In human neuroimaging studies, atlas registration enables mapping MRI scans to a common coordinate frame, which is necessary to aggregate data from multiple subjects. Machine learning registration methods have achieved excellent speed and accuracy but lack interpretability and flexibility at test time (since their deformation model is fixed). More recently, keypoint-based methods have been proposed to tackle these issues, but their accuracy is still subpar, particularly when fitting nonlinear transforms. Here we propose Registration by Regression (RbR), a novel atlas registration framework that: is highly robust and flexible; can be trained with cheaply obtained data; and operates on a single channel, such that it can also be used as pretraining for other tasks. RbR predicts the (x, y, z) atlas coordinates for every voxel of the input scan (i.e., every voxel is a keypoint), and then uses closed-form expressions to quickly fit transforms using a wide array of possible deformation models, including affine and nonlinear (e.g., Bspline, Demons, invertible diffeomorphic models, etc.). Robustness is provided by the large number of voxels informing the registration and can be further increased by robust estimators like RANSAC. Experiments on independent public datasets show that RbR yields more accurate registration than competing keypoint approaches, over a wide range of deformation models.
在人类神经成像研究中，地图集注册可以将MRI扫描映射到一个共同的坐标框架，这对于汇总来自多个受试者的数据是必要的。机器学习配准方法取得了优异的速度和准确性，但在测试时缺乏可解释性和灵活性（因为它们的变形模型是固定的）。最近，已经提出了基于关键点的方法来解决这些问题，但它们的精度仍然低于标准，特别是在拟合非线性变换时。在这里，我们提出了一种新的图谱配准框架——回归配准（RbR），它具有高度的鲁棒性和灵活性；可以用廉价获得的数据进行训练；并且在单一通道上运行，因此它也可以用于其他任务的预训练。RbR预测输入扫描的每个体素的（x, y, z）图谱坐标（即，每个体素是一个关键点），然后使用封闭形式的表达式来快速拟合变换，使用广泛的可能的变形模型，包括仿射和非线性（例如，b样条，Demons，可逆微分同态模型等）。鲁棒性是由大量的体素提供的，并且可以通过像RANSAC这样的鲁棒估计器进一步提高。在独立的公共数据集上的实验表明，在广泛的变形模型上，RbR比竞争的关键点方法获得更准确的配准。
{"title":"Registration by Regression (RbR): a framework for interpretable and flexible atlas registration.","authors":"Karthik Gopinath, Xiaoling Hu, Malte Hoffmann, Oula Puonti, Juan Eugenio Iglesias","doi":"10.1007/978-3-031-73480-9_16","DOIUrl":"https://doi.org/10.1007/978-3-031-73480-9_16","url":null,"abstract":"<p><p>In human neuroimaging studies, atlas registration enables mapping MRI scans to a common coordinate frame, which is necessary to aggregate data from multiple subjects. Machine learning registration methods have achieved excellent speed and accuracy but lack interpretability and flexibility at test time (since their deformation model is fixed). More recently, keypoint-based methods have been proposed to tackle these issues, but their accuracy is still subpar, particularly when fitting nonlinear transforms. Here we propose Registration by Regression (RbR), a novel atlas registration framework that: is highly robust and flexible; can be trained with cheaply obtained data; and operates on a single channel, such that it can also be used as pretraining for other tasks. RbR predicts the (<i>x, y, z</i>) atlas coordinates for every voxel of the input scan (i.e., every voxel is a keypoint), and then uses closed-form expressions to quickly fit transforms using a wide array of possible deformation models, including affine and nonlinear (e.g., Bspline, Demons, invertible diffeomorphic models, etc.). Robustness is provided by the large number of voxels informing the registration and can be further increased by robust estimators like RANSAC. Experiments on independent public datasets show that RbR yields more accurate registration than competing keypoint approaches, over a wide range of deformation models.</p>","PeriodicalId":90799,"journal":{"name":"Biomedical image registration, ... proceedings. WBIR (Workshop : 2006- )","volume":"15249 ","pages":"205-215"},"PeriodicalIF":0.0,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12380105/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144981992","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-05-13DOI: 10.1007/978-3-030-50120-4_11
