-z","article-title":"Deep learning to detect anterior cruciate ligament tear on knee MRI: multi-continental external vali-dation","volume":"32","author":"Tran A","year":"2022","unstructured":"Tran A, Lassalle L, Zille P, at al. Deep learning to detect anterior cruciate ligament tear on knee MRI: multi-continental external vali-dation. Eur Radiol 2022; 32: 8394-8403.","journal-title":"Eur Radiol"},{"key":"ref-1157146","doi-asserted-by":"crossref","first-page":"1353","DOI":"10.1007\/s00330-021-08190-4","article-title":"Deep learning in knee imaging: a systematic review utilizing a Checklist for Artificial Intelligence in Medical Imaging (CLAIM)","volume":"32","author":"Si L","year":"2022","unstructured":"Si L, Zhong J, Huo J, et al. Deep learning in knee imaging: a systematic review utilizing a Checklist for Artificial Intelligence in Medical Imaging (CLAIM). Eur Radiol 2022; 32: 1353-1361.","journal-title":"Eur Radiol"}],"container-title":["Polish Journal of Radiology"],"original-title":[],"link":[{"URL":"https:\/\/www.polradiol.com\/pdf-171854-109066","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.polradiol.com\/pdf-171854-109066","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,5,22]],"date-time":"2024-05-22T11:25:39Z","timestamp":1716377139000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.polradiol.com\/Feasibility-of-the-fat-suppression-image-subtraction-method-using-deep-learning-for,171854,0,2.html"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,12,8]]},"references-count":30,"alternative-id":["171854"],"URL":"https:\/\/doi.org\/10.5114\/pjr.2023.133660","relation":{},"ISSN":["1899-0967"],"issn-type":[{"value":"1899-0967","type":"print"}],"subject":[],"published":{"date-parts":[[2023,12,8]]}}}