基于Demons的微分同胚非刚性配准研究

Demons算法的一个局限是它无法处理大形变,且不能产生微分同胚的变换以满足计算解剖学的形态分析需要。利用李群中的指数映射,把原来Demons形变场相加的更新方式改进为若干次形变场间的复合,同时又保证了较高的运算效率。实验表明,新算法能配准大形变问题,且真实颅脑CT的实验结果与Demons算法的结果相近,但产生的形变场光滑可逆,并有相对更小的形变能量。     

基于Demons算法的多模态医学图像非刚性配准研究

非刚性医学图像配准是医学图像研究领域的热门专题之一,具有重要的临床应用价值。本文提出一种改进的Demons算法,将灰度守恒模型和局部结构张量守恒模型结合起来,构造一个新的能量函数处理多模态配准问题,然后采用L-BFGS算法优化能量函数,解决复杂三维数据的优化问题,并采用多尺度分层细化的思想解决大形变配准。实验表明,本文算法对大形变和多模态三维医学图像配准都有较好的效果。     

GPU框架下基于改进Demons算法的CT-CBCT图像变形配准研究

目的:针对自适应放射治疗中的关键技术--CT和CBCT图像变形配准问题,提出一种基于图形处理器GPU的改进Demons配准算法。方法:通过匹配CT和CBCT图像相应体素局部邻域点集的k阶样本矩,计算CBCT图像每一个体素CT值的线性变换系数,并在每一次Demons迭代过程中,对原CBCT图像逐体素做CT值线性变换,最后利用Demons公式计算变形场。结果:5例临床头颈部肿瘤患者的CT和CBCT图像配准结果表明,改进后算法不受CT和CBCT图像CT值强度不一致的影响,能快速、精确的完成图像的变形配准。结论:基于GPU框架的改进Demons算法可以快速精确完成CT-CBCT图像变形配准,较好的满足了临床对于快速变形图像配准的要求。     

基于Demons算法的弹性配准在放疗锥形束CT图像配准中的应用

将基于经典Demons算法与加速Demons算法的弹性配准模型应用到放疗患者不同分次间的锥形束CT(CBCT)图像中,为进一步分析肿瘤及危及器官的变化提供软件支持。通过Matlab软件编写三维弹性配准程序,并用此程序对宫颈癌放疗患者不同分次时的两组锥形束CT图像进行仿真实验验证。结果表明经典Demons算法配准前后对比,最小均方误(MSE)减少59.7%,相关系数(CC)提高了11.0%;加速Demons算法配准前后对比,MSE减少40.1%,CC提高7.2%。实验验证上述两种基于Demons模型的弹性配准在CBCT图像配准中取得较好的配准效果,但对细微差别处仍显得精度不够,且整个配准时间较长,如要应用到在线自适应放疗中仍需进一步提高形变的精度并减少配准时间。     

基于回归预测的肺4D-CT多相位配准

针对由于灰度不均和局部形变较大引起的肺4D-CT图像配准精度不足问题,提出基于回归的逐块预测初始形变的方法。新方法的核心思想是:配准一幅浮动图像至参考图像时,利用与浮动图像相对应的不同相位的图像信息进行形变场预测。首先,利用已有配准算法配准不同相位的图像至参考图像,得到各图像对应的形变场;再将图像和对应形变场分块作为训练集,利用多维支持向量回归机建立回归模型;将浮动图像分块输入回归模型中,预测出初始形变场,从而得到中间图像,并最终细化配准中间图像与参考图像。采用由德克萨斯安德森肿瘤中心DIR实验室采集并公开的数据集,评价所提出的算法。实验量化评价结果表明,与传统的Active Demons算法、Spectral Log-Demons算法相比,图像的均方误差平方和显著降低(Active Demons算法49.34±23.92,Spectral Log-Demons算法31.81±15.09,所提出算法18.97±5.75,P<0.05),相关系数显著提高(Active Demons算法0.952±0.022,Spectral Log-Demons算法0.967±0.015,所提出算法0.980±0.006,P<0.05)。同时,视觉评价结果显示,所提出算法能够获得更准确的配准图像。     

