Ebrahimi Mehran, Siegler Peter, Modhafar Amen, Holloway Claire M B, Plewes Donald B, Martel Anne L
Faculty of Science, University of Ontario Institute of Technology, 2000 Simcoe St N, Oshawa, ON, Canada, L1H 7K4.
Phys Med Biol. 2014 Apr 7;59(7):1589-605. doi: 10.1088/0031-9155/59/7/1589. Epub 2014 Mar 10.
Breast MRI is frequently performed prior to breast conserving surgery in order to assess the location and extent of the lesion. Ideally, the surgeon should also be able to use the image information during surgery to guide the excision and this requires that the MR image is co-registered to conform to the patient's position on the operating table. Recent progress in MR imaging techniques has made it possible to obtain high quality images of the patient in the supine position which significantly reduces the complexity of the registration task. Surface markers placed on the breast during imaging can be located during surgery using an external tracking device and this information can be used to co-register the images to the patient. There remains the problem that in most clinical MR scanners the arm of the patient has to be placed parallel to the body whereas the arm is placed perpendicular to the patient during surgery. The aim of this study is to determine the accuracy of co-registration based on a surface marker approach and, in particular, to determine what effect the difference in a patient's arm position makes on the accuracy of tumour localization. Obtaining a second MRI of the patient where the patient's arm is perpendicular to body axes (operating room position) is not possible. Instead we obtain a secondary MRI scan where the patient's arm is above the patient's head to validate the registration. Five patients with enhancing lesions ranging from 1.5 to 80 cm(3) in size were imaged using contrast enhanced MRI with their arms in two positions. A thin-plate spline registration scheme was used to match these two configurations. The registration algorithm uses the surface markers only and does not employ the image intensities. Tumour outlines were segmented and centre of mass (COM) displacement and Dice measures of lesion overlap were calculated. The relationship between the number of markers used and the COM-displacement was also studied. The lesion COM-displacements ranged from 0.9 to 9.3 mm and the Dice overlap score ranged from 20% to 80%. The registration procedure took less than 1 min to run on a standard PC. Alignment of pre-surgical supine MR images to the patient using surface markers on the breast for co-registration therefore appears to be feasible.
保乳手术前常进行乳腺磁共振成像（MRI），以评估病变的位置和范围。理想情况下，外科医生在手术过程中也应能够利用图像信息来指导切除操作，这就要求将磁共振图像进行配准，使其与患者在手术台上的位置相符。磁共振成像技术的最新进展使得在仰卧位获取患者的高质量图像成为可能，这显著降低了配准任务的复杂性。成像过程中放置在乳房上的表面标记物可在手术期间使用外部跟踪设备进行定位，该信息可用于将图像与患者进行配准。然而，仍存在一个问题，即在大多数临床磁共振扫描仪中，患者的手臂必须与身体平行放置，而在手术过程中手臂是垂直于患者放置的。本研究的目的是确定基于表面标记物方法的配准准确性，特别是确定患者手臂位置的差异对肿瘤定位准确性有何影响。在患者手臂垂直于身体轴线（手术室位置）的情况下获取第二次MRI是不可能的。相反，我们获取了一次辅助MRI扫描，其中患者的手臂位于患者头部上方，以验证配准情况。对5例增强病变大小在1.5至80立方厘米之间的患者，在其手臂处于两个位置时使用对比增强MRI进行成像。采用薄板样条配准方案来匹配这两种配置。配准算法仅使用表面标记物，不采用图像强度。对肿瘤轮廓进行分割，并计算质心（COM）位移和病变重叠的骰子系数。还研究了所用标记物数量与COM位移之间的关系。病变COM位移范围为0.9至9.3毫米，骰子重叠分数范围为20%至80%。在标准个人电脑上运行配准程序耗时不到1分钟。因此，利用乳房上的表面标记物将术前仰卧位磁共振图像与患者进行配准似乎是可行的。
