{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,4]],"date-time":"2026-06-04T11:27:39Z","timestamp":1780572459023,"version":"3.54.1"},"reference-count":44,"publisher":"Wiley","issue":"2","license":[{"start":{"date-parts":[[2025,2,3]],"date-time":"2025-02-03T00:00:00Z","timestamp":1738540800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Magnetic Resonance Imaging"],"published-print":{"date-parts":[[2025,8]]},"abstract":"<jats:sec><jats:title>Background<\/jats:title><jats:p>Postencephalitic epilepsy (PEE) is a severe neurological complication following encephalitis. Early identification of individuals at high risk for PEE is important for timely intervention.<\/jats:p><\/jats:sec><jats:sec><jats:title>Purpose<\/jats:title><jats:p>To develop a large self\u2010supervised vision foundation model using a big dataset of multi\u2010contrast head MRI scans, followed by fine\u2010tuning with MRI data and follow\u2010up outcomes from patients with PEE to develop a PEE association model.<\/jats:p><\/jats:sec><jats:sec><jats:title>Study Type<\/jats:title><jats:p>Retrospective.<\/jats:p><\/jats:sec><jats:sec><jats:title>Population<\/jats:title><jats:p>Fifty\u2010seven thousand six hundred twenty\u2010one contrast\u2010enhanced head MRI scans from 34,871 patients for foundation model construction, and head MRI scans from 144 patients with encephalitis (64 PEE, 80 N\u2010PEE) for the PEE association model.<\/jats:p><\/jats:sec><jats:sec><jats:title>Field Strength\/Sequence<\/jats:title><jats:p>1.5\u2010T, 3\u2010T, T1\u2010weighted imaging, T2\u2010weighted imaging, fluid attenuated inversion recovery, T1\u2010weighted contrast\u2010enhanced imaging.<\/jats:p><\/jats:sec><jats:sec><jats:title>Assessment<\/jats:title><jats:p>The foundation model was developed using self\u2010supervised learning and cross\u2010contrast context recovery. Patients with encephalitis were monitored for a median of 3.7\u2009years (range 0.7\u20137.5\u2009years), with epilepsy diagnosed according to International League Against Epilepsy. Occlusion sensitivity mapping highlighted brain regions involved in PEE classifications. Model performance was compared with DenseNet without pre\u2010trained weights.<\/jats:p><\/jats:sec><jats:sec><jats:title>Statistical Tests<\/jats:title><jats:p>Performance was assessed via confusion matrices, accuracy, sensitivity, specificity, precision, F1 score, and area under the receiver operating characteristic curve (AUC). The DeLong test evaluated AUC between the two models (<jats:italic>P<\/jats:italic>\u2009&lt;\u20090.05 for statistical significance).<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>The PEE association model achieved accuracy, sensitivity, specificity, precision, F1 score, and AUC of 79.3% (95% CI: 0.71\u20130.92), 92.3% (95% CI: 0.80\u20131.00), 68.8% (95% CI: 0.55\u20130.87), 70.6% (95% CI: 0.61\u20130.90), 80.0% (95% CI: 0.71\u20130.93), and 81.0% (95% CI: 0.68\u20130.92), respectively. A significant AUC improvement was found compared to DenseNet (Delong test, <jats:italic>P<\/jats:italic>\u2009=\u20090.03). The association model focused on brain regions affected by encephalitis.<\/jats:p><\/jats:sec><jats:sec><jats:title>Data Conclusion<\/jats:title><jats:p>Using extensive unlabeled data via self\u2010supervised learning addressed the limitations of supervised tasks with limited data. The fine\u2010tuned foundation model outperformed DenseNet, which was trained exclusively on task data.<\/jats:p><\/jats:sec><jats:sec><jats:title>Plain Language Summary<\/jats:title><jats:p>This research develops a model to assess the occurrence epilepsy after encephalitis, a severe brain inflammation condition. By using over 57,000 brain scans, the study trains a computer program to recognize patterns in brain images. The model analyzes whole\u2010brain scans to identify areas commonly affected by the disease, such as the temporal and frontal lobes. It was tested on data from patients with encephalitis and showed better performance than older methods. The model can assess the risk of secondary epilepsy in patients with encephalitis, allowing doctors to intervene early and improve treatment outcomes for those affected by this condition.