{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,21]],"date-time":"2026-05-21T17:04:44Z","timestamp":1779383084940,"version":"3.53.1"},"reference-count":34,"publisher":"Wiley","issue":"7","license":[{"start":{"date-parts":[[2025,4,3]],"date-time":"2025-04-03T00:00:00Z","timestamp":1743638400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2025,7]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>In external x\u2010ray radiation therapy, the administered dose distribution can deviate from the planned dose due to alterations in patient positioning, changes in intra\u2010fraction anatomy, and the limited precision of the beam delivery system in spatial terms. Adaptive radiation therapy (ART) can potentially improve dose delivery accuracy by re\u2010optimizing the treatment plan before each fraction, maximizing the dose to the target volume while minimizing exposure to surrounding radiosensitive organs. However, to effectively implement ART, the stages of the radiation therapy pipeline, including image acquisition, segmentation, physician directive generation, and treatment plan generation, must be optimized for maximum speed and accuracy to ensure feasibility prior to each treatment fraction. In this work, we focus on image segmentation. By reducing the segmentation computation time, one can reproduce the planning process for each session, enabling routine customization for individual patients, achieving safe dose escalation, better cancer control, and reduced risk of severe radiotoxicity.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>The aim of this study is to develop a fast point\u2010cloud\u2010based segmentation model with novel in\u2010silico\u2010aided data augmentation and demonstrate it on pelvic computed tomography (CT) patient data used in prostate cancer (PCa) treatment. This model can be implemented during ART because it requires only a few seconds to perform organ\u00a0segmentation.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>In this study, a dataset of pelvic CT images was obtained from Order of St. Francis (OSF) Healthcare Hospital (Peoria, IL, USA), comprising 38 images in total. These were divided into 25 for training, seven for validation, and six for testing the developed model. A novel point\u2010cloud\u2010based model was used to reduce the prostate segmentation time, cross\u2010validation was implemented to ensure the robustness of the model. The developed point\u2010cloud\u2010based network is a novel deep\u2010learning (DL) model that adds a loss function that combines region\u2010based with a new boundary loss function. The region\u2010based loss enables the identification of large volumes while the boundary loss, whose relative weight increases with the epochs, increases the network training ability of uneven surfaces, like the interface between the prostate bladder and rectum, which are challenging to resolve. We introduced a new data\u2010augmentation approach to expand the training set. This fully automated method generates synthetic 3\u2010D CT images by creating relevant organs in the extended cardiac\u2010torso (XCAT) computational phantom. The Dice similarity coefficient was used as an assessment metric and compared to state\u2010of\u2010the\u2010art segmentation models. The doses to the prostate and organs at risk (i.e., bladder and rectum) were also calculated for both our automated segmentation and manual expert segmentation to evaluate the practical feasibility of the point\u2010cloud\u2010based approach.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Our model achieves the segmentation results (Dice coefficient) of 0.92\u00a0\u00a00.04, 0.89\u00a0\u00a00.05, and 0.84\u00a0\u00a00.07 for bladder, prostate, and rectum, respectively. The accuracy of the prostate segmentation outperforms the voxel\u2010based segmentation models reported in the literature. More importantly, the average segmentation time of the point\u2010cloud model for a single 3\u2010D CT data set was 1.8 times faster than 2\u2010D fully convolutional network (FCN), and 11 times faster than 3\u2010D U\u2010Net. The improved loss function and in\u2010silico\u2010based training data augmentation approach effectively enabled the model to learn features of outlier data sets, thereby improving the model's robustness across diverse images. The developed fast and robust point\u2010cloud segmentation model can potentially be applied to ART to improve the treatment\u00a0workflow.