{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,3]],"date-time":"2026-03-03T14:40:56Z","timestamp":1772548856110,"version":"3.50.1"},"reference-count":49,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2023,1,18]],"date-time":"2023-01-18T00:00:00Z","timestamp":1674000000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100006095","name":"Helse S\u00f8r-\u00d8st RHF","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100006095","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neuroinform."],"abstract":"<jats:sec><jats:title>Introduction<\/jats:title><jats:p>Management of patients with brain metastases is often based on manual lesion detection and segmentation by an expert reader. This is a time- and labor-intensive process, and to that end, this work proposes an end-to-end deep learning segmentation network for a varying number of available MRI available sequences.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>We adapt and evaluate a 2.5D and a 3D convolution neural network trained and tested on a retrospective multinational study from two independent centers, in addition, nnU-Net was adapted as a comparative benchmark. Segmentation and detection performance was evaluated by: (1) the dice similarity coefficient, (2) a per-metastases and the average detection sensitivity, and (3) the number of false positives.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>The 2.5D and 3D models achieved similar results, albeit the 2.5D model had better detection rate, whereas the 3D model had fewer false positive predictions, and nnU-Net had fewest false positives, but with the lowest detection rate. On MRI data from center 1, the 2.5D, 3D, and nnU-Net detected 79%, 71%, and 65% of all metastases; had an average per patient sensitivity of 0.88, 0.84, and 0.76; and had on average 6.2, 3.2, and 1.7 false positive predictions per patient, respectively. For center 2, the 2.5D, 3D, and nnU-Net detected 88%, 86%, and 78% of all metastases; had an average per patient sensitivity of 0.92, 0.91, and 0.85; and had on average 1.0, 0.4, and 0.1 false positive predictions per patient, respectively.<\/jats:p><\/jats:sec><jats:sec><jats:title>Discussion\/Conclusion<\/jats:title><jats:p>Our results show that deep learning can yield highly accurate segmentations of brain metastases with few false positives in multinational data, but the accuracy degrades for metastases with an area smaller than 0.4 cm<jats:sup>2<\/jats:sup>.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fninf.2022.1056068","type":"journal-article","created":{"date-parts":[[2023,1,18]],"date-time":"2023-01-18T05:20:32Z","timestamp":1674019232000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":27,"title":["2.5D and 3D segmentation of brain metastases with deep learning on multinational MRI data"],"prefix":"10.3389","volume":"16","author":[{"given":"Jon Andr\u00e9","family":"Ottesen","sequence":"first","affiliation":[]},{"given":"Darvin","family":"Yi","sequence":"additional","affiliation":[]},{"given":"Elizabeth","family":"Tong","sequence":"additional","affiliation":[]},{"given":"Michael","family":"Iv","sequence":"additional","affiliation":[]},{"given":"Anna","family":"Latysheva","sequence":"additional","affiliation":[]},{"given":"Cathrine","family":"Saxhaug","sequence":"additional","affiliation":[]},{"given":"Kari Dolven","family":"Jacobsen","sequence":"additional","affiliation":[]},{"given":"\u00c5slaug","family":"Helland","sequence":"additional","affiliation":[]},{"given":"Kyrre Eeg","family":"Emblem","sequence":"additional","affiliation":[]},{"given":"Daniel L.","family":"Rubin","sequence":"additional","affiliation":[]},{"given":"Atle","family":"Bj\u00f8rnerud","sequence":"additional","affiliation":[]},{"given":"Greg","family":"Zaharchuk","sequence":"additional","affiliation":[]},{"given":"Endre","family":"Gr\u00f8vik","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2023,1,18]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"683","DOI":"10.1109\/ISBI.2019.8759329","article-title":"A Novel Focal Tversky loss function with improved Attention U-Net for lesion segmentation","volume":"2019","author":"Abraham","year":"2018","journal-title":"Proc. Int. Symp. Biomed. Imaging"},{"key":"B2","doi-asserted-by":"publisher","first-page":"R97","DOI":"10.1088\/0031-9155\/58\/13\/R97","article-title":"A survey of MRI-based medical image analysis for brain tumor