基于多分辨率GMI Demons算法的18F-FDG PET-CT图像配准在食道癌病例中的应用

18F-FDG PET和CT图像的精确配准在肿瘤的放射治疗中具有重要的临床研究意义,本研究采用全局刚性粗配准对食道癌病例中的PET和CT图像进行预处理,尽可能地减小摆位误差,然后使用基于互信息梯度的Demons算法(GMI Demons)进行局部形变配准,有效弥补内部器官误差,另外为了加快配准过程,保持图像的鲁棒性的同时避免局部极值,在形变配准前使用多分辨率图像金字塔结构。通过对10例食道癌病例的定量分析,最大互信息值结果说明经GMI Demons算法配准之后的图像精度比基于MI算法要提高8.046%±0.041%,配准前后临床上肿瘤靶区(GTV)大小的变化,说明经GMI Demons算法配准之后的GTV大小比基于MI算法配准之后的精度提高8.022%±0.044%。两种定量结果的一致性和通过对图像的定性分析,说明该配准策略可以快速地精确肿瘤靶区位置,在制定精确的放疗计划和实际的临床应用中具有研究意义。     

基于自适应微分同胚多分辨率Demons算法的多模态磁共振图像配准

微分同胚Demons能保证变形场的光滑可逆性,避免不合理形变的产生,但迭代次数需手动设定,且对配准结果影响较大。为解决此问题,提出自适应微分同胚的多分辨率Demons算法。首先利用非刚性配准的优化理论框架和多分辨率的策略,引入基于灰度的相似度能量函数,并设置配准终止条件,最终实现迭代次数的自适应。用check board图像,相同模态和不同模态的磁共振图像进行测试,采用配准评价指标进行定量分析,并分析不同驱动力和参数对配准结果的影响。实验结果表明:对于相同模态的磁共振图像,均方误差为514.796 5,归一化互相关系数为0.999 3,结构相似度为0.994 8;对于不同模态的磁共振图像,均方误差为1 354.1,归一化互相关系数为0.593 5,结构相似度为0.511 6;该算法均方误差最小,归一化相关系数、结构相似度最高;该算法具有高效性和鲁棒性,可用于磁共振图像的非刚性配准。     

带有预处理的计划CT和在线CT的形变配准方法

CT扫描中,水溶性碘造影的存在使得计划CT和在线CT图像中血管内的HU值出现非常大的偏差,从而导致计划CT和在线CT图像错配。针对该问题,本研究提出了一种基于预处理的计划CT和在线CT形变配准方法。首先,根据CT图像组织和结构的信息,利用阈值分割方法分割出血管,并将所有分割中最大的联通区域作为初始分割的强化血管;其次,利用分割得到的强化血管区域外扩5 mm,作为外扩的强化血管,并将血管用固定的HU值进行填充;最后,对完成填充后的图像利用Demons算法进行形变配准。实验结果显示本文提出的带有预处理的形变配准方法,可以较好地解决水溶性碘造影剂引起的CT错配问题。     

基于微分同胚Demons形变配准算法获取肺通气图

基于四维计算机断层扫描图像(4D-CT)生成肺通气功能图的研究是医学图像处理领域的一个热点问题。肺通气图可以有效识别肺的不同功能区域,依此制定的肺癌个体化放疗计划可以有针对性的保护肺的高功能区,减轻放疗对肺功能的影响。本研究提出一种基于微分同胚Demons形变配准算法获取肺通气图的方法。首先使用阈值法对4D-CT呼气末时相和吸气末时相的肺体积进行分割;然后使用微分同胚Demons形变配准算法对两个时相分割出来的肺体积进行形变配准获取其形变函数φ;最后应用雅可比行列式分析φ,经重建获得肺通气图的三维分布。结果发现该方法充分考虑了图像的空间信息及灰度信息,无需借助其他功能影像设备,即可快速获取高分辨率的肺通气功能图像。     

基于局部联合熵梯度的双向多分辨率Demons算法

针对active demons算法易受到参数设置的影响,无法有效解决大形变场的配准问题,本研究提出了基于局部联合熵梯度的双向多分辨率demons算法。利用在配准过程中两幅图像的互信息不断增加,局部联合熵增大的规律,本研究引入两幅图像局部联合熵参数,将图像局部联合熵的梯度附加到demons驱动力中,实现了基于局部联合熵梯度的双向多分辨率demons算法。利用自然图像、MRI图像和CT图像测试本算法的优越性,与active demons,diffeomorphic demons算法进行对比分析,采用均方误差、归一化互相关系数和结构相似度对配准结果进行定量评价。本算法归一化互相关系数和结构相似度最高,均方误差最小。通过分析权重系数的影响和设置合适的参数,本算法可应用于大形变的医学图像配准,具有一定的临床应用价值。     

