{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,21]],"date-time":"2026-04-21T08:20:07Z","timestamp":1776759607236,"version":"3.51.2"},"reference-count":22,"publisher":"Springer Science and Business Media LLC","issue":"11","license":[{"start":{"date-parts":[[2023,6,20]],"date-time":"2023-06-20T00:00:00Z","timestamp":1687219200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,6,20]],"date-time":"2023-06-20T00:00:00Z","timestamp":1687219200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/100008375","name":"University of Basel","doi-asserted-by":"crossref","id":[{"id":"10.13039\/100008375","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Int J CARS"],"abstract":"<jats:title>Abstract<\/jats:title><jats:sec>\n                <jats:title>Purpose<\/jats:title>\n                <jats:p>Automated distinct bone segmentation from CT scans is widely used in planning and navigation workflows. U-Net variants are known to provide excellent results in supervised semantic segmentation. However, in distinct bone segmentation from upper-body CTs a large field of view and a computationally taxing 3D architecture are required. This leads to low-resolution results lacking detail or localisation errors due to missing spatial context when using high-resolution inputs.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Methods<\/jats:title>\n                <jats:p>We propose to solve this problem by using end-to-end trainable segmentation networks that combine several 3D U-Nets working at different resolutions. Our approach, which extends and generalizes HookNet and MRN, captures spatial information at a lower resolution and skips the encoded information to the target network, which operates on smaller high-resolution inputs. We evaluated our proposed architecture against single-resolution networks and performed an ablation study on information concatenation and the number of context networks.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Results<\/jats:title>\n                <jats:p>Our proposed best network achieves a median DSC of 0.86 taken over all 125 segmented bone classes and reduces the confusion among similar-looking bones in different locations. These results outperform our previously published 3D U-Net baseline results on the task and distinct bone segmentation results reported by other groups.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Conclusion<\/jats:title>\n                <jats:p>The presented multi-resolution 3D U-Nets address current shortcomings in bone segmentation from upper-body CT scans by allowing for capturing a larger field of view while avoiding the cubic growth of the input pixels and intermediate computations that quickly outgrow the computational capacities in 3D. The approach thus improves the accuracy and efficiency of distinct bone segmentation from upper-body CT.<\/jats:p>\n              <\/jats:sec>","DOI":"10.1007\/s11548-023-02957-4","type":"journal-article","created":{"date-parts":[[2023,6,20]],"date-time":"2023-06-20T11:02:58Z","timestamp":1687258978000},"page":"2091-2099","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":6,"title":["Improved distinct bone segmentation in upper-body CT through multi-resolution networks"],"prefix":"10.1007","volume":"18","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-0226-9519","authenticated-orcid":false,"given":"Eva","family":"Schnider","sequence":"first","affiliation":[]},{"given":"Julia","family":"Wolleb","sequence":"additional","affiliation":[]},{"given":"Antal","family":"Huck","sequence":"additional","affiliation":[]},{"given":"Mireille","family":"Toranelli","sequence":"additional","affiliation":[]},{"given":"Georg","family":"Rauter","sequence":"additional","affiliation":[]},{"given":"Magdalena","family":"M\u00fcller-Gerbl","sequence":"additional","affiliation":[]},{"given":"Philippe C.","family":"Cattin","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2023,6,20]]},"reference":[{"issue":"5","key":"2957_CR1","doi-asserted-by":"publisher","first-page":"1417","DOI":"10.1007\/s11517-022-02529-9","volume":"60","author":"Y Deng","year":"2022","unstructured":"Deng Y, Wang L, Zhao C, Tang S, Cheng X, Deng H-W, Zhou W (2022) A deep learning-based approach to automatic proximal femur segmentation in quantitative ct images. Med Biol Eng Comput 60(5):1417\u201329","journal-title":"Med Biol Eng Comput"},{"issue":"1","key":"2957_CR2","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s11657-022-01063-3","volume":"17","author":"K Uemura","year":"2022","unstructured":"Uemura K, Otake Y, Takao M, Makino H, Soufi M, Iwasa M, Sugano N, Sato Y (2022) Development of an open-source measurement system to assess the areal bone mineral density of the proximal femur from clinical ct images. Arch Osteoporos 