Ringenberg Jordan, Deo Makarand, Devabhaktuni Vijay, Berenfeld Omer, Boyers Pamela, Gold Jeffrey
EECS Department, College of Engineering, University of Toledo, 2801 W. Bancroft Street, Toledo, OH 43606, United States.
Department of Engineering, Norfolk State University, 700 Park Avenue, Norfolk, VA 23504, United States.
Comput Med Imaging Graph. 2014 Apr;38(3):190-201. doi: 10.1016/j.compmedimag.2013.12.011. Epub 2014 Jan 2.
This paper presents a fully automatic method to segment the right ventricle (RV) from short-axis cardiac MRI. A combination of a novel window-constrained accumulator thresholding technique, binary difference of Gaussian (DoG) filters, optimal thresholding, and morphology are utilized to drive the segmentation. A priori segmentation window constraints are incorporated to guide and refine the process, as well as to ensure appropriate area confinement of the segmentation. Training and testing were performed using a combined 48 patient datasets supplied by the organizers of the MICCAI 2012 right ventricle segmentation challenge, allowing for unbiased evaluations and benchmark comparisons. Marked improvements in speed and accuracy over the top existing methods are demonstrated.
本文提出了一种从短轴心脏磁共振成像中自动分割右心室（RV）的方法。该方法结合了一种新颖的窗口约束累积器阈值技术、高斯差分（DoG）二元滤波器、最优阈值处理和形态学方法来驱动分割过程。引入了先验分割窗口约束来指导和优化该过程，并确保分割区域的适当限制。使用由2012年医学图像计算与计算机辅助干预国际会议（MICCAI）右心室分割挑战赛的组织者提供的48个患者数据集组合进行训练和测试，从而实现无偏评估和基准比较。结果表明，与现有的最佳方法相比，该方法在速度和准确性上有显著提高。
