Simon Fraiser University, Burnaby, BC, Canada.
Ecole de Technologie Superieure (ETS), Montreal, QC, Canada.
Comput Med Imaging Graph. 2016 Dec;54:27-34. doi: 10.1016/j.compmedimag.2016.09.004. Epub 2016 Oct 1.
This study investigates a fast integral-kernel algorithm for classifying (labeling) the vertebra and disc structures in axial magnetic resonance images (MRI). The method is based on a hierarchy of feature levels, where pixel classifications via non-linear probability product kernels (PPKs) are followed by classifications of 2D slices, individual 3D structures and groups of 3D structures. The algorithm further embeds geometric priors based on anatomical measurements of the spine. Our classifier requires evaluations of computationally expensive integrals at each pixel, and direct evaluations of such integrals would be prohibitively time consuming. We propose an efficient computation of kernel density estimates and PPK evaluations for large images and arbitrary local window sizes via integral kernels. Our method requires a single user click for a whole 3D MRI volume, runs nearly in real-time, and does not require an intensive external training. Comprehensive evaluations over T1-weighted axial lumbar spine data sets from 32 patients demonstrate a competitive structure classification accuracy of 99%, along with a 2D slice classification accuracy of 88%. To the best of our knowledge, such a structure classification accuracy has not been reached by the existing spine labeling algorithms. Furthermore, we believe our work is the first to use integral kernels in the context of medical images.
本研究提出了一种快速积分核算法，用于对轴向磁共振图像（MRI）中的脊椎和椎间盘结构进行分类（标记）。该方法基于特征层次结构，其中通过非线性概率乘积核（PPK）对像素进行分类，然后对 2D 切片、单个 3D 结构和 3D 结构组进行分类。该算法进一步基于脊柱的解剖学测量嵌入了几何先验知识。我们的分类器在每个像素都需要评估计算成本高昂的积分，如果直接评估这些积分将非常耗时。我们提出了一种有效的方法，通过积分核来计算大规模图像和任意局部窗口大小的核密度估计和 PPK 评估。我们的方法仅需用户点击一次即可处理整个 3D MRI 容积，几乎可以实时运行，并且不需要密集的外部训练。通过对来自 32 名患者的 T1 加权轴向腰椎数据集的综合评估，我们的结构分类准确率达到 99%，2D 切片分类准确率达到 88%。据我们所知，现有的脊柱标记算法尚未达到如此高的结构分类准确率。此外，我们相信我们的工作是首次在医学图像上下文中使用积分核。
