D'Agostino Emiliano, Maes Frederik, Vandermeulen Dirk, Suetens Paul
Katholieke Universiteit Leuven, Faculties of Medicine and Engineering, Medical Imaging Center (Radiology - ESAT/PSI), University Hospital Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium.
Inf Process Med Imaging. 2007;20:320-32. doi: 10.1007/978-3-540-73273-0_27.
In this paper an algorithm for atlas-to-image non-rigid registration based on regional entropy minimization is presented. Tissue class probabilities in the atlas are registered with the intensities in the target image. The novel aspect of the paper consists in using tissue class probability maps that include the three main regions (for the brain, white matter, gray matter and csf) and a further partitioning thereof. For example, gray matter is further subdivided into basal ganglia (each of them defining its own class) and the rest (of gray matter). This guarantees a regional entropy minimization instead of just a global one. In other words, the local labels in the atlas will be adjusted in order to obtain the best explanation for the intensity distribution in the corresponding subregion of the target image.
本文提出了一种基于区域熵最小化的图谱到图像非刚性配准算法。图谱中的组织类别概率与目标图像中的强度进行配准。本文的新颖之处在于使用了组织类别概率图，该图包括三个主要区域（针对大脑、白质、灰质和脑脊液）及其进一步划分。例如，灰质进一步细分为基底神经节（每个基底神经节定义其自己的类别）和其余部分（灰质）。这保证了区域熵最小化，而不仅仅是全局熵最小化。换句话说，将调整图谱中的局部标签，以便为目标图像相应子区域中的强度分布获得最佳解释。
