一种基于模式控制自由变形的非刚性医学图像配准算法

目的 对手术前MRI／CT图像和手术中超声图像两种模态下都可见的血管结构进行配准。方法 提出一种基于自由变形模型的多模态医学图像的非刚性配准的方法。当两个图像对准时，一种图像中的血管中心点就会对应着另一种图像下灰度脊点。对于全局变换采用刚性变换，而对于局部的形变，采用一种基于模式控制B样条的自由变形模型（FFD）来描述。配准算法采用遗传算法和共轭梯度法结合的优化策略来最小化目标函数。结果 我们设计了两个实验，分别应用于体模和临床数据来评价我们的算法。这种方法是连续而且准确的。最后的变换参数的均方差值是亚像素、亚毫米级的，在0．010弧度以内的。结论 实验结果显示本方法从配准精度和收敛速度上都得到了良好的效果。可以有效地应用于超声图像导航手术系统。     

建立PACS系统的构思

PACS全称为医学影像存档与通讯系统。是近年来随着数字成像技术、计算机技术和网络技术的进步而迅速发展起来的、它能够解决医学图像的获取、显示、存贮、传送和管理，是一种高效率、无胶片化影像系统。其主要作用有：联接不同的影像设备（CT、MR、XRAY、超声、核医学等）；存储与管理图像；图像的调用与后处理。PACS在结构上主要由医学图像获取设备、大容量数据存储及数据库管理系统、图像显示和处理系统以及影像传输网络等多个部分组成，保证PACS成为全开放式系统的重要的网络标准和协议DICOM3．0。     

CT、MR图像融合技术临床应用研究

目的 利用医学图像融合技术为临床提供新的诊断信息。方法 选取30例（男18例、女12例）颅脑病变患者为研究对象,其中20例在1－2周内分别进行了CT和MR检查,10例在CT确诊后,行MR复查;将此图像数据运用Legendre矩找出图像的质心和主轴,进而完成图像的平移、缩放和旋转,以实现CT和MR图像的融合。结果 在30例CT和MR图像融合中,使二者图像信息相互补充的有28例,较单纯地观察CT或MR图像能明确判断病变发展趋势的有19例,手术证实的4例,但有2例图像融合后无明显的优越性。结论 不同来源的多模态图像进行融合,可为临床医师明确诊断、设计手术、放疗方案提供有利信息。在融合算法上,利用Legendre矩完成运算不失为一种比较直接、快速、简洁的方法。     

超声弹性成像技术在乳腺疾病诊断与鉴别诊断中的应用

超声弹性成像是一种新的超声成像方法,它通过获取有关组织弹性信息进行成像,弥补了X射线、超声成像（US）、磁共振成像（MRI）、计算机断层扫描（CT）等传统医学成像模态不能直接提供组织弹性信息的不足,具有无创、简单、价廉、容易应用等优点,被广泛应用于临床,     

基于解剖知识模型的医学图像分割方法研究

目的提高图像分割技术的自动化程度和可靠性。方法提出了一个基于知识模型的医学CT图像的分害方法,系统由解剖知识模型、图像处理程序和推理机组成。模块之间的通讯由黑板控制。结果通过在胸部CT图像处理中的应用,该方法减少了人工干预,得到较满意的分割结果一结论该方法提高了医学图像分割的自动化程度和可靠性。由于具有扩展性,该方法为基于知识医学图像的处理提供一个通用的模式。     

