Accelerating image registration of MRI by GPU-based parallel computation

Automatic image registration for MRI applications generally requires many iteration loops and is, therefore, a time-consuming task. This drawback prolongs data analysis and delays the workflow of clinical routines. Recent advances in the massively parallel computation of graphic processing units (GPUs) may be a solution to this problem. This study proposes a method to accelerate registration calculations, especially for the popular statistical parametric mapping (SPM) system. This study reimplemented the image registration of SPM system to achieve an approximately 14-fold increase in speed in registering single-modality intrasubject data sets. The proposed program is fully compatible with SPM, allowing the user to simply replace the original image registration library of SPM to gain the benefit of the computation power provided by commodity graphic processors. In conclusion, the GPU computation method is a practical way to accelerate automatic image registration. This technology promises a broader scope of application in the field of image registration.     

低分辨率医学图像的多图谱分割方法

基于多图谱的图像分割方法因其分割精度高和鲁棒性强,在医学图像分割领域被广泛研究,主要包含图像配准和标签融合两个步骤.目前对多图谱分割方法的研究通常都是在图谱图像和待分割目标图像具有相同分辨率的情况下展开的.然而,由于受图像采集时间,采集设备等影响,临床实践中采集的影像大多是低分辨率数据,使得目前在影像研究中广泛使用的方法无法有效应用于临床实践.因此,针对这一问题,我们结合图像超分辨率恢复方法,提出了精确鲁棒的低分辨率医学图像的多图谱分割方法,实验结果显示提出的方法显著地提高了多图谱分割方法的分割精度.     

一种基于物体质心的图像自动化校准方法

图像校准是将任意相邻两图像都具有公共部分的一列图像合成一副全图的图像处理过程.而图像自动化校准的难点在于对一组图像公共部分特征信息的提取和自动匹配,尤其是自动匹配和图像校准的误差分析.本文基于放射变换不变性,给出了一种基于特征信息质心的自动匹配方法,与目前国际上主流特征提取和特征校准方法相比,本文方法较准确,且适用于卫星图片及一些MRI和CT 图片.本文方法的另一优点是具有线性计算复杂度.     

Measurement of Spatial-Frequency and Amplitude Responses of a CCD-Matrix Image-Registration Device

A unique method for studying spatial-frequency and amplitude responses of an image-registration device on the basis of a CCD matrix is developed. A monochromatic light-intensity distribution arising from Fraunhofer diffraction on two equal slits is used as an input signal. A quantitative approach to a spatial resolution of the CCD matrix for registration of a periodic image is developed. The dependence of the rms contrast degree of the output signal on the spatial frequency of the image recorded is studied experimentally. The optical scheme developed and the method proposed for treatment of a digitized signal allow one to determine the amplitude response of the registration device over the whole surface of the CCD matrix and to determine the admissible level of the average intensity of the radiation registered.     

基于AM测度及混合优化算法的图像配准方法

介绍了一种基于边缘特征的图像配准方法。首先利用Canny算子提取图像的边缘;然后以AM测度作为配准准则,在优化AM峰值和配准参数的时候,先用遗传算法将配准参数收敛到最优值附近,再使用Powell法进行局部搜索;最后对待配准图像进行刚体变换及双线性插值,从而实现图像配准。实验结果表明,此优化方法不仅提高了计算速度,而且提高了配准精度,配准结果达到了亚像素级水平。     

Feature tracking for projection registration in laboratory-scale reflective tomography laser radar imaging

In range-resolved reflective tomography laser radar imaging system, for accurate image reconstruction, all projections must be aligned to the center of target rotation. Due to the unknown location of the axis of rotation and random target translation, the center of target rotation in each projection cannot be measured accurately and images reconstructed based on range-resolved projections without registration would be dislocated. Here, we apply a feature tracking method to align projections data for tomographic reconstruction in laboratory-scale laser radar imaging. The target was placed on a spin table with an unknown and fixed axis, and the oscillatory motion of the target translation was simulated by a second order inertia filter to accord with the real satellites vibrations. The experimental simulation results demonstrate the effectiveness of this method to improve images reconstruction equality. The application of this method to actual radar imaging in remote distance was discussed and the differences were compared with the lab setups.     

Interpolation algorithm for recovering the missing phase values in 3D measurement

In the three-dimensional (3D) phase measurement, some marks are usually adhered to the object in order to make the 3D registration process faster and easier. As covered with marks, local phase data are missing and have to be interpolated later. Considering the phase distribution nearby the marks, a gradient estimate (GE) interpolation algorithm is provided here. This algorithm recovers one pixel's missing phase value with the average of the estimated values which is calculated by gradients in eight directions nearby. Since this algorithm is a local processing, the missing phase values should be interpolated from the edge of the marks to the center. In the computer simulation and the practical experiment, compared with the same-size neighborhood mean (NM) algorithm and the Gerchberg-Saxton (GS) algorithm, this new algorithm achieves very good fit results with the least time. So it can be used as a practical tool for automatic missing phase interpolation.     

Enhanced FEM-based modeling of brain shift deformation in Image-Guided Neurosurgery

We consider the problem of improving outcomes for neurosurgery patients by enhancing intraoperative navigation and guidance. Current navigation systems do not accurately account for intraoperative brain deformation. We focus on the brain shift deformation that occurs just after the opening of the skull and dura. The heart of our system is a nonrigid registration technique using a biomechanical model. We specifically work on two axes: the representation of the structures in the biomechanical model and the evaluation of the surface landmark displacement fields between intraoperative MR images. Using the modified Hausdorff distance as an image similarity measure, we demonstrate that our approach significantly improves the alignment of the intraoperative images.     

针对平移和旋转的快速电子稳像方法

针对视频图像中存在的平移抖动和旋转抖动,提出了一种基于灰度投影和模板匹配的电子稳像方法。用网格法搜索图像中灰度特征明显的区域,结合分象限法提高系统的鲁棒性,在不同坐标系下分别投影,纠正平移和旋转偏差。给出了模板制作方法和配准步骤。实验证明,该方法具有较高的精度和很高的处理速度,能够实时去除由视频图像抖动引起的噪声。     

基于均值距离测度的医学图像配准

针对互信息测度在配准医学图像时易陷入局部极值、速度慢的缺点,提出了基于均值不等式的均值距离测度.首先根据均值不等式推导出5种均值距离测度：方根-算术均值距离(SAM)、方根-几何均值距离(SGM)、方根-调和均值距离(SHM)、算术-几何均值距离(AGM)、算术-调和均值距离(AHM).然后通过人体脑部CT/MR和MR-T1/PD图像的刚体配准实验,从函数曲线、配准精度、计算时间和收敛性能方面,对互信息与5种均值距离信息测度进行了比较与分析.实验结果表明,在不损失配准精度的前提下,AHM和SAM测度可以获得更快的配准速度,对噪声有很强的鲁棒性.