{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,23]],"date-time":"2026-01-23T00:26:23Z","timestamp":1769127983383,"version":"3.49.0"},"reference-count":33,"publisher":"Oxford University Press (OUP)","issue":"2","license":[{"start":{"date-parts":[[2025,4,4]],"date-time":"2025-04-04T00:00:00Z","timestamp":1743724800000},"content-version":"vor","delay-in-days":34,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"publisher","award":["R01AR078762"],"award-info":[{"award-number":["R01AR078762"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,3,19]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:sec>\n                  <jats:title>Background<\/jats:title>\n                  <jats:p>Accurate interpretation of meniscal anomalies on knee MRI is critical for diagnosis and treatment planning, with artificial intelligence emerging as a promising tool to support and enhance this process through automated anomaly detection.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Purpose<\/jats:title>\n                  <jats:p>To evaluate the impact of an artificial intelligence (AI) anomaly detection assistant on radiologists\u2019 interpretation of meniscal anomalies in undersampled, deep learning (DL)-reconstructed knee MRI and assess the relationship between reconstruction quality metrics and anomaly detection performance.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Materials and Methods<\/jats:title>\n                  <jats:p>This retrospective study included 947 knee MRI examinations; 51 were excluded for poor image quality, leaving 896 participants (mean age, 44.7\u2009\u00b1\u200915.3\u2009years; 472 women). Using 8-fold undersampled data, DL-based reconstructed images were generated. An object detection model was trained on original, fully sampled images and evaluated on 1 original and 14 DL-reconstructed test sets to identify meniscal lesions. Standard reconstruction metrics (normalized root mean square error, peak signal-to-noise ratio, and structural similarity index) and anomaly detection metrics (mean average precision, F1 score) were quantified and compared. Two radiologists independently reviewed a stratified sample of 50 examinations unassisted and assisted with AI-predicted anomaly boxes. McNemar\u2019s test evaluated differences in diagnostic performance; Cohen\u2019s kappa assessed interrater agreement.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>On the original images, the anomaly detection model achieved the following: 70.53% precision, 72.17% recall, 63.09% mAP, and a 71.34% F1 score. Comparing performance among the undersampled reconstruction datasets, box-based reconstruction metrics showed better correlation with detection performance than traditional image-based metrics (mAP to box-based SSIM, r\u2009=\u20090.81, P\u2009&amp;lt;\u2009.01; mAP to image-based SSIM, r\u2009=\u20090.64, P\u2009=\u2009.01). In 50 participants, AI assistance improved radiologists\u2019 accuracy on reconstructed images. Sensitivity increased from 77.27% (95% CI, 65.83-85.72; 51\/66) to 80.30% (95% CI, 69.16-88.11; 53\/66), and specificity improved from 88.46% (95% CI, 83.73-91.95; 207\/234) to 90.60% (95% CI, 86.18-93.71; 212\/234) (P &amp;lt; .05).<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Conclusion<\/jats:title>\n                  <jats:p>AI-assisted meniscal anomaly detection enhanced radiologists\u2019 interpretation of undersampled, DL-reconstructed knee MRI. Anomaly detection may serve as a complementary tool alongside other reconstruction metrics to assess the preservation of clinically important features in reconstructed images, warranting further investigation.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1093\/radadv\/umaf015","type":"journal-article","created":{"date-parts":[[2025,4,4]],"date-time":"2025-04-04T19:59:05Z","timestamp":1743796745000},"source":"Crossref","is-referenced-by-count":3,"title":["Improving radiologist detection of meniscal abnormality on undersampled, deep learning reconstructed knee MRI"],"prefix":"10.1093","volume":"2","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-5937-0194","authenticated-orcid":false,"given":"Natalia","family":"Konovalova","sequence":"first","affiliation":[{"name":"Radiology and Biomedical Imaging Department, University of California, San Francisco , San Francisco, CA,","place":["United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Aniket","family":"Tolpadi","sequence":"additional","affiliation":[{"name":"Radiology