Alena Uus, Maximilian Pietsch, I. Grigorescu, Daan Christiaens, J. Tournier, Lucilio Cordero-Grande, J. Hutter, A. Edwards, J. Hajnal, M. Deprez
{"title":"Multi-channel Registration for Diffusion MRI: Longitudinal Analysis for the Neonatal Brain","authors":"Alena Uus, Maximilian Pietsch, I. Grigorescu, Daan Christiaens, J. Tournier, Lucilio Cordero-Grande, J. Hutter, A. Edwards, J. Hajnal, M. Deprez","doi":"10.1007/978-3-030-50120-4_11","DOIUrl":"https://doi.org/10.1007/978-3-030-50120-4_11","url":null,"abstract":"","PeriodicalId":90799,"journal":{"name":"Biomedical image registration, ... proceedings. WBIR (Workshop : 2006- )","volume":"1 1","pages":"111 - 121"},"PeriodicalIF":0.0,"publicationDate":"2020-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82292941","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-05-13DOI: 10.1007/978-3-030-50120-4_14
F. Lange, Stephen M. Smith, M. Bertelsen, A. Khrapitchev, P. Manger, R. Mars, J. Andersson
{"title":"Multimodal MRI Template Creation in the Ring-Tailed Lemur and Rhesus Macaque","authors":"F. Lange, Stephen M. Smith, M. Bertelsen, A. Khrapitchev, P. Manger, R. Mars, J. Andersson","doi":"10.1007/978-3-030-50120-4_14","DOIUrl":"https://doi.org/10.1007/978-3-030-50120-4_14","url":null,"abstract":"","PeriodicalId":90799,"journal":{"name":"Biomedical image registration, ... proceedings. WBIR (Workshop : 2006- )","volume":"124 1","pages":"141 - 150"},"PeriodicalIF":0.0,"publicationDate":"2020-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86289591","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-05-13DOI: 10.1007/978-3-030-50120-4_8
Diana Waldmannstetter, F. Navarro, B. Wiestler, J. Kirschke, A. Sekuboyina, Ester Molero, Bjoern H Menze
{"title":"Reinforced Redetection of Landmark in Pre- and Post-operative Brain Scan Using Anatomical Guidance for Image Alignment","authors":"Diana Waldmannstetter, F. Navarro, B. Wiestler, J. Kirschke, A. Sekuboyina, Ester Molero, Bjoern H Menze","doi":"10.1007/978-3-030-50120-4_8","DOIUrl":"https://doi.org/10.1007/978-3-030-50120-4_8","url":null,"abstract":"","PeriodicalId":90799,"journal":{"name":"Biomedical image registration, ... proceedings. WBIR (Workshop : 2006- )","volume":"48 1","pages":"81 - 90"},"PeriodicalIF":0.0,"publicationDate":"2020-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88125756","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-05-13DOI: 10.1007/978-3-030-50120-4_12
H. Breman, J. Mulders, Levin Fritz, J. Peters, John A. Pyles, J. Eck, M. Bastiani, A. Roebroeck, J. Ashburner, R. Goebel
{"title":"An Image Registration-Based Method for EPI Distortion Correction Based on Opposite Phase Encoding (COPE)","authors":"H. Breman, J. Mulders, Levin Fritz, J. Peters, John A. Pyles, J. Eck, M. Bastiani, A. Roebroeck, J. Ashburner, R. Goebel","doi":"10.1007/978-3-030-50120-4_12","DOIUrl":"https://doi.org/10.1007/978-3-030-50120-4_12","url":null,"abstract":"","PeriodicalId":90799,"journal":{"name":"Biomedical image registration, ... proceedings. WBIR (Workshop : 2006- )","volume":"16 1","pages":"122 - 130"},"PeriodicalIF":0.0,"publicationDate":"2020-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73965835","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号