{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,19]],"date-time":"2026-10-19T23:15:26Z","timestamp":1792451726180,"version":"4.5.0"},"reference-count":61,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2023,12,11]],"date-time":"2023-12-11T00:00:00Z","timestamp":1702252800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100020950","name":"National Science and Technology Council","doi-asserted-by":"publisher","award":["NSTC 111-2221-E-182-057 MOST 110-2221-E-182-034 MOST 109-2221-E-182-025 MOST 108-2221-E-182-061"],"award-info":[{"award-number":["NSTC 111-2221-E-182-057 MOST 110-2221-E-182-034 MOST 109-2221-E-182-025 MOST 108-2221-E-182-061"]}],"id":[{"id":"10.13039\/501100020950","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Bioeng. Biotechnol."],"abstract":"<jats:p>Creating a personalized implant for cranioplasty can be costly and aesthetically challenging, particularly for comminuted fractures that affect a wide area. Despite significant advances in deep learning techniques for 2D image completion, generating a 3D shape inpainting remains challenging due to the higher dimensionality and computational demands for 3D skull models. Here, we present a practical deep-learning approach to generate implant geometry from defective 3D skull models created from CT scans. Our proposed 3D reconstruction system comprises two neural networks that produce high-quality implant models suitable for clinical use while reducing training time. The first network repairs low-resolution defective models, while the second network enhances the volumetric resolution of the repaired model. We have tested our method in simulations and real-life surgical practices, producing implants that fit naturally and precisely match defect boundaries, particularly for skull defects above the Frankfort horizontal plane.<\/jats:p>","DOI":"10.3389\/fbioe.2023.1297933","type":"journal-article","created":{"date-parts":[[2023,12,11]],"date-time":"2023-12-11T05:07:42Z","timestamp":1702271262000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":14,"title":["Creating high-resolution 3D cranial implant geometry using deep learning techniques"],"prefix":"10.3389","volume":"11","author":[{"given":"Chieh-Tsai","family":"Wu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yao-Hung","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yau-Zen","family":"Chang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2023,12,11]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1803.08375","article-title":"Deep learning using rectified linear units (relu)","author":"Agarap","year":"2018"},{"key":"B2","doi-asserted-by":"publisher","first-page":"445","DOI":"10.1016\/j.wneu.2020.04.211","article-title":"Cranioplasty: a comprehensive review of the history, materials, surgical aspects, and complications","volume":"139","author":"Alkhaibary","year":"2020","journal-title":"World Neurosurg."},{"key":"B3","doi-asserted-by":"publisher","first-page":"37","DOI":"10.5555\/3045796.3045801","article-title":"Autoencoders, unsupervised learning, and deep architectures","volume":"27","author":"Baldi","year":"2011","journal-title":"Proc. ICML Workshop Unsupervised Transf. Learn"},{"key":"B4","doi-asserted-by":"publisher","first-page":"8242","DOI":"10.1038\/s41598-020-64803-w","article-title":"Evaluating white matter lesion segmentations with refined s\u00f8rensen-dice analysis","volume":"10","author":"Carass","year":"2020","journal-title":"Sci. Rep."},{"key":"B5","first-page":"77","article-title":"PointNet: deep learning on point sets for 3D classification and segmentation","author":"Charles","year":"2017"},{"key":"B6","doi-asserted-by":"publisher","first-page":"4199","DOI":"10.1038\/s41598-017-04454-6","article-title":"Computer-aided implant design for the restoration of cranial defects","volume":"23","author":"Chen","year":"2017","journal-title":"Sci. Rep."},{"key":"B7","doi-asserted-by":"publisher","first-page":"2388","DOI":"10.1016\/S0140-6736(18)31645-3","article-title":"Deep learning algorithms for detection of critical findings in head CT scans: a retrospective study","volume":"392","author":"Chilamkurthy","year":"2018","journal-title":"Lancet"},{"key":"B8","first-page":"6545","article-title":"Shape Completion using 3D-encoder-predictor CNNs and shape synthesis","author":"Dai","year":"2017"},{"key":"B9","doi-asserted-by":"publisher","first-page":"3244","DOI":"10.1364\/BOE.9.003244","article-title":"DRUNET: a dilated-residual U-Net deep learning network to segment optic nerve head tissues in optical coherence tomography images","volume":"9","author":"Devalla","year":"2018","journal-title":"Biomed. Opt. Express."