<\/jats:p><\/jats:sec><jats:sec><jats:title>Evidence Level<\/jats:title><jats:p>4<\/jats:p><\/jats:sec><jats:sec><jats:title>Technical Efficacy<\/jats:title><jats:p>Stage 1<\/jats:p><\/jats:sec>","DOI":"10.1002\/jmri.29734","type":"journal-article","created":{"date-parts":[[2025,2,3]],"date-time":"2025-02-03T07:25:12Z","timestamp":1738567512000},"page":"494-505","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":5,"title":["Associations of Postencephalitic Epilepsy Using Multi\u2010Contrast Whole Brain <scp>MRI<\/scp>: A Large Self\u2010Supervised Vision Foundation Model Strategy"],"prefix":"10.1002","volume":"62","author":[{"given":"Ronghui","family":"Gao","sequence":"first","affiliation":[{"name":"Department of Radiology, West China Hospital Sichuan University  Chengdu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anjiao","family":"Peng","sequence":"additional","affiliation":[{"name":"Department of Neurology, West China Hospital Sichuan University  Chengdu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yifei","family":"Duan","sequence":"additional","affiliation":[{"name":"Department of Neurology, West China Hospital Sichuan University  Chengdu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mengyao","family":"Chen","sequence":"additional","affiliation":[{"name":"Huaxi MR Research Center (HMRRC), West China Hospital Sichuan University  Chengdu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tao","family":"Zheng","sequence":"additional","affiliation":[{"name":"IT Center, West China Hospital Sichuan University  Chengdu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Meng","family":"Zhang","sequence":"additional","affiliation":[{"name":"NVIDIA Corp Beijing Representative Office  Beijing China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lei","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Neurology, West China Hospital Sichuan University  Chengdu China"},{"name":"Pazhou Lab  Guangzhou China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0687-6065","authenticated-orcid":false,"given":"Huaiqiang","family":"Sun","sequence":"additional","affiliation":[{"name":"Department of Radiology, West China Hospital Sichuan University  Chengdu China"},{"name":"Huaxi MR Research Center (HMRRC), West China Hospital Sichuan University  Chengdu China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2025,2,3]]},"reference":[{"key":"e_1_2_10_2_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00281-009-0176-1"},{"key":"e_1_2_10_3_1","doi-asserted-by":"publisher","DOI":"10.1542\/peds.2013-3010"},{"key":"e_1_2_10_4_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jneuroim.2023.578089"},{"key":"e_1_2_10_5_1","doi-asserted-by":"publisher","DOI":"10.1186\/s12974-018-1192-7"},{"key":"e_1_2_10_6_1","doi-asserted-by":"publisher","DOI":"10.1111\/epi.13965"},{"key":"e_1_2_10_7_1","doi-asserted-by":"publisher","DOI":"10.1111\/epi.13253"},{"key":"e_1_2_10_8_1","doi-asserted-by":"publisher","DOI":"10.1111\/epi.12879"},{"key":"e_1_2_10_9_1","doi-asserted-by":"publisher","DOI":"10.1177\/7010.2006.00223"},{"key":"e_1_2_10_10_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.clineuro.2023.107854"},{"key":"e_1_2_10_11_1","doi-asserted-by":"publisher","DOI":"10.3389\/fneur.2018.01136"},{"key":"e_1_2_10_12_1","doi-asserted-by":"publisher","DOI":"10.1136\/jnnp.2003.034280"},{"key":"e_1_2_10_13_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-023-00811-0"},{"key":"e_1_2_10_14_1","doi-asserted-by":"crossref","unstructured":"HeK ChenX XieS LiY Doll\u00e1rP GirshickR editors.Masked autoencoders are scalable vision learners. Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition2022.","DOI":"10.1109\/CVPR52688.2022.01553"},{"key":"e_1_2_10_15_1","doi-asserted-by":"publisher","DOI":"10.1093\/cid\/cit458"},{"key":"e_1_2_10_16_1","doi-asserted-by":"publisher","DOI":"10.1111\/epi.12550"},{"key":"e_1_2_10_17_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10916-023-02017-z"},{"key":"e_1_2_10_18_1","unstructured":"DosovitskiyAJ.An image is worth 16\u2009\u00d7\u200916 words: Transformers for image recognition at scale. arXiv preprint arXiv: 2010.119292020."},{"key":"e_1_2_10_19_1","unstructured":"LoshchilovI HutterF.Fixing weight decay regularization in adam. ArXiv preprint arXiv2017."