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Our proposed method demonstrates favorable performance in the segmentation of X\u2010ray CT data. Results confirmed that the point\u2010cloud\u2010model is faster than voxel\u2010based segmentation algorithms while achieving comparable or better segmentation results. The segmentation approach can be integrated into ART workflow, ultimately reducing the workload of clinicians and\u00a0radiologists.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.17815","type":"journal-article","created":{"date-parts":[[2025,4,5]],"date-time":"2025-04-05T09:19:30Z","timestamp":1743844770000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Point\u2010cloud segmentation with in\u2010silico data augmentation for prostate cancer treatment"],"prefix":"10.1002","volume":"52","author":[{"given":"Jianxin","family":"Zhou","sequence":"first","affiliation":[{"name":"Department of Nuclear, Plasma, and Radiological Engineering University of Illinois at Urbana\u2010Champaign Urbana Illinois USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Massimiliano","family":"Salvatori","sequence":"additional","affiliation":[{"name":"IFC\u2010National Research Council Pisa Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kadishe","family":"Fejza","sequence":"additional","affiliation":[{"name":"OSF HealthCare, Department of Radiation Oncology Peoria Illinois USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gregory M.","family":"Hermann","sequence":"additional","affiliation":[{"name":"OSF HealthCare, Department of Radiation Oncology Peoria Illinois USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Angela Di","family":"Fulvio","sequence":"additional","affiliation":[{"name":"Department of Nuclear, Plasma, and Radiological Engineering University of Illinois at Urbana\u2010Champaign Urbana Illinois USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2025,4,3]]},"reference":[{"key":"e_1_2_8_2_1","doi-asserted-by":"publisher","DOI":"10.1088\/0031-9155\/42\/1\/008"},{"key":"e_1_2_8_3_1","unstructured":"Investing in clinical research to improve patient outcomes: etho adaptive therapy clinical research.https:\/\/www.varian.com\/resources\u2010support\/blogs\/investing\u2010clinical\u2010research\u2010improve\u2010patient\u2010outcomes\u2010ethostm"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.13479"},{"key":"e_1_2_8_5_1","article-title":"Making on\u2010line adaptive radiotherapy possible using artificial intelligence and machine learning for efficient daily re\u2010planning","author":"Archambault Y","year":"2020","journal-title":"Medical Physics International Journal"},{"key":"e_1_2_8_6_1","doi-asserted-by":"crossref","unstructured":"MilletariF NavabN AhmadiSA.V\u2010Net: Fully convolutional neural networks for volumetric medical image segmentation. In:Proceedings \u2010 2016 4th International Conference on 3D Vision 3DV 2016.IEEE;2016:565\u2010571.","DOI":"10.1109\/3DV.2016.79"},{"key":"e_1_2_8_7_1","series-title":"Lecture Notes in Computer Science (including sub series Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)","volume-title":"U\u2010net: convolutional networks for biomedical image segmentation","author":"Ronneberger O","year":"2015"},{"key":"e_1_2_8_8_1","doi-asserted-by":"crossref","unstructured":"DaiA ChangAX SavvaM HalberM FunkhouserT Nie\u00dfnerM.ScanNet: richly\u2010annotated 3D reconstructions of indoor scenes. In:2017 IEEE Conference on Computer Vision and Pattern Recognition.IEEE;2017:2432\u20102443.","DOI":"10.1109\/CVPR.2017.261"},{"key":"e_1_2_8_9_1","doi-asserted-by":"crossref","unstructured":"CaesarH BankitiV LangAH et\u00a0al.nuScenes: a multimodal dataset for autonomous driving. In:2020 IEEE\/CVF Conference of Computer Vision and Pattern Recognition.IEEE;2020.","DOI":"10.1109\/CVPR42600.2020.01164"},{"key":"e_1_2_8_10_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2015.06.026"},{"key":"e_1_2_8_11_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.autcon.2020.103210"},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2021.09.027"},{"key":"e_1_2_8_13_1","doi-asserted-by":"crossref","unstructured":"YueX WuB SeshiaSA KeutzerK Sangiovanni\u2010VincentelliAL.A LiDAR point cloud generator: from a virtual world to autonomous driving. In:2018 ACM International Conference on Multimedia Retrieval (ICMR).ICMR;2018:458\u2010464.","DOI":"10.1145\/3206025.3206080"},{"key":"e_1_2_8_14_1","doi-asserted-by":"publisher","DOI":"10.1109\/TIM.2021.3123218"},{"key":"e_1_2_8_15_1","unstructured":"CharlesQR LiY HaoS LeonidasGJ.PointNet++: deep hierarchical feature learning on point sets in a metric space. In:2017 Conference on Neural Information Processing Systems (NIPS).NIPS;2017."