studies","volume":"58","author":"Bauer","year":"2013","journal-title":"Phys. Med. Biol"},{"key":"B3","doi-asserted-by":"publisher","first-page":"49","DOI":"10.1097\/RLI.0b013e3181c02ed5","article-title":"Intra- and interobserver variability of linear and volumetric measurements of brain metastases using contrast-enhanced magnetic resonance imaging","volume":"45","author":"Bauknecht","year":"2010","journal-title":"Invest. Radiol"},{"key":"B4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13014-020-01514-6","article-title":"Deep convolutional neural networks for automated segmentation of brain metastases trained on clinical data","volume":"15","author":"Bousabarah","year":"2020","journal-title":"Radiat. Oncol"},{"key":"B5","doi-asserted-by":"publisher","first-page":"015003","DOI":"10.1088\/1361-6560\/abca53","article-title":"Automatic detection and segmentation of multiple brain metastases on magnetic resonance image using asymmetric UNet architecture","volume":"66","author":"Cao","year":"2021","journal-title":"Phys. Med. Biol"},{"key":"B6","doi-asserted-by":"publisher","first-page":"43","DOI":"10.1016\/j.compbiomed.2018.02.004","article-title":"Automatic detection and segmentation of brain metastases on multimodal MR images with a deep convolutional neural network","volume":"95","author":"Charron","year":"2018","journal-title":"Comput. Biol. Med"},{"key":"B7","doi-asserted-by":"publisher","first-page":"4314","DOI":"10.3389\/fonc.2021.739639","article-title":"Deep learning-based computer-aided detection system for automated treatment response assessment of brain metastases on 3D MRI","volume":"11","author":"Cho","year":"2021","journal-title":"Front. Oncol"},{"key":"B8","author":"Consortium","year":"2022","journal-title":"MONAI: Medical Open Network"},{"key":"B9","unstructured":"CordtsM.\n            OmranM.\n            RamosS.\n            RehfeldT.\n            EnzweilerM.\n            BenensonR.\n          The Cityscapes Dataset for Semantic Urban Scene Understanding,\u201d in Proceedings of the IEEE conference on computer vision and pattern recognition, 3213\u201332232016"},{"key":"B10","doi-asserted-by":"publisher","first-page":"2883","DOI":"10.1109\/JBHI.2020.2982103","article-title":"Automated Brain Metastases Detection Framework for T1-Weighted Contrast-Enhanced 3D MRI","volume":"24","author":"Dikici","year":"2020","journal-title":"IEEE J. Biomed. Heal. Informa."},{"key":"B11","unstructured":"DosovitskiyA.\n            BeyerL.\n            KolesnikovA.\n            WeissenbornD.\n            ZhaiX.\n            UnterthinerT.\n          An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale2020"},{"key":"B12","unstructured":"\u201cRich feature hierarchies for accurate object detection and semantic segmentation,\u201d2328\n            GirshickR.\n            DonahueJ.\n            DarrellT.\n            MalikJ.\n          Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition2013"},{"key":"B13","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41746-021-00398-4","article-title":"Handling missing MRI sequences in deep learning segmentation of brain metastases: a multicenter study","volume":"4","author":"Gr\u00f8vik","year":"2021","journal-title":"npJ. Digit. Med. 2021 41"},{"key":"B14","doi-asserted-by":"publisher","first-page":"175","DOI":"10.1002\/jmri.26766","article-title":"Deep learning enables automatic detection and segmentation of brain metastases on multisequence MRI","volume":"51","author":"Gr\u00f8vik","year":"2020","journal-title":"J. Magn. Reson. Imaging"},{"key":"B15","unstructured":"\u201cDeep residual learning for image recognition,\u201d2730\n            HeK.\n            ZhangX.\n            RenS.\n            SunJ.\n          32166560Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition2016"},{"key":"B16","doi-asserted-by":"publisher","first-page":"66","DOI":"10.1088\/1361-6560\/ac1835","article-title":"Automatic segmentation of brain metastases using T1 magnetic resonance and computed tomography images","author":"Hsu","year":"2021","journal-title":"Phys. Med. Biol"},{"key":"B17","doi-asserted-by":"publisher","first-page":"203","DOI":"10.1038\/s41592-020-01008-z","article-title":"nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation","volume":"18","author":"Isensee","year":"","journal-title":"Nat. Methods 2020 182"},{"key":"B18","doi-asserted-by":"publisher","first-page":"118","DOI":"10.1007\/978-3-030-72087-2_11","article-title":"nnU-Net for Brain Tumor Segmentation","author":"Isensee","year":"","journal-title":"Lect. Notes Comput. Sci."