医学图像配准

前列腺超声-磁共振图像配准

前列腺超声-磁共振图像配准能为影像引导下的前列腺介入手术提供丰富准确的信息。然而，该配准极具挑战性，因为超声和磁共振图像表现形式差异巨大、两者图像中缺少对应标记点、以及前列腺形态多样、力学属性各异。我们创新性的提出模型-表面配准方法，基于个性化统计形变模型和混合点云匹配的方法实现MR-TRUS图像配准，配准结果完全满足临床需求。

                    The construction of personalized statistical deformable model (PSDM)

                          Towards PSDM and hybrid point matching for robust MR-TRUS registration

（2）相关研究老师：倪东、王毅

（3）相关文章：

[1]Yi Wang, Jie-Zhi Cheng, Dong Ni, Muqing Lin, Jing Qin, Xiongbiao Luo, Ming Xu, XiaoyanXie, Pheng-Ann Heng, Towards Personalized Statistical Deformable Model and Hybrid Point Matching for Robust MR-TRUS Registration, IEEE Transactions on Medical Imaging, 2016, 35(2): 589-604.

[2]Yi Wang, Qingqing Zheng, Pheng-Ann Heng, Online Robust Projective Dictionary Learning: Shape Modeling for MR-TRUS Registration, IEEE Transactions on Medical Imaging, 2018, 37(4): 1067-1078.