},{"key":"B10","doi-asserted-by":"publisher","first-page":"297","DOI":"10.2307\/1932409","article-title":"Measures of the amount of ecologic association between species","volume":"26","author":"Dice","year":"1945","journal-title":"Ecology"},{"key":"B11","first-page":"47","article-title":"Deep learning using augmentation via registration: 1st place solution to the AutoImplant 2020 challenge","volume-title":"Lecture notes in computer science, LNCS","author":"Ellis","year":"2020"},{"key":"B12","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-92652-6_2","article-title":"Qualitative criteria for feasible cranial implant designs","volume-title":"Towards the automatization of cranial implant design in cranioplasty II, LNCS 13123","author":"Ellis","year":"2021"},{"key":"B13","doi-asserted-by":"publisher","first-page":"1054","DOI":"10.1016\/j.jcms.2019.03.027","article-title":"Virtual reconstruction of bilateral midfacial defects by using statistical shape modeling","volume":"47","author":"Fuessinger","year":"2019","journal-title":"J. Craniomaxillofac. Surg."},{"key":"B14","volume-title":"Cranial defect datasets","author":"Gall","year":"2019"},{"key":"B15","first-page":"2261","article-title":"Densely connected convolutional networks","author":"Gao Huang","year":"2017"},{"key":"B16","doi-asserted-by":"publisher","first-page":"3664","DOI":"10.1109\/TIP.2021.3064223","article-title":"Distributed Connected component filtering and analysis in 2D and 3D tera-scale data Sets","volume":"30","author":"Gazagnes","year":"2021","journal-title":"IEEE Trans. Image Process."},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1406.2661","article-title":"Generative adversarial networks","author":"Goodfellow","year":"2014"},{"key":"B18","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3306346.3322959","article-title":"MeshCNN: a network with an edge","volume":"38","author":"Hanocka","year":"2019","journal-title":"ACM Trans. Graph."},{"key":"B19","first-page":"770","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"key":"B20","doi-asserted-by":"publisher","first-page":"504","DOI":"10.1126\/science.1127647","article-title":"Reducing the dimensionality of data with neural networks","volume":"313","author":"Hinton","year":"2006","journal-title":"Science"},{"key":"B21","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3072959.3073659","article-title":"Globally and locally consistent image completion","volume":"36","author":"Iizuka","year":"2017","journal-title":"ACM Trans. Graph."},{"key":"B22","doi-asserted-by":"publisher","first-page":"22884","DOI":"10.1109\/ACCESS.2020.2970169","article-title":"Image inpainting based on generative adversarial networks","volume":"8","author":"Jiang","year":"2020","journal-title":"IEEE Access"},{"key":"B23","doi-asserted-by":"publisher","first-page":"106902","DOI":"10.1016\/j.dib.2021.106902","article-title":"SkullBreak\/SkullFix \u2013 dataset for automatic cranial implant design and a benchmark for volumetric shape learning tasks","volume":"35","author":"Kodym","year":"2021","journal-title":"Data Brief"},{"key":"B24","doi-asserted-by":"crossref","DOI":"10.1109\/ICUMT48472.2019.8970918","article-title":"Upsampling algorithms for autoencoder segmentation neural networks: a comparison study","author":"Kolarik","year":"2019"},{"key":"B25","doi-asserted-by":"publisher","first-page":"56","DOI":"10.1016\/j.jocn.2008.04.001","article-title":"Cranioplasty using polymethyl methacrylate prostheses","volume":"16","author":"Lee","year":"2009","journal-title":"J. Clin. Neurosci."},{"key":"B26","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-64327-0","volume-title":"Towards the automatization of cranial implant design in cranioplasty I","author":"Li","year":"2020"},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2204.05703","article-title":"Back to the Roots: reconstructing large and complex cranial defects using an image-based statistical shape model","author":"Li","year":"2022"},{"key":"B28","doi-asserted-by":"publisher","first-page":"36","DOI":"10.1038\/s41597-021-00806-0","article-title":"Synthetic skull bone defects for automatic patient-specific craniofacial implant design","volume":"8","author":"Li","year":"","journal-title":"Sci. Data"},{"key":"B29","doi-asserted-by":"publisher","first-page":"107524","DOI":"10.1016\/j.dib.2021.107524","article-title":"MUG500+: database of 500 high-resolution healthy human skulls and 29 craniotomy skulls and implants","volume":"39","author":"Li","year":"","journal-title":"Data Brief."