and Biomedical Imaging Department, University of California, San Francisco , San Francisco, CA,","place":["United States"]},{"name":"Bioengineering Department, University of California, Berkeley , Berkeley, CA,","place":["United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Felix","family":"Liu","sequence":"additional","affiliation":[{"name":"Radiology and Biomedical Imaging Department, University of California, San Francisco , San Francisco, CA,","place":["United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6483-3381","authenticated-orcid":false,"given":"Zehra","family":"Akkaya","sequence":"additional","affiliation":[{"name":"Radiology and Biomedical Imaging Department, University of California, San Francisco , San Francisco, CA,","place":["United States"]},{"name":"Faculty of Medicine, Radiology Department, Ankara University , Ankara,","place":["Turkey"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Johanna","family":"Luitjens","sequence":"additional","affiliation":[{"name":"Radiology and Biomedical Imaging Department, University of California, San Francisco , San Francisco, CA,","place":["United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Felix","family":"Gassert","sequence":"additional","affiliation":[{"name":"Radiology and Biomedical Imaging Department, University of California, San Francisco , San Francisco, CA,","place":["United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Paula","family":"Giesler","sequence":"additional","affiliation":[{"name":"Radiology and Biomedical Imaging Department, University of California, San Francisco , San Francisco, CA,","place":["United States"]},{"name":"Faculty of Medicine, University of Freiburg Medical Center , Freiburg,","place":["Germany"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Rupsa","family":"Bhattacharjee","sequence":"additional","affiliation":[{"name":"Radiology and Biomedical Imaging Department, University of California, San Francisco , San Francisco, CA,","place":["United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Misung","family":"Han","sequence":"additional","affiliation":[{"name":"Radiology and Biomedical Imaging Department, University of California, San Francisco , San Francisco, CA,","place":["United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Emma","family":"Bahroos","sequence":"additional","affiliation":[{"name":"Radiology and Biomedical Imaging Department, University of California, San Francisco , San Francisco, CA,","place":["United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0201-871X","authenticated-orcid":false,"given":"Sharmila","family":"Majumdar","sequence":"additional","affiliation":[{"name":"Radiology and Biomedical Imaging Department, University of California, San Francisco , San Francisco, CA,","place":["United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Valentina","family":"Pedoia","sequence":"additional","affiliation":[{"name":"Radiology and Biomedical Imaging Department, University of California, San Francisco , San Francisco, CA,","place":["United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2025,4,4]]},"reference":[{"key":"2025042501063080300_umaf015-B1","doi-asserted-by":"crossref","first-page":"495","DOI":"10.1146\/annurev-bioeng-060418-052547","article-title":"The meniscus in normal and osteoarthritic tissues: facing the structure property challenges and current treatment trends","volume":"21","author":"Murphy","year":"2019","journal-title":"Annu Rev Biomed Eng"},{"issue":"2","key":"2025042501063080300_umaf015-B2","doi-asserted-by":"crossref","first-page":"307","DOI":"10.1016\/j.mric.2021.11.009","article-title":"MRI of the knee meniscus","volume":"30","author":"Trunz","year":"2022","journal-title":"Magn Reson Imaging Clin N Am"},{"issue":"6","key":"2025042501063080300_umaf015-B3","doi-asserted-by":"crossref","first-page":"3055","DOI":"10.1002\/mrm.26977","article-title":"Learning a variational network for reconstruction of accelerated MRI data","volume":"79","author":"Hammernik","year":"2018","journal-title":"Magn Reson Med"},{"issue":"5","key":"2025042501063080300_umaf015-B4","doi-asserted-by":"crossref","first-page":"564","DOI":"10.1111\/1754-9485.13276","article-title":"Deep learning in magnetic resonance image reconstruction","volume":"65","author":"Chandra","year":"2021","journal-title":"J Med Imaging Radiat