Direct aperture optimization for online adaptive radiation therapy

This paper is the first investigation of using direct aperture optimization (DAO) for online adaptive radiation therapy (ART). A geometrical model representing the anatomy of a typical prostate case was created. To simulate interfractional deformations, four different anatomical deformations were created by systematically deforming the original anatomy by various amounts (0.25, 0.50, 0.75, and 1.00 cm). We describe a series of techniques where the original treatment plan was adapted in order to correct for the deterioration of dose distribution quality caused by the anatomical deformations. We found that the average time needed to adapt the original plan to arrive at a clinically acceptable plan is roughly half of the time needed for a complete plan regeneration, for all four anatomical deformations. Furthermore, through modification of the DAO algorithm the optimization search space was reduced and the plan adaptation was significantly accelerated. For the first anatomical deformation (0.25 cm), the plan adaptation was six times more efficient than the complete plan regeneration. For the 0.50 and 0.75 cm deformations, the optimization efficiency was increased by a factor of roughly 3 compared to the complete plan regeneration. However, for the anatomical deformation of 1.00 cm, the reduction of the optimization search space during plan adaptation did not result in any efficiency improvement over the original (nonmodified) plan adaptation. The anatomical deformation of 1.00 cm demonstrates the limit of this approach. We propose an innovative approach to online ART in which the plan adaptation and radiation delivery are merged together and performed concurrently-adaptive radiation delivery (ARD). A fundamental advantage of ARD is the fact that radiation delivery can start almost immediately after image acquisition and evaluation. Most of the original plan adaptation is done during the radiation delivery, so the time spent adapting the original plan does not increase the overall time the patient has to spend on the treatment couch. As a consequence, the effective time allotted for plan adaptation is drastically reduced. For the 0.25, 0.5, and 0.75 cm anatomical deformations, the treatment time was increased by only 2, 4, and 6 s, respectively, as compared to no plan adaptation. For the anatomical deformation of 1.0 cm the time increase was substantially larger. The anatomical deformation of 1.0 cm represents an extreme case, which is rarely observed for the prostate, and again demonstrates the limit of this approach. ARD shows great potential for an online adaptive method with minimal extension of treatment time.     

基于一种新型模糊增强算法的眼底图像增强

眼底图像是细节较多的复杂图像,Ｐａｌ等人的提出的模糊增强算法对增强这类图像有着较好的效果。但这种算法存在着处理速度较性和容易损失信息的缺陷。针对这一情况,本文提出了一种改进后的模糊增强算法。实验证明,改进后的模糊增强算法对眼底图像的增强有着更好的效果。     

一种改进的心脏电兴奋矢量传播算法

矢量传播算法是一种新的心脏电兴奋传播算法,针对其缺陷,通过引入判断新波源的方法,提出了改进的矢量传播算法.仿真结果表明,该算法克服了原算法中不能准确计算介质中有阻碍情况下兴奋传播过程的局限性,同时保留了原算法精确快速的优点,有助于更准确地仿真某些病理波形.     

Quantitative evaluation of free-form deformation registration for dynamic contrast-enhanced MR mammography

In this paper, we present an evaluation study of a set of registration strategies for the alignment of sequences of 3D dynamic contrast-enhanced magnetic resonance breast images. The accuracy of the optimal registration strategies was determined on unseen data. The evaluation is based on the simulation of physically plausible breast deformations using finite element methods and on contrast-enhanced image pairs without visually detectable motion artifacts. The configuration of the finite element model was chosen according to its ability to predict in vivo breast deformations for two volunteers. We computed transformations for ten patients with 12 simulated deformations each. These deformations were applied to the postcontrast image to model patient motion occurring between pre- and postcontrast image acquisition. The original precontrast images were registered to the corresponding deformed postcontrast images. The performance of several registration configurations (rigid, affine, B-spline based nonrigid, single-resolution, multi-resolution, and volume-preserving) was optimized for five of the ten patients. The images were most accurately aligned with volume-preserving single-resolution nonrigid registration employing 40 or 20 mm control point spacing. When tested on the remaining five patients the optimal configurations reduced the average mean registration error from 1.40 to 0.45 mm for the whole breast tissue and from 1.20 to 0.32 mm for the enhancing lesion. These results were obtained on average within 26 (81) min for 40 (20) mm control point spacing. The visual appearance of the difference images from 30 patients was significantly improved after 20 mm volume-preserving single-resolution nonrigid registration in comparison to no registration or rigid registration. No substantial volume changes within the region of the enhancing lesions were introduced by this nonrigid registration.     