CT影像组学模型及深度学习技术预测肺腺癌EGFR突变

【摘要】目的：探讨基于CT图像建立的影像组学模型及深度学习模型在预测肺腺癌表皮生长因子受体（EGFR）突变中的价值。方法：回顾性分析228例经手术病理证实的肺腺癌患者的CT图像,其中EGFR突变型116例,野生型112例。由两位放射科医师各自独立在CT图像上沿肺癌病灶边缘手动逐层勾画感兴趣区（ROI）获得病灶全容积ROI后提取影像组学特征。比较两位医师提取的影像组学特征的一致性,自高年资医师提取的特征中选取组内相关系数大于0.7的影像组学特征纳入研究。分别按照70%和30%的比例将所有病灶随机划分为训练集和验证集。在训练集中利用LASSO回归方法对影像组学特征进行筛选后,分别建立影像组学评分（Radscore）、随机森林（RF）和支持向量机（SVM）三种影像组学模型。此外,将训练集图像输入ResNet深度学习网络中建立深度学习模型。在验证集中对上述4个模型进行验证,计算敏感度、特异度和ROC曲线下面积（AUC）来评价不同模型的预测效能。结果：自CT图像中共提取了306个一致性良好的影像组学特征,经筛选后获得9个最佳特征用于建立影像组学模型。在验证集中,SVM模型的AUC（0.813）高于Radscore（0.761）和RF模型（0.775）,但差异无统计学意义（P=0.089和0.330）;ResNet模型的AUC为0.916,高于SVM模型（0.813）、Radscore（0.761）和RF模型（0.775）。ResNet模型与Radscore和RF模型间AUC的差异具有统计学意义（P=0.031和0.043）,与SVM模型间AUC的差异无统计学意义（P=0.106）。ResNet模型的敏感度为0.879,高于SVM模型（0.771）、Radscore（0.818）和RF模型（0.743）。ResNet模型的特异度为0.914,高于SVM模型（0.758）、Radscore（0.714）和RF模型（0.727）。结论：基于CT图像的影像组学模型能够较好地预测EGFR基因突变,深度学习技术可以有效提高模型的预测准确性。     

基于深度学习方法的伪CT图像合成技术研究及在放疗中的应用进展

随着医学图像合成任务复杂度的提高和对临床放疗精度的需求,深度学习算法在伪CT图像合成与分析中的角色越发重要。本文根据图像的模态种类对基于深度学习方法下的伪CT图像合成技术进行归类与分析,并介绍其在放疗应用中的最新进展。     

曝光量:自动化解剖学特异性CT射线曝光量选取的质量保证和辐射监测

摘要目的开发和验证一种信息工具库,使得能从光学字符识别的现有数字图像档案中获取解剖学特异性CT的射线曝光量(容积CT剂量指数和剂量长度积)CT剂量报告的屏幕截图(剂量屏幕)结合数字成像和通信医学属性。材料与方法这一机构审查委员会批准的遵循HIPAA的研究在一个大型城市保健医疗提供的服务网络中完成。数据来自2000—2010年之间随机抽样的CT样本;这些样本的图像包含在企业图像档案中,其中包括从一个成熟的学术性三级中心医院及其附属网站获得的图像,涵盖有肿瘤中心、社区医院、门诊影像中心,以及其他设施所引入的图像。     

肺弥漫性病变的MIP、MinIP重建

最大密度投影（MIP）及最小密度投影（MinIP）是一种应用广泛的CT及MR图像后处理技术,本文综述近年来的文献,就CT的MIP及MinIP图像与螺旋CT、高分辨率CT（HRCT）图像相比较,对肺弥漫性疾病的诊断价值进行了总结。     

采用CT断层扫描图像构建DeBakeyⅢ型主动脉夹层三维血流动力学数值模拟分析模型

目的 探索简便、精确的方法构建仿真的DeBakey Ⅲ型主动脉夹层血流动力学计算机模拟分析三维模型的方法.方法 利用Siemens Sensation Cardiac.64层螺旋CT薄层扫描技术,基于1 mm层厚获取4种临床常见典型形态的DeBakey Ⅲ型主动脉夹层连续断层DICOM格式图像,导入医学图像处理软件Materialise MIMICS v12.11,界定目标区域后生成三维动脉模型,经网格优化处理去除低质量及相交面网格,保存结果输出,导入Tgrid 5.0软件,对主动脉面网格模型进行几何修复,使面网格扭曲率<0.75,采用非结构化四面体网格生成DeBakey Ⅲ型主动脉夹层血流动力学分析计算机模拟体网格模型,并对所构建模型进行血流属性、流场边界等界定,初步验证模型的有效性.结果 通过初步计算求解,确定所构建的DeBakey Ⅲ型主动脉模型分别包含1857030,1820501,1844181及1814914 4个四面体单元.结论 利用64层螺旋CT薄层扫描技术获取DICOM格式连续断层CT图像,可快速、准确地构建DeBakeyⅢ型主动脉夹层血流动力学分析计算机模型,为进一步的计算流体力学分析奠定了良好的基础.     