},{"key":"e_1_2_8_16_1","doi-asserted-by":"crossref","unstructured":"HuaBS TranMK YeungSK.Pointwise convolutional neural networks. In:2018 IEEE Conference on Computer Vision and Pattern Recognition.IEEE;2017:984\u2010993.","DOI":"10.1109\/CVPR.2018.00109"},{"key":"e_1_2_8_17_1","doi-asserted-by":"crossref","unstructured":"KlokovR LempitskyV.Escape from cells: deep Kd\u2010networks for the recognition of 3D point cloud models. In:2017 IEEE International Conference of Computer Vision.IEEE;2017:863\u2010872.","DOI":"10.1109\/ICCV.2017.99"},{"key":"e_1_2_8_18_1","doi-asserted-by":"crossref","unstructured":"YangJ GuS WeiD PfisterH NiB.RibSeg dataset and strong point cloud baselines for rib segmentation from CT scans. In:Medical Image Computing and Computer Assisted Intervention\u2013MICCAI 2021.Springer;2021.","DOI":"10.1007\/978-3-030-87193-2_58"},{"key":"e_1_2_8_19_1","unstructured":"Ghazvinian ZanjaniF Anssari MoinD VerheijB et\u00a0al.Deep learning approach to semantic segmentation in 3D point cloud intra\u2010oral scans of teeth. In:Proceedings of The 2nd International Conference on Medical Imaging with Deep Learning;102 of Proceedings of Machine Learning Research Proceedings of Machine Learning Research.PMLR;2019:557\u2010571."},{"key":"e_1_2_8_20_1","unstructured":"Mask R\u2010CNN\u2013Torchvision main documentation.https:\/\/pytorch.org\/vision\/main\/models\/mask_rcnn.html"},{"key":"e_1_2_8_21_1","article-title":"Rethinking network design and local geometry in point cloud: a simple residual MLP framework","author":"Ma X","year":"2022","journal-title":"ArXiv"},{"key":"e_1_2_8_22_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-024-51258-6"},{"key":"e_1_2_8_23_1","unstructured":"AbdullaW.Mask R\u2010CNN for object detection and instance segmentation on Keras and TensorFlow.2017.https:\/\/github.com\/matterport\/Mask_RCNN"},{"key":"e_1_2_8_24_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2020.101851"},{"key":"e_1_2_8_25_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.3480985"},{"key":"e_1_2_8_26_1","doi-asserted-by":"publisher","DOI":"10.1007\/BF02573985"},{"key":"e_1_2_8_27_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.2955743"},{"key":"e_1_2_8_28_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41597-019-0235-y"},{"key":"e_1_2_8_29_1","doi-asserted-by":"publisher","DOI":"10.2307\/1932409"},{"key":"e_1_2_8_30_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/abcad9"},{"key":"e_1_2_8_31_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13014-022-01985-9"},{"key":"e_1_2_8_32_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejmp.2020.09.004"},{"key":"e_1_2_8_33_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2019.03.017"},{"key":"e_1_2_8_34_1","unstructured":"WadaK.Pytorch\u2010fcn: PyTorch implementation of fully convolutional networks.2017.https:\/\/github.com\/wkentaro\/pytorch\u2010fcn"},{"key":"e_1_2_8_35_1","doi-asserted-by":"publisher","DOI":"10.7554\/eLife.57613"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.17815","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,8]],"date-time":"2025-12-08T10:30:58Z","timestamp":1765189858000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.17815"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,3]]},"references-count":34,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2025,7]]}},"alternative-id":["10.1002\/mp.17815"],"URL":"https:\/\/doi.org\/10.1002\/mp.17815","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,4,3]]},"assertion":[{"value":"2024-09-27","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-03-23","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-04-03","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e17815"}}