},{"key":"B30","doi-asserted-by":"publisher","first-page":"2329","DOI":"10.1109\/TMI.2021.3077047","article-title":"AutoImplant 2020-First MICCAI challenge on automatic cranial implant design","volume":"40","author":"Li","year":"","journal-title":"IEEE Trans. Med. Imaging."},{"key":"B31","doi-asserted-by":"publisher","first-page":"102171","DOI":"10.1016\/j.media.2021.102171","article-title":"Automatic skull defect restoration and cranial implant generation for cranioplasty","volume":"73","author":"Li","year":"","journal-title":"Med. Image Anal."},{"key":"B32","first-page":"3911","article-title":"Generative face completion","author":"Li","year":"2017"},{"key":"B33","first-page":"1251","article-title":"Learning effective binary descriptors via cross entropy","author":"Liu","year":"2017"},{"key":"B34","doi-asserted-by":"crossref","first-page":"63","DOI":"10.1007\/978-3-030-92652-6_6","article-title":"A U-Net based system for cranial implant design with pre-processing and learned implant filtering","volume-title":"Towards the automatization of cranial implant design in cranioplasty IILNCS, 13123","author":"Mahdi","year":"2021"},{"key":"B35","first-page":"922","article-title":"VoxNet: a 3D convolutional neural network for real-time object recognition","author":"Maturana","year":"2015"},{"key":"B36","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-64327-0_5","article-title":"Cranial implant design via virtual craniectomy with shape priors","volume-title":"Towards the automatization of cranial implant design in cranioplasty, lecture notes in computer science, LNCS 12439","author":"Matzkin","year":""},{"key":"B37","first-page":"390","article-title":"Self-supervised skull reconstruction in brain CT Images with decompressive craniectomy","volume-title":"Med. Image. Comput. Comput. Assist. Interv., lecture notes in computer science, LNCS 12262","author":"Matzkin","year":""},{"key":"B38","doi-asserted-by":"publisher","first-page":"5597","DOI":"10.1109\/IEMBS.2007.4353615","article-title":"Hausdorff distance based 3D quantification of brain tumor evolution from MRI images","volume":"2007","author":"Morain-Nicolier","year":"2007","journal-title":"Proc. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc."},{"key":"B39","first-page":"151","article-title":"Automated computer-aided design of cranial implants using a deep volumetric convolutional denoising autoencoder","volume-title":"Proc. The world conf. Inf. Syst. And technol. (WorldCIST'19 2019)","author":"Morais","year":"2019"},{"key":"B40","doi-asserted-by":"publisher","first-page":"11542","DOI":"10.1038\/s41598-021-90985-y","article-title":"Reproducibility of CT-based radiomic features against image resampling and perturbations for tumour and healthy kidney in renal cancer patients","volume":"11","author":"Mottola","year":"2021","journal-title":"Sci. Rep."},{"key":"B41","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2016.278","article-title":"Context encoders: feature learning by inpainting","author":"Pathak","year":"2016"},{"key":"B42","doi-asserted-by":"publisher","first-page":"239","DOI":"10.1093\/ejo\/cjx066","article-title":"Three-dimensional Frankfort horizontal plane for 3D cephalometry: a comparative assessment of conventional versus novel landmarks and horizontal planes","volume":"40","author":"Pittayapat","year":"2018","journal-title":"Eur. J. Orth."},{"key":"B43","doi-asserted-by":"publisher","first-page":"e027904","DOI":"10.1136\/bmjopen-2018-027904","article-title":"Necessity of dural tenting sutures in modern neurosurgery: protocol for a systematic review","volume":"9","author":"Przepi\u00f3rka","year":"2019","journal-title":"BMJ Open"},{"key":"B44","first-page":"5105","article-title":"PointNet++: deep hierarchical feature learning on point sets in a metric space","volume-title":"Proc. 31st int. Conf. Neural inf. Process. Syst. (NIPS 2017)","author":"Qi","year":"2017"},{"key":"B45","first-page":"234","article-title":"U-Net: convolutional networks for biomedical image segmentation","volume-title":"Int. Conf. Medical imag. Comput. And comput.-assisted intervention (MICCAI 2015), LNCS 9351","author":"Ronneberger","year":"2015"},{"key":"B46","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1186\/s40537-019-0197-0","article-title":"A survey on image data augmentation for deep learning","volume":"6","author":"Shorten","year":"2019","journal-title":"J. Big Data"},{"key":"B47","doi-asserted-by":"publisher","first-page":"1900","DOI":"10.1097\/SCS.0000000000002019","article-title":"Comparison of direct and digital measures of cranial vault asymmetry for assessment of plagiocephaly","volume":"26","author":"Skolnick","year":"2015","journal-title":"J. Craniofac. Surg."