Dual-modality brain PET-CT image segmentation based on adaptive use of functional and anatomical information

Dual medical imaging modalities, such as PET-CT, are now a routine component of clinical practice. Medical image segmentation methods, however, have generally only been applied to single modality images. In this paper, we propose the dual-modality image segmentation model to segment brain PET-CT images into gray matter, white matter and cerebrospinal fluid. This model converts PET-CT image segmentation into an optimization process controlled simultaneously by PET and CT voxel values and spatial constraints. It is innovative in the creation and application of the modality discriminatory power (MDP) coefficient as a weighting scheme to adaptively combine the functional (PET) and anatomical (CT) information on a voxel-by-voxel basis. Our approach relies upon allowing the modality with higher discriminatory power to play a more important role in the segmentation process. We compared the proposed approach to three other image segmentation strategies, including PET-only based segmentation, combination of the results of independent PET image segmentation and CT image segmentation, and simultaneous segmentation of joint PET and CT images without an adaptive weighting scheme. Our results in 21 clinical studies showed that our approach provides the most accurate and reliable segmentation for brain PET-CT images.     

磁共振模拟定位及磁共振诊断影像与定位CT融合的精准性比较

目的 与磁共振成像（MRI）诊断影像（MRIdiag）相比较,评估MRI模拟定位影像（MRIsim）与模拟CT融合的精准性,为MRIsim的进一步应用提供参考。方法 选择行MRIsim同时又有MRIdiag的患者24例,其中脑胶质瘤、鼻咽癌和前列腺癌各8例。将每位患者的MRIsim、MRIdiag影像分别与模拟CT融合,在3种图像上分别勾画危及器官（OAR）,在CT与MRIsim融合影像（F_CTMsim）、CT与MRIdiag融合影像（F_CTMdiag）上分别勾画靶区。评估基于MRIsim、MRIdiag与CT之间OAR的一致性指数（CI）、形似指数（DSC）、图像相似性指数（S）。基于F_CTMsim的靶区和CT的OAR设计IMRT计划,评估靶区和OAR的剂量学差异。结果 除了鼻咽癌患者的全脑和前列腺癌患者的盆骨外,其余OAR基于3种图像的体积值无明显差异（P>0.05）;MRIsim的所有OAR与CT比较的CI和DSC值均高于MRIdiag,其中50%的OAR差异有统计学意义（t=2.58~5.47,P<0.05）。MRIsim、MRIdiag与CT比较,S值分别为0.89、0.83（t=5.77,P<0.05）,相对于MRIdiag,MRIsim改善了S值10%（2%~56%）。OAR和靶区剂量学差异无统计学意义（P>0.05）。结论 相对于MRIdiag,将MRIsim引入放疗与模拟CT融合可以明显改善图像匹配的精准性。但两者基于刚性匹配肿瘤区域结合手动调整方法所勾画的靶区之间没有剂量学差异。     

Clinical evaluation of the computed tomography attenuation correction map for myocardial perfusion imaging: the potential for incidental pathology detection