Oncol"},{"key":"2025042501063080300_umaf015-B5","doi-asserted-by":"crossref","first-page":"107452","DOI":"10.1016\/j.cmpb.2023.107452","article-title":"Deep learning based MRI reconstruction with transformer","volume":"233","author":"Wu","year":"2023","journal-title":"Comput Methods Programs Biomed"},{"issue":"5","key":"2025042501063080300_umaf015-B6","doi-asserted-by":"crossref","first-page":"1853","DOI":"10.1109\/TMI.2024.3351702","article-title":"High-frequency space diffusion model for accelerated MRI","volume":"43","author":"Cao","year":"2024","journal-title":"IEEE Trans Med Imaging"},{"issue":"3","key":"2025042501063080300_umaf015-B7","doi-asserted-by":"crossref","first-page":"e230181","DOI":"10.1148\/ryai.230181","article-title":"Impact of deep learning image reconstruction methods on MRI throughput","volume":"6","author":"Yang","year":"2024","journal-title":"Radiol Artif Intell"},{"issue":"3","key":"2025042501063080300_umaf015-B8","doi-asserted-by":"crossref","first-page":"335","DOI":"10.1007\/s10334-024-01173-8","article-title":"Deep learning for accelerated and robust MRI reconstruction","volume":"37","author":"Heckel","year":"2024","journal-title":"MAGMA"},{"issue":"3","key":"2025042501063080300_umaf015-B9","doi-asserted-by":"crossref","first-page":"e219001","DOI":"10.1148\/ryai.2021219001","article-title":"Automatic deep learning-assisted detection and grading of abnormalities in knee MRI studies","volume":"3","author":"Astuto","year":"2021","journal-title":"Radiol Artif Intell"},{"issue":"4","key":"2025042501063080300_umaf015-B10","doi-asserted-by":"crossref","first-page":"243","DOI":"10.1016\/j.diii.2019.02.007","article-title":"Artificial intelligence to diagnose meniscus tears on MRI","volume":"100","author":"Roblot","year":"2019","journal-title":"Diagn Interv Imaging"},{"issue":"6","key":"2025042501063080300_umaf015-B11","doi-asserted-by":"crossref","first-page":"1137","DOI":"10.1109\/TPAMI.2016.2577031","article-title":"Faster R-CNN: towards real-time object detection with region proposal networks","volume":"39","author":"Ren","year":"2017","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"2025042501063080300_umaf015-B12","first-page":"779","author":"Redmon"},{"issue":"1","key":"2025042501063080300_umaf015-B13","doi-asserted-by":"crossref","first-page":"e230764","DOI":"10.1148\/radiol.230764","article-title":"How AI may transform musculoskeletal imaging","volume":"310","author":"Guermazi","year":"2024","journal-title":"Radiology"},{"key":"2025042501063080300_umaf015-B14","first-page":"113","article-title":"Breaking speed limits with simultaneous ultra-fast MRI reconstruction and tissue segmentation","author":"Caliva","year":"2020"},{"issue":"2","key":"2025042501063080300_umaf015-B15","doi-asserted-by":"crossref","first-page":"267","DOI":"10.3390\/bioengineering10020267","article-title":"K2S challenge: from undersampled K-space to automatic segmentation","volume":"10","author":"Tolpadi","year":"2023","journal-title":"Bioengineering"},{"issue":"32","key":"2025042501063080300_umaf015-B16","doi-asserted-by":"crossref","first-page":"8352","DOI":"10.1364\/AO.36.008352","article-title":"Invariant error metrics for image reconstruction","volume":"36","author":"Fienup","year":"1997","journal-title":"Appl Opt"},{"issue":"1","key":"2025042501063080300_umaf015-B17","doi-asserted-by":"crossref","first-page":"12","DOI":"10.1049\/iet-ipr.2012.0489","article-title":"Is there a relationship between peak-signal-to-noise ratio and structural similarity index measure?","volume":"7","author":"Hor\u00e9","year":"2013","journal-title":"IET Image Process"},{"issue":"1","key":"2025042501063080300_umaf015-B18","doi-asserted-by":"crossref","first-page":"81","DOI":"10.1007\/s11760-009-0144-1","article-title":"A comprehensive assessment of the structural similarity index","volume":"5","author":"Dosselmann","year":"2011","journal-title":"Signal Image Video Process"},{"issue":"6","key":"2025042501063080300_umaf015-B19","doi-asserted-by":"crossref","first-page":"3054","DOI":"10.1002\/mrm.28338","article-title":"Advancing machine learning for MR image reconstruction with an open competition: overview of the 2019 fastMRI challenge","volume":"84","author":"Knoll","year":"2020","journal-title":"Magn Reson Med"},{"issue":"2","key":"2025042501063080300_umaf015-B20","doi-asserted-by":"crossref","first-page":"382","DOI":"10.1002\/mrm.21481","article-title":"Comparison of reconstruction accuracy and efficiency among autocalibrating data-driven parallel imaging