},{"key":"B48","first-page":"2107","article-title":"Octree generating networks: efficient convolutional architectures for high-resolution 3D outputs","author":"Tatarchenko","year":"2017"},{"key":"B49","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3072959.3073608","article-title":"O-CNN: octree-based convolutional neural networks for 3D shape analysis","volume":"36","author":"Wang","year":"2017","journal-title":"ACM Trans. Graph."},{"key":"B50","first-page":"1074","article-title":"Deep octree-based CNNs with output-guided skip connections for 3D shape and scene completion","author":"Wang","year":"2020"},{"key":"B51","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3272127.3275050","article-title":"Adaptive O-CNN: a patch-based deep representation of 3D shapes","volume":"37","author":"Wang","year":"2018","journal-title":"ACM Trans. Graph."},{"key":"B52","doi-asserted-by":"publisher","first-page":"e416","DOI":"10.1016\/j.wneu.2020.12.072","article-title":"Patient-specific three-dimensional printing guide for single-stage skull bone tumor surgery: novel software workflow with manufacturing of prefabricated jigs for bone resection and reconstruction","volume":"147","author":"Wu","year":"2021","journal-title":"World Neurosurg."},{"key":"B53","doi-asserted-by":"publisher","first-page":"2683","DOI":"10.1038\/s41598-022-06606-9","article-title":"Three-dimensional deep learning to automatically generate cranial implant geometry","volume":"12","author":"Wu","year":"2022","journal-title":"Sci. Rep."},{"key":"B54","doi-asserted-by":"publisher","first-page":"362","DOI":"10.1109\/TBME.2020.2990586","article-title":"Estimating reference shape model for personalized surgical reconstruction of craniomaxillofacial defects","volume":"68","author":"Xiao","year":"2021","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"B55","first-page":"14976","article-title":"Generative PointNet: deep energy-based learning on unordered point sets for 3D generation, reconstruction and classification","author":"Xie","year":"2021"},{"key":"B56","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-01264-9_1","article-title":"Shift-Net: image inpainting via deep feature rearrangement","volume-title":"Computer vision \u2013 ECCV 2018, lecture notes in computer science, LNCS 11218","author":"Yan","year":"2018"},{"key":"B57","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2017.434","article-title":"High-resolution image inpainting using multi-scale neural patch synthesis","author":"Yang","year":"2017"},{"key":"B58","doi-asserted-by":"publisher","first-page":"e841","DOI":"10.1016\/j.wneu.2019.05.005","article-title":"Long-term complications of cranioplasty using stored autologous bone graft, three-dimensional polymethyl methacrylate, or titanium mesh after decompressive craniectomy: a single-center experience after 596 procedures","volume":"128","author":"Yeap","year":"2019","journal-title":"World Neurosurg."},{"key":"B59","doi-asserted-by":"publisher","first-page":"989","DOI":"10.1007\/s00276-015-1430-y","article-title":"A new three-dimensional measurement in evaluating the cranial asymmetry caused by craniosynostosis","volume":"37","author":"Yin","year":"2015","journal-title":"Surg. Radiol. Anat."},{"key":"B60","article-title":"Multi-scale context aggregation by dilated convolutions","volume-title":"Proc. 4th int. Conf. Learn. Rep. (ICLR 2016)","author":"Yu","year":"2016"},{"key":"B61","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1212.5701","article-title":"ADADELTA: an adaptive learning rate method","author":"Zeiler","year":"2012"}],"container-title":["Frontiers in Bioengineering and Biotechnology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fbioe.2023.1297933\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,12,11]],"date-time":"2023-12-11T05:07:53Z","timestamp":1702271273000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fbioe.2023.1297933\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,12,11]]},"references-count":61,"alternative-id":["10.3389\/fbioe.2023.1297933"],"URL":"https:\/\/doi.org\/10.3389\/fbioe.2023.1297933","relation":{},"ISSN":["2296-4185"],"issn-type":[{"value":"2296-4185","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,12,11]]},"article-number":"1297933"}}