The benefits of hybrid imaging in nuclear medicine have been proven to increase the diagnostic accuracy and sensitivity of many procedures by localizing or characterizing lesions or by correcting emission data to more accurately represent radiopharmaceutical distribution. Single-photon emission computed tomography/computed tomography (SPECT/CT) has a significant role in the diagnosis and follow-up of ischaemic heart disease with attenuation correction data being obtained on an integrated CT scanner. Initially, the CT component of hybrid SPECT/CT systems was what could be described as low specification utilizing fixed output parameters. As technology has progressed, the CT component of newer systems has specifications that are identical to that of stand-alone diagnostic systems. Irrespective of the type of scanner used, the computed tomography attenuation correction (CTAC) for myocardial perfusion imaging produces low-quality, limited-range CT images of the chest that include the mediastinum, lung fields and surrounding soft tissues. The diagnostic potential of this data set is unclear; yet, examples exist whereby significant pathology can be identified and investigated further. Despite guidance from a number of professional bodies suggesting that evaluation of the resulting images for every medical exposure be carried out, there is no indication as to whether this should include the evaluation of CTAC images. This review aims to initiate discussion by examining the ethical, legal, financial and practical issues (e.g. CT specification and image quality) surrounding the clinical evaluation of the CTAC for myocardial perfusion imaging images. Reference to discussions that have taken place, and continue to take place, in other modalities, current European and UK legislations, and guidelines and research in the field will be made.     

Novel detector technology for clinical PET

Introduction  

Positron emission tomography (PET) is the most sensitive of all medical imaging modalities for quantitatively probing biologic processes at the molecular level. However, spatial resolution in PET is significantly inferior to that of other imaging modalities that can provide exquisite images of the anatomy, such as X-ray computed tomography (CT) or magnetic resonance (MR) imaging.     

A graphical user interface for automatic image registration software designed for radiotherapy treatment planning

Medical imaging forms a vital component of radiotherapy treatment planning and its evaluation. The integration of the useful data obtained from multiple imaging modalities for radiotherapy planning is achieved by image registration softwares. In radiotherapy planning systems, normally the computed tomography (CT) slices are kept as a standard upon which other modality images (magnetic resonance imaging [MRI], single photon emission computed tomography [SPECT], positron emission tomography [PET], etc.) are aligned—automatically or interactively. Following validation of successful registration, they are resampled and reformatted, as per the requirements. This paper defines the minimum requirements of automatic image registration software for 3-dimensional (3D) radiotherapy planning and describes the implementation of a suitable graphical user interface developed in Visual Basic (version 5). The automatic image registration (AIR) routines freely available from Dr. Roger P. Woods, UCLA, (USA) were used in this software. This software could be easily implemented and was easy to use for image processing suitable for radiotherapy planning systems.     

Image preprocessing for a picture archiving and communication system.

OBJECTIVES AND RATIONALE. In a picture archiving and communication system (PACS), images are acquired from multiple modalities and displayed on an electronic workstation. Each modality has different image characteristics. This variability must be addressed before the image is displayed. METHODS. The authors developed methods to automatically process magnetic resonance (MR), computed tomographic (CT), and computed radiography (CR) images before display and subjectively evaluated their effectiveness. RESULTS. Unwanted background successfully was automatically removed from 89.5% of 615 CR images. Of 803 chest, abdomen, and hand images 93% were automatically rotated to the correct orientation. CONCLUSIONS. Automated preprocessing of PACS images can be performed successfully, improving speed and convenience for the radiologist interpreting images at an electronic workstation.     

A graphical user interface for automatic image registration software designed for radiotherapy treatment planning

Medical imaging forms a vital component of radiotherapy treatment planning and its evaluation. The integration of the useful data obtained from multiple imaging modalities for radiotherapy planning is achieved by image registration softwares. In radiotherapy planning systems, normally the computed tomography (CT) slices are kept as a standard upon which other modality images (magnetic resonance imaging [MRI], single photon emission computed tomography [SPECT], positron emission tomography [PET], etc.) are aligned—automatically or interactively. Following validation of successful registration, they are resampled and reformatted, as per the requirements. This paper defines the minimum requirements of automatic image registration software for 3-dimensional (3D) radiotherapy planning and describes the implementation of a suitable graphical user interface developed in Visual Basic (version 5). The automatic image registration (AIR) routines freely available from Dr. Roger P. Woods, UCLA, (USA) were used in this software. This software could be easily implemented and was easy to use for image processing suitable for radiotherapy planning systems.     