17(1):1\u201311","journal-title":"Arch Osteoporos"},{"issue":"12","key":"2957_CR3","doi-asserted-by":"publisher","first-page":"2914","DOI":"10.1002\/jor.25303","volume":"40","author":"Z Su","year":"2022","unstructured":"Su Z, Liu Z, Wang M, Li S, Lin L, Yuan Z, Pang S, Feng Q, Chen T, Lu H (2022) Three-dimensional reconstruction of kambin\u2019s triangle based on automated magnetic resonance image segmentation. J Orthop Res 40(12):2914\u20132923","journal-title":"J Orthop Res"},{"issue":"1","key":"2957_CR4","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1007\/s11548-018-1883-7","volume":"14","author":"A Klein","year":"2019","unstructured":"Klein A, Warszawski J, Hillenga\u00df J, Maier-Hein KH (2019) Automatic bone segmentation in whole-body ct images. Int J Comput Assist Radiol Surg 14(1):21\u201329","journal-title":"Int J Comput Assist Radiol Surg"},{"issue":"5","key":"2957_CR5","doi-asserted-by":"publisher","first-page":"055010","DOI":"10.1088\/2057-1976\/ac37ab","volume":"8","author":"P Leydon","year":"2021","unstructured":"Leydon P, O\u2019Connell M, Greene D, Curran KM (2021) Bone segmentation in contrast enhanced whole-body computed tomography. Biomed Phys Eng Exp 8(5):055010","journal-title":"Biomed Phys Eng Exp"},{"key":"2957_CR6","doi-asserted-by":"publisher","first-page":"103767","DOI":"10.1016\/j.compbiomed.2020.103767","volume":"121","author":"S Noguchi","year":"2020","unstructured":"Noguchi S, Nishio M, Yakami M, Nakagomi K, Togashi K (2020) Bone segmentation on whole-body ct using convolutional neural network with novel data augmentation techniques. Comput Biol Med 121:103767","journal-title":"Comput Biol Med"},{"key":"2957_CR7","doi-asserted-by":"crossref","unstructured":"Payer C, Stern D, Bischof H, Urschler M (2020) Coarse to fine vertebrae localization and segmentation with spatialconfiguration-net and u-net. In: VISIGRAPP (5: VISAPP), pp 124\u2013133","DOI":"10.5220\/0008975201240133"},{"key":"2957_CR8","doi-asserted-by":"crossref","unstructured":"Yang J, Gu S, Wei D, Pfister H, Ni B (2021) Ribseg dataset and strong point cloud baselines for rib segmentation from ct scans. In: International conference on medical image computing and computer-assisted intervention, pp 611\u2013621. Springer","DOI":"10.1007\/978-3-030-87193-2_58"},{"key":"2957_CR9","doi-asserted-by":"crossref","unstructured":"Faisal A, Khalil A, Chai HY, Lai KW (2021) X-ray carpal bone segmentation and area measurement. Multim Tools Appl 1\u201312","DOI":"10.1007\/s11042-021-11281-5"},{"issue":"11","key":"2957_CR10","doi-asserted-by":"publisher","first-page":"1125","DOI":"10.6009\/jjrt.2020_JSRT_76.11.1125","volume":"76","author":"Y Wakamatsu","year":"2020","unstructured":"Wakamatsu Y, Kamiya N, Zhou X, Hara T, Fujita H (2020) Semantic segmentation of eight regions of upper and lower limb bones using 3d u-net in whole-body ct images. Nihon Hoshasen Gijutsu Gakkai Zasshi 76(11):1125\u20131132","journal-title":"Nihon Hoshasen Gijutsu Gakkai Zasshi"},{"issue":"7","key":"2957_CR11","doi-asserted-by":"publisher","first-page":"2812","DOI":"10.1088\/1361-6560\/aa6055","volume":"62","author":"Y Fu","year":"2017","unstructured":"Fu Y, Liu S, Li HH, Yang D (2017) Automatic and hierarchical segmentation of the human skeleton in CT images. Phys Med Biol 62(7):2812\u20132833","journal-title":"Phys Med Biol"},{"key":"2957_CR12","doi-asserted-by":"publisher","first-page":"89","DOI":"10.1016\/j.ejrad.2019.01.028","volume":"113","author":"S Lindgren Belal","year":"2019","unstructured":"Lindgren Belal S, Sadik M, Kaboteh R, Enqvist O, Ul\u00e9n J, Poulsen MH, Simonsen J, H\u00f8ilund-Carlsen PF, Edenbrandt L, Tr\u00e4g\u00e5rdh E (2019) Deep learning for segmentation of 49 selected bones in CT scans: First step in automated PET\/CT-based 3D quantification of skeletal metastases. Eur J Radiol 113:89\u201395","journal-title":"Eur J Radiol"},{"key":"2957_CR13","doi-asserted-by":"crossref","unstructured":"Kamiya N, Kume M, Zheng G, Zhou X, Kato H, Chen H, Muramatsu C, Hara T, Miyoshi T, Matsuo M, Fujita H (2018) Automated recognition of erector spinae muscles and their skeletal attachment region via deep learning in torso ct images. In: International workshop on computational methods and clinical applications in musculoskeletal imaging, pp 1\u201310. Springer","DOI":"10.1007\/978-3-030-11166-3_1"},{"key":"2957_CR14","doi-asserted-by":"publisher","first-page":"102364","DOI":"10.1016\/j.artmed.2022.102364","volume":"132","author":"A Boutillon","year":"2022","unstructured":"Boutillon A, Borotikar B, Burdin V, Conze P-H (2022) Multi-structure bone segmentation in pediatric mr images with combined regularization from shape priors and adversarial network. Artif Intell Med 132:102364","journal-title":"Artif Intell