由CT三维数据重建物体曲面剖面技术研究

用X-CT设备对被检测物体扫描后获得了一系列的二维图像,然后通过计算机图形图像学中的可视化技术从这些二维图像中重建出物体内部的三维结构。这正成为无损检测中一项重要的技术。而重建被检测物体的任意剖面的技术正是三维可视化技术的重要组成部分。本文在总结以前绘制物体任意剖面的众多方法基础上,对其中主要两种任意剖面方法进行深入探讨和比较研究,针对绘制曲面剖面的重采样困难,并结合基于体元的剖面直接生成方法的内在特点,在不影响剖面图像质量的前提下,提出了绘制曲面剖面的近似方法,而且使用数据预处理技术对其算法进行改进,大大提高了运算速度,达到实时绘制的要求。这改进的方法不仅能够绘制任意平面剖面,而且能够绘制包括球面等曲面剖面,因而它具有良好的扩展性和广阔的应用前景。     

用于颅颌面种植外科的三维立体可视化系统

目的论文对在个人PC机上对颅颌面医学图像的可视化进行研究,开发出用于颅颌面种植外科的CT断层图像三维可视化系统。方法应用3D纹理映射的方法对颅颌面CT断层图像进行三维重建,并得到其XYZ方向上的切面图像。然后利用边缘检测的方法得到眼眶的位置,并在眼眶位置进行扇形切割,得到18幅扇形切割图像。结果该系统可以帮助医生进行手术前的骨质、骨量分析,设计种植手术的过程及模拟、预测手术的种植效果。结论建立的颅颌面种植外科三维可视化系统以其硬件配置合理、软件设计新颖、多维视角、便捷快速精确等为特色。具有较强的应用价值。     

基于形变模型Level Set方法的图像分割

本文阐述了基于形变模型(Deformable Models)的Level Set分割方法的基本原理及其特点,介绍了在图像域的实现方法,实现并改进了基于该模型的Narrow Band快速算法。该算法的基本流程是:先在需要分割的目标内或外给定一封闭的初始曲线,通过Gaussian滤波后计算图像的梯度,最