Regions-based illustrative visualization of multimodal datasets

We present as novel method for the exploration of multiple overlapping volumes that provides flexibility to merge data in different ways in different regions. In each region, either one of the modalities is rendered alone or the fusion of two modalities is shown. In the regions where data is fused, the relative weights of each modality are defined with a 2D transfer function depending on the voxel's pair of property values. The regions can be defined interactively by painting on the volume. Alternatively, when one of the modalities has been pre-classified, a graph representation of the dataset is constructed, and regions can be defined as sets of voxels fulfilling a specific combination of classification criteria. In both cases, a different fusion and shading function can be defined for each region. In this way, illustrative images of the dataset can be easily generated applying effects of cutting away, ghosting and modality enhancement.     

Multimodality Image Fusion and Planning and Dose Delivery for Radiation Therapy

Image-guided radiation therapy (IGRT) relies on the quality of fused images to yield accurate and reproducible patient setup prior to dose delivery. The registration of 2 image datasets can be characterized as hardware-based or software-based image fusion. Hardware-based image fusion is performed by hybrid scanners that combine 2 distinct medical imaging modalities such as positron emission tomography (PET) and computed tomography (CT) into a single device. In hybrid scanners, the patient maintains the same position during both studies making the fusion of image data sets simple. However, it cannot perform temporal image registration where image datasets are acquired at different times. On the other hand, software-based image fusion technique can merge image datasets taken at different times or with different medical imaging modalities. Software-based image fusion can be performed either manually, using landmarks, or automatically. In the automatic image fusion method, the best fit is evaluated using mutual information coefficient. Manual image fusion is typically performed at dose planning and for patient setup prior to dose delivery for IGRT. The fusion of orthogonal live radiographic images taken prior to dose delivery to digitally reconstructed radiographs will be presented. Although manual image fusion has been routinely used, the use of fiducial markers has shortened the fusion time. Automated image fusion should be possible for IGRT because the image datasets are derived basically from the same imaging modality, resulting in further shortening the fusion time. The advantages and limitations of both hardware-based and software-based image fusion methodologies are discussed.     

Positron emission tomography/computed tomography

Accurate anatomical localization of functional abnormalities obtained with the use of positron emission tomography (PET) is known to be problematic. Although tracers such as (18)F-fluorodeoxyglucose ((18)F-FDG) visualize certain normal anatomical structures, the spatial resolution is generally inadequate for accurate anatomic localization of pathology. Combining PET with a high-resolution anatomical imaging modality such as computed tomography (CT) can resolve the localization issue as long as the images from the two modalities are accurately coregistered. However, software-based registration techniques have difficulty accounting for differences in patient positioning and involuntary movement of internal organs, often necessitating labor-intensive nonlinear mapping that may not converge to a satisfactory result. Acquiring both CT and PET images in the same scanner obviates the need for software registration and routinely provides accurately aligned images of anatomy and function in a single scan. A CT scanner positioned in line with a PET scanner and with a common patient couch and operating console has provided a practical solution to anatomical and functional image registration. Axial translation of the couch between the 2 modalities enables both CT and PET data to be acquired during a single imaging session. In addition, the CT images can be used to generate essentially noiseless attenuation correction factors for the PET emission data. By minimizing patient movement between the CT and PET scans and accounting for the axial separation of the two modalities, accurately registered anatomical and functional images can be obtained. Since the introduction of the first PET/CT prototype more than 6 years ago, numerous patients with cancer have been scanned on commercial PET/CT devices worldwide. The commercial designs feature multidetector spiral CT and high-performance PET components. Experience has demonstrated an increased level of accuracy and confidence in the interpretation of the combined study as compared with studies acquired separately, particularly in distinguishing pathology from normal, physiologic tracer uptake and precisely localizing abnormal foci. Combined PET/CT scanners represent an important evolution in technology that has helped to bring molecular imaging to the forefront in cancer diagnosis, staging and therapy monitoring.