Med"},{"issue":"2","key":"2957_CR15","doi-asserted-by":"publisher","first-page":"203","DOI":"10.1038\/s41592-020-01008-z","volume":"18","author":"F Isensee","year":"2021","unstructured":"Isensee F, Jaeger PF, Kohl SA, Petersen J, Maier-Hein KH (2021) nnu-net: a self-configuring method for deep learning-based biomedical image segmentation. Nat Methods 18(2):203\u2013211","journal-title":"Nat Methods"},{"key":"2957_CR16","doi-asserted-by":"crossref","unstructured":"Gu F, Burlutskiy N, Andersson M, Wil\u00e9n LK (2018) Multi-resolution networks for semantic segmentation in whole slide images. Comput Pathol Ophthalmic Med Image Anal 11\u201318","DOI":"10.1007\/978-3-030-00949-6_2"},{"key":"2957_CR17","doi-asserted-by":"publisher","first-page":"101890","DOI":"10.1016\/j.media.2020.101890","volume":"68","author":"M Van Rijthoven","year":"2021","unstructured":"Van Rijthoven M, Balkenhol M, Sili\u0146a K, Van Der Laak J, Ciompi F (2021) Hooknet: Multi-resolution convolutional neural networks for semantic segmentation in histopathology whole-slide images. Med Image Anal 68:101890","journal-title":"Med Image Anal"},{"key":"2957_CR18","unstructured":"Faludi B, Zentai N, Zelechowski M, Zam A, Rauter G, Griessen M, Cattin PC (2021) Transfer-function-independent acceleration structure for volume rendering in virtual reality. In: Proceedings of the conference on high-performance graphics, pp 1\u201310"},{"issue":"4","key":"2957_CR19","doi-asserted-by":"publisher","first-page":"451","DOI":"10.1080\/21681163.2021.2002192","volume":"10","author":"M \u017belechowski","year":"2021","unstructured":"\u017belechowski M, Karnam M, Faludi B, Gerig N, Rauter G, Cattin PC (2021) Patient positioning by visualising surgical robot rotational workspace in augmented reality. Comput Methods Biomech Biomed Eng Imaging Vis 10(4):451\u20137","journal-title":"Comput Methods Biomech Biomed Eng Imaging Vis"},{"issue":"11","key":"2957_CR20","doi-asserted-by":"publisher","first-page":"2113","DOI":"10.1007\/s11548-022-02650-y","volume":"17","author":"E Schnider","year":"2022","unstructured":"Schnider E, Huck A, Toranelli M, Rauter G, M\u00fcller-Gerbl M, Cattin PC (2022) Improved distinct bone segmentation from upper-body ct using binary-prediction-enhanced multi-class inference. Int J Comput Assist Radiol Surg 17(11):2113\u20132120","journal-title":"Int J Comput Assist Radiol Surg"},{"key":"2957_CR21","doi-asserted-by":"crossref","unstructured":"Schnider E, Horv\u00e1th A, Rauter G, Zam A, M\u00fcller-Gerbl M, Cattin PC (2020) 3d segmentation networks for excessive numbers of classes: Distinct bone segmentation in upper bodies. In: International workshop on machine learning in medical imaging, pp 40\u201349. Springer","DOI":"10.1007\/978-3-030-59861-7_5"},{"key":"2957_CR22","doi-asserted-by":"crossref","unstructured":"Milletari F, Navab N, Ahmadi S-A (2016) V-net: Fully convolutional neural networks for volumetric medical image segmentation. In: 2016 fourth international conference on 3D vision (3DV), pp 565\u2013571. IEEE","DOI":"10.1109\/3DV.2016.79"}],"container-title":["International Journal of Computer Assisted Radiology and Surgery"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11548-023-02957-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s11548-023-02957-4\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11548-023-02957-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,10,21]],"date-time":"2023-10-21T03:24:01Z","timestamp":1697858641000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s11548-023-02957-4"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,6,20]]},"references-count":22,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2023,11]]}},"alternative-id":["2957"],"URL":"https:\/\/doi.org\/10.1007\/s11548-023-02957-4","relation":{},"ISSN":["1861-6429"],"issn-type":[{"value":"1861-6429","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,6,20]]},"assertion":[{"value":"10 January 2023","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"9 May 2023","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"20 June 2023","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"None of the authors have competing interests to declare that are relevant to the content of this article.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}},{"value":"Body donours signed informed consent regarding publications using their data.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"Informed consent was obtained from all individual body donours included in the study.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent to participate"}},{"value":"This research study was conducted retrospectively from CT data routinely obtained from body donours. No ethical approval is required.","order":5,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethical approval"}}]}}