Oakland University, Rochester, MI 48309, USA.
IEEE Trans Biomed Eng. 2011 Oct;58(10):2876-84. doi: 10.1109/TBME.2011.2162063. Epub 2011 Jul 14.
Prior to performing an MRI-guided breast biopsy, the radiologist has to locate the suspect lesion with the breast compressed between rigid plates. However, the suspect lesion is typically identified from a diagnostic MRI exam with the breast hanging freely under the force of gravity. There are several challenges associated with localizing suspect lesions including, patient positioning, the visibility of the lesion may fade after contrast injection, menstrual cycles, and lesion deformation. Researchers have developed finite element analysis (FEA) methodologies that simulate breast compression with the intent of reducing these challenges. In this paper, we constructed a patient-specific finite element (FE) breast model from diagnostic MR images. In addition, we constructed surfaces corresponding to the biopsy MR volume and used them to deform the FE breast mesh. The predicted results suggest that the FE breast model, in its uncompressed configuration, can be compressed to replicate the perimeter of the biopsy MR volume. The simulated lesion displacement was within 3 mm of its actual position.
在进行 MRI 引导下的乳腺活检之前，放射科医生必须将乳房压在刚性板之间以定位可疑病变。然而，可疑病变通常是从诊断性 MRI 检查中确定的，此时乳房在重力作用下自然下垂。定位可疑病变存在一些挑战，包括患者体位、病变的可见性在对比剂注射后可能会减弱、月经周期以及病变变形等。研究人员已经开发出有限元分析 (FEA) 方法，旨在模拟乳房压缩，以减少这些挑战。在本文中，我们从诊断性磁共振图像构建了一个特定于患者的有限元 (FE) 乳房模型。此外，我们还构建了与活检磁共振体积相对应的表面，并使用它们来变形 FE 乳房网格。预测结果表明，FE 乳房模型在未压缩状态下可以被压缩以复制活检磁共振体积的周长。模拟的病变位移与其实际位置相差在 3 毫米以内。