},{"key":"e_1_2_10_20_1","doi-asserted-by":"crossref","unstructured":"HuangG LiuZ Van Der MaatenL WeinbergerKQ editors.Densely connected convolutional networks. Proceedings of the IEEE conference on computer vision and pattern recognition2017.","DOI":"10.1109\/CVPR.2017.243"},{"key":"e_1_2_10_21_1","doi-asserted-by":"crossref","unstructured":"ZeilerM editor.Visualizing and understanding convolutional networks. European conference on computer vision\/arXiv2014.","DOI":"10.1007\/978-3-319-10590-1_53"},{"key":"e_1_2_10_22_1","doi-asserted-by":"publisher","DOI":"10.1109\/SSCI.2018.8628742"},{"key":"e_1_2_10_23_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-33128-3_1"},{"key":"e_1_2_10_24_1","first-page":"28","article-title":"Transfer learning in medical image classification: Challenges and opportunities","volume":"5","author":"Godasu R","year":"2020","journal-title":"Epilepsia"},{"key":"e_1_2_10_25_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2020.104115"},{"key":"e_1_2_10_26_1","unstructured":"RaghuM ZhangC KleinbergJ BengioS.Transfusion: Understanding transfer learning for medical imaging.2019. 32."},{"key":"e_1_2_10_27_1","doi-asserted-by":"crossref","unstructured":"HajaA Van Der WoudeB SchomakerL editors.Organoids segmentation using self\u2010supervised learning: How complex should the pretext task be?Proceedings of the 2023 10th International Conference on Biomedical and Bioinformatics Engineering2023.","DOI":"10.1145\/3637732.3637772"},{"key":"e_1_2_10_28_1","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2021.3134634"},{"key":"e_1_2_10_29_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2019.101539"},{"key":"e_1_2_10_30_1","first-page":"1","article-title":"MaeFE: Masked autoencoders family of electrocardiogram for self\u2010supervised pretraining and transfer learning","volume":"72","author":"Zhang H","year":"2023","journal-title":"IEEE Trans Instrum Meas"},{"key":"e_1_2_10_31_1","doi-asserted-by":"crossref","unstructured":"XieR PangK BaderGD WangB editors.Maester: Masked autoencoder guided segmentation at pixel resolution for accurate self\u2010supervised subcellular structure recognition. Proc IEEE Conf Comput Vis Pattern Recogn2023.","DOI":"10.1109\/CVPR52729.2023.00321"},{"key":"e_1_2_10_32_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jalz.2019.02.007"},{"key":"e_1_2_10_33_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2022.103565"},{"key":"e_1_2_10_34_1","unstructured":"BanerjeeS MitraS MasulliF RovettaSJ.Deep radiomics for brain tumor detection and classification from multi\u2010sequence MRI. arXiv preprint arXiv:1903.092402019."},{"key":"e_1_2_10_35_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.nic.2011.01.011"},{"key":"e_1_2_10_36_1","doi-asserted-by":"publisher","DOI":"10.1002\/jmri.22830"},{"key":"e_1_2_10_37_1","doi-asserted-by":"publisher","DOI":"10.1109\/JBHI.2019.2891049"},{"key":"e_1_2_10_38_1","unstructured":"YangS LiuX ZhengZ WangW MaXJ.Fusing medical image features and clinical features with deep learning for computer\u2010aided diagnosis. arXiv preprint arXiv:2103.058552021."},{"key":"e_1_2_10_39_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0139974"},{"key":"e_1_2_10_40_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.seizure.2022.05.017"},{"key":"e_1_2_10_41_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.seizure.2020.12.004"},{"key":"e_1_2_10_42_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.seizure.2018.11.005"},{"key":"e_1_2_10_43_1","first-page":"2019","article-title":"Seizures and epilepsy in herpes simplex virus encephalitis: Current concepts and future directions of pathogenesis and management","volume":"259","author":"Sellner J","year":"2012","journal-title":"Epilepsia"},{"key":"e_1_2_10_44_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.expneurol.2011.09.033"},{"key":"e_1_2_10_45_1","doi-asserted-by":"publisher","DOI":"10.1684\/j.1950-6945.2000.tb00375.x"}],"container-title":["Journal of Magnetic Resonance Imaging"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/jmri.29734","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,29]],"date-time":"2025-09-29T16:24:55Z","timestamp":1759163095000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/jmri.29734"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,2,3]]},"references-count":44,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2025,8]]}},"alternative-id":["10.1002\/jmri.29734"],"URL":"https:\/\/doi.org\/10.1002\/jmri.29734","archive":["Portico"],"relation":{},"ISSN":["1053-1807","1522-2586"],"issn-type":[{"value":"1053-1807","type":"print"},{"value":"1522-2586","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,2,3]]},"assertion":[{"value":"2024-10-17","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-01-23","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-02-03","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}