},{"key":"B19","doi-asserted-by":"publisher","first-page":"4952","DOI":"10.1002\/hbm.2475","article-title":"Automated brain extraction of multisequence MRI using artificial neural networks","volume":"40","author":"Isensee","year":"2019","journal-title":"Hum. Brain. Mapp"},{"key":"B20","doi-asserted-by":"publisher","first-page":"58","DOI":"10.1007\/978-3-030-72087-2_6","article-title":"H2NF-Net for Brain Tumor Segmentation using Multimodal MR Imaging: 2nd Place Solution to BraTS Challenge 2020 Segmentation Task","author":"Jia","year":"2020","journal-title":"Lect. Notes Comput. Sci."},{"key":"B21","doi-asserted-by":"publisher","first-page":"337","DOI":"10.1016\/S1042-3680(18)30365-6","article-title":"Demographics of Brain Metastasis","volume":"7","author":"Johnson","year":"1996","journal-title":"Neurosurg. Clin. N. Am"},{"key":"B22","doi-asserted-by":"publisher","first-page":"1608","DOI":"10.1002\/jmri.27741","article-title":"Fully automated mr detection and segmentation of brain metastases in non-small cell lung cancer using deep learning","volume":"54","author":"J\u00fcnger","year":"2021","journal-title":"J. Magn. Reson. Imaging"},{"key":"B23","doi-asserted-by":"publisher","first-page":"61","DOI":"10.1016\/j.media.2016.10.004","article-title":"Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation","volume":"36","author":"Kamnitsas","year":"2017","journal-title":"Med. Image Anal"},{"key":"B24","doi-asserted-by":"publisher","first-page":"84","DOI":"10.1145\/3065386","volume":"60","author":"Krizhevsky","year":"2017","journal-title":"ImageNet classification with deep convolutional neural networks. Commun ACM"},{"key":"B25","doi-asserted-by":"publisher","first-page":"e270","DOI":"10.1016\/S1470-2045(15)70057-4","article-title":"Response assessment criteria for brain metastases: proposal from the RANO group","volume":"16","author":"Lin","year":"2015","journal-title":"Lancet Oncol"},{"key":"B26","doi-asserted-by":"publisher","first-page":"3475","DOI":"10.1200\/JCO.2015.60.9503","article-title":"Treatment of brain metastases","volume":"33","author":"Lin","year":"2015","journal-title":"J. Clin. Oncol"},{"key":"B27","unstructured":"LoshchilovI.\n            HutterF.\n          SGDR: Stochastic Gradient Descent with Warm Restarts. 5th Int Conf Learn Represent ICLR 2017 - Conf Track Proc2016"},{"key":"B28","unstructured":"Decoupled Weight Decay Regularization. 7th Int. Conf. Learn. Represent\n            LoshchilovI.\n            HutterF.\n          ICLR 20192017"},{"key":"B29","doi-asserted-by":"publisher","first-page":"102035","DOI":"10.1016\/j.media.2021.102035","article-title":"Loss odyssey in medical image segmentation","volume":"71","author":"Ma","year":"2021","journal-title":"Med. Image Anal"},{"key":"B30","doi-asserted-by":"publisher","first-page":"1993","DOI":"10.1109\/TMI.2014.2377694","article-title":"The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS)","volume":"34","author":"Menze","year":"2015","journal-title":"IEEE Trans. Med. Imaging"},{"key":"B31","unstructured":"V-Net: Fully convolutional neural networks for volumetric medical image segmentation565571\n            MilletariF.\n            NavabN.\n            AhmadiS. A.\n          Proc - 2016 4th Int Conf 3D Vision, 3DV2016"},{"key":"B32","doi-asserted-by":"publisher","first-page":"655","DOI":"10.3174\/ajnr.A6982","article-title":"Automated Detection and Segmentation of Brain Metastases in Malignant Melanoma: Evaluation of a Dedicated Deep Learning Model","volume":"42","author":"Pennig","year":"2021","journal-title":"Am J. Neuroradiol"},{"key":"B33","unstructured":"ReddiS. J.\n            KaleS.\n            KumarS.\n          26586620On the Convergence of Adam and Beyond2018"},{"key":"B34","doi-asserted-by":"crossref","unstructured":"779788\n            RedmonJ.\n            DivvalaS.\n            GirshickR.\n            FarhadiA.\n          You Only Look Once: Unified, Real-Time Object Detection2016","DOI":"10.1109\/CVPR.2016.91"},{"key":"B35","doi-asserted-by":"publisher","first-page":"234","DOI":"10.1007\/978-3-319-24574-4_28","article-title":"U-Net: Convolutional Networks for Biomedical Image Segmentation","volume":"9351","author":"Ronneberger","year":"2015","journal-title":"Lect. Notes. Comput. Sci."