miLBP: a robust and fast modality-independent 3D LBP for multimodal deformable registration.
miLBP: a robust and fast modality-independent 3D LBP for multimodal deformable registration.
复制标题
miLBP：用于多模态变形配准的鲁棒且快速的独立于模态的 3D LBP
DOI：
10.1007/s11548-016-1407-2
复制
发表时间：
2016-06
影响因子：
3
通讯作者：
Song Z
中科院分区：
文献类型：
--
作者：
Jiang D;Shi Y;Yao D;Wang M;Song Z
目的计算机辅助干预通常依赖于多模态形变配准来提供补充信息。然而，多模态可变形配准仍然是一项具有挑战性的任务。方法引入一种新的鲁棒、快速的与模态无关的三维二值描述子miLBP，该描述子将局部自相似原理与局部二值模式形式相结合，能够鲁棒地提取不同模态下三维体的相似几何特征。miLBP是一个位串，可以通过简单地设定体素距离的阈值来计算。此外，利用汉明距离可以有效地评价描述子的相似度。结果将milbp与向量值自相似上下文（SSC）在人工图像和临床环境中进行比较。结果表明，miLBP在多模态提取局部几何特征方面比SSC具有更强的鲁棒性，在不同的配准场景下取得了更高的配准精度。此外，在术前磁共振成像和术中超声图像之间最具挑战性的配准中，我们的方法在准确性和速度方面都明显优于最先进的方法（一个病例29.2秒）。结论miLBP的配准性能和配准速度表明该方法具有应用于时效性强的术中计算机辅助干预的潜力。
PurposeComputer-assisted intervention often depends on multimodal deformable registration to provide complementary information. However, multimodal deformable registration remains a challenging task.MethodsThis paper introduces a novel robust and fast modality-independent 3D binary descriptor, called miLBP, which integrates the principle of local self-similarity with a form of local binary pattern and can robustly extract the similar geometry features from 3D volumes across different modalities. miLBP is a bit string that can be computed by simply thresholding the voxel distance. Furthermore, the descriptor similarity can be evaluated efficiently using the Hamming distance.ResultsmiLBP was compared to vector-valued self-similarity context (SSC) in artificial image and clinical settings. The results show that miLBP is more robust than SSC in extracting local geometry features across modalities and achieved higher registration accuracy in different registration scenarios. Furthermore, in the most challenging registration between preoperative magnetic resonance imaging and intra-operative ultrasound images, our approach significantly outperforms the state-of-the-art methods in terms of both accuracy () and speed (29.2 s for one case).ConclusionsRegistration performance and speed indicate that miLBP has the potential of being applied to the time-sensitive intra-operative computer-assisted intervention.