methods","volume":"59","author":"Brau","year":"2008","journal-title":"Magn Reson Med"},{"key":"2025042501063080300_umaf015-B21","first-page":"566","article-title":"Learning data augmentation strategies for object detection","author":"Zoph","year":"2020"},{"key":"2025042501063080300_umaf015-B22","author":"Tolpadi"},{"issue":"5","key":"2025042501063080300_umaf015-B23","doi-asserted-by":"crossref","first-page":"2188","DOI":"10.1002\/mrm.27201","article-title":"KIKI-net: cross-domain convolutional neural networks for reconstructing undersampled magnetic resonance images","volume":"80","author":"Eo","year":"2018","journal-title":"Magn Reson Med"},{"issue":"5","key":"2025042501063080300_umaf015-B24","doi-asserted-by":"crossref","first-page":"1391","DOI":"10.2214\/ajr.184.5.01841391","article-title":"Reader agreement studies","volume":"184","author":"Crewson","year":"2005","journal-title":"Am J Roentgenol"},{"issue":"5","key":"2025042501063080300_umaf015-B25","doi-asserted-by":"crossref","first-page":"323","DOI":"10.1037\/h0028106","article-title":"Large sample standard errors of kappa and weighted kappa","volume":"72","author":"Fleiss","year":"1969","journal-title":"Psychol Bull"},{"key":"2025042501063080300_umaf015-B26","doi-asserted-by":"publisher","author":"Zhao","DOI":"10.48550\/arXiv.2109.11524"},{"issue":"4","key":"2025042501063080300_umaf015-B27","doi-asserted-by":"crossref","DOI":"10.3390\/jcm10040606","article-title":"Diagnostic accuracy of magnetic resonance imaging in the detection of type and location of meniscus tears: comparison with arthroscopic findings","volume":"10","author":"Kim","year":"2021","journal-title":"J Clin Med"},{"issue":"4","key":"2025042501063080300_umaf015-B28","doi-asserted-by":"crossref","first-page":"571","DOI":"10.1007\/s10278-019-00232-0","article-title":"RIL-Contour: a medical imaging dataset annotation tool for and with deep learning","volume":"32","author":"Philbrick","year":"2019","journal-title":"J Digit Imaging"},{"issue":"4","key":"2025042501063080300_umaf015-B29","doi-asserted-by":"crossref","first-page":"e250","DOI":"10.1016\/S2589-7500(21)00041-8","article-title":"Deep learning-based artificial intelligence model to assist thyroid nodule diagnosis and management: a multicentre diagnostic study","volume":"3","author":"Peng","year":"2021","journal-title":"Lancet Digit Health"},{"issue":"12","key":"2025042501063080300_umaf015-B30","doi-asserted-by":"crossref","first-page":"1985","DOI":"10.1007\/s00234-021-02697-9","article-title":"Deep learning assistance increases the detection sensitivity of radiologists for secondary intracranial aneurysms in subarachnoid hemorrhage","volume":"63","author":"Pennig","year":"2021","journal-title":"Neuroradiology"},{"issue":"3","key":"2025042501063080300_umaf015-B31","doi-asserted-by":"crossref","first-page":"552","DOI":"10.2214\/AJR.17.18228","article-title":"Advanced imaging techniques in the knee: benefits and limitations of new rapid acquisition strategies for routine knee MRI","volume":"209","author":"Garwood","year":"2017","journal-title":"Am J Roentgenol"},{"issue":"10","key":"2025042501063080300_umaf015-B32","doi-asserted-by":"crossref","first-page":"2804","DOI":"10.1109\/TMI.2022.3180228","article-title":"Multimodal transformer for accelerated MR imaging","volume":"42","author":"Feng","year":"2023","journal-title":"IEEE Trans Med Imaging"},{"key":"2025042501063080300_umaf015-B33","first-page":"280","author":"Johnson"}],"container-title":["Radiology Advances"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/radadv\/advance-article-pdf\/doi\/10.1093\/radadv\/umaf015\/62871844\/umaf015.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/radadv\/article-pdf\/2\/2\/umaf015\/62871844\/umaf015.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/radadv\/article-pdf\/2\/2\/umaf015\/62871844\/umaf015.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,4,25]],"date-time":"2025-04-25T05:06:43Z","timestamp":1745557603000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/radadv\/article\/doi\/10.1093\/radadv\/umaf015\/8106504"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3]]},"references-count":33,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2025,3,19]]}},"URL":"https:\/\/doi.org\/10.1093\/radadv\/umaf015","relation":{},"ISSN":["2976-9337"],"issn-type":[{"value":"2976-9337","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2025,3]]},"published":{"date-parts":[[2025,3]]},"article-number":"umaf015"}}