},{"key":"B36","doi-asserted-by":"publisher","first-page":"e200204","DOI":"10.1148\/ryai.2021200204","article-title":"Three-dimensional u-net convolutional neural network for detection and segmentation of intracranial metastases","volume":"3","author":"Rudie","year":"2021","journal-title":"Radiol. Artif. Intell"},{"key":"B37","doi-asserted-by":"publisher","first-page":"211","DOI":"10.1007\/s11263-015-0816-y","article-title":"ImageNet Large Scale Visual Recognition Challenge","volume":"115","author":"Russakovsky","year":"2014","journal-title":"Int J. Comput. Vis"},{"key":"B38","doi-asserted-by":"publisher","first-page":"379","DOI":"10.1007\/978-3-319-67389-9_44","article-title":"Tversky loss function for image segmentation using 3D fully convolutional deep networks","author":"Salehi","year":"2017","journal-title":"Lect. Notes Comput. Sci."},{"key":"B39","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s10014-015-0235-3","article-title":"Brain metastasis: clinical characteristics, pathological findings and molecular subtyping for therapeutic implications","volume":"33","author":"Takei","year":"2016","journal-title":"Brain Tumor Pathol"},{"key":"B40","unstructured":"\u201cEfficientNet: rethinking model scaling for convolutional neural networks,\u201d1069110700\n            TanM.\n            LeQ. V.\n          International conference on machine learning2019"},{"key":"B41","doi-asserted-by":"publisher","first-page":"3349","DOI":"10.1109\/TPAMI.2020.2983686","article-title":"Deep high-resolution representation learning for visual recognition","volume":"43","author":"Wang","year":"2019","journal-title":"IEEE Trans.Pattern. Anal. Mach. Intell"},{"key":"B42","unstructured":"WangL.\n            LeeC.-.\n            Y TuZ.\n            LazebnikS.\n          Training Deeper Convolutional Networks with Deep Supervision2015"},{"key":"B43","doi-asserted-by":"publisher","first-page":"505","DOI":"10.1093\/neuonc\/noz234","article-title":"Deep learning-based detection and segmentation-assisted management of brain metastases","volume":"22","author":"Xue","year":"2020","journal-title":"Neuro Oncol"},{"key":"B44","doi-asserted-by":"publisher","first-page":"6020","DOI":"10.1002\/mp.15136","article-title":"MRI pulse sequence integration for deep-learning-based brain metastases segmentation","volume":"48","author":"Yi","year":"2021","journal-title":"Med. Phys"},{"key":"B45","first-page":"1","article-title":"Segmentation Transformer: Object-Contextual Representations for Semantic Segmentation","author":"Yuan","year":"2019","journal-title":"arXiv"},{"key":"B46","unstructured":"\u201cmixup: beyond empirical risk minimization,\u201d\n            ZhangH.\n            CisseM.\n            DauphinY. N.\n            Lopez-PazD.\n          6th International Conference of Learning Represention. ICLR 2018 - Conference Track Proceedings2017"},{"key":"B47","doi-asserted-by":"publisher","first-page":"1227","DOI":"10.1002\/jmri.27129","article-title":"Deep-learning detection of cancer metastases to the brain on MRI","volume":"52","author":"Zhang","year":"2020","journal-title":"J. Magn. Reson. Imaging"},{"key":"B48","doi-asserted-by":"publisher","first-page":"189","DOI":"10.1016\/j.radonc.2020.09.016","article-title":"MetNet: Computer-aided segmentation of brain metastases in post-contrast T1-weighted magnetic resonance imaging","volume":"153","author":"Zhou","year":"","journal-title":"Radiother. Oncol"},{"key":"B49","doi-asserted-by":"publisher","first-page":"407","DOI":"10.1148\/radiol.2020191479","article-title":"Computer-aided Detection of Brain Metastases in T1-weighted MRI for Stereotactic Radiosurgery Using Deep Learning Single-Shot Detectors","volume":"295","author":"Zhou","year":"","journal-title":"Radiology"}],"container-title":["Frontiers in Neuroinformatics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fninf.2022.1056068\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,1,18]],"date-time":"2023-01-18T05:20:47Z","timestamp":1674019247000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fninf.2022.1056068\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,1,18]]},"references-count":49,"alternative-id":["10.3389\/fninf.2022.1056068"],"URL":"https:\/\/doi.org\/10.3389\/fninf.2022.1056068","relation":{},"ISSN":["1662-5196"],"issn-type":[{"value":"1662-5196","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,1,18]]},"article-number":"1056068"}}