Receiver operating characteristic study of image preprocessing for teleradiology and digital workstations

The purpose of this study was to evaluate whether digitized analog images displayed on a digital workstation can be improved by using a preprocessing algorithm, and if so, whether the quality of the resulting images can reach that of the original films. The material contained 120 difficult cases (about 50% with selected pathology). Four radiologists each evaluated half of the randomly ordered cases with the digital workstation and half of the cases with the original radiographs. The data were compared with a previous similar study, where the workstation had no option for preprocessed images. Preprocessed digital images were clearly superior to digital images without preprocessing, although for those of the highest diagnostic difficulty they were inferior to the original films. The preprocessing algorithm has improved the diagnostic quality of the digital workstation. There is room yet for improvement compared to plain films, although the current setup may be sufficient in some settings.     

基于CT图像的颅内血肿检测与分割

目的:为提高血肿分割精度,提出一种基于改进Canny算子的颅内类血肿噪声检测方法。方法:首先用区域生长算 法分割出颅脑组织,去掉颅骨等干扰信息。然后使用基于改进Canny边缘检测的方法检测颅脑边缘类血肿噪声,并与原 图像进行与运算消除该噪声。最后,通过使用OTSU适应度函数的遗传算法精准分割出颅内血肿。结果:该方法在随机 抽取的200例颅脑血肿图像中,血肿检测的准确率达到96.3%,Dice相似度达到93.5%。结论:该方法能准确、有效地检测 并分割出颅内血肿。     

改进的遗传模糊聚类算法及其在MR脑组织分割中的应用

为提高MR图像中脑组织分割的精度,针对目前遗传模糊聚类算法存在的问题,提出了改进的遗传模糊聚类算法。首先,通过完全改变遗传算法的编码方式、变异方式和交叉方式,对现有遗传算法进行改进,从而使遗传算法能获得最小的适应度函数值;然后,结合模糊聚类算法,最终得到改进的遗传模糊聚类算法。将改进的遗传模糊聚类算法应用于MR脑图像的分割,结果表明,改进的遗传模糊聚类算法的分割质量高于现有的遗传模糊聚类算法和快速模糊聚类算法。改进的遗传模糊聚类算法可以做为一种快速、全自动的MR脑图像分割工具。     

平均血糖波动幅度计算机辅助计算算法及实现

以传统人工计算方法为基础,结合利用改进微分法,寻找动态血糖数据极值点,并完善血糖有效波动的判断条件,提出一种能利用计算机辅助计算平均血糖波动幅度(MAGE)的算法,利用Visual C# 2008实现该算法的软件应用,并利用传统人工计算方法(MAGEo)与计算机辅助计算方法(MAGEc),对76例临床动态血糖监测数据(正常成人27例,2型糖尿病成人患者25例,正常孕妇24例)进行比较计算,对结果进行相关分析及两种方法的一致性评估.比较结果显示:1)MAGEc方法与MAGEo方法的计算结果对于不同受监测者分组均显著相关(正常成人组r=0.994,P＜0.01;2型糖尿病成人患者组r=0.997,P＜0.01;正常孕妇组r=0.998,P＜0.01;所有受监测者r=0.997,P＜0.01).2)Bland-Altman法分析可知,落在一致性界限外的点分别为:正常成人组3个,占11.1％ (3/27);2型糖尿病成人患者组1个,占4％( 1/25);正常孕妇组1个,占4.2％( 1/24);所有受监测者6个,占7.9％(6/76).两种方法之间具有较好的一致性.利用MAGE计算机辅助计算算法,能解决临床科研工作中通过人工计算MAGE带来的计算时间冗长、计算结果准确性低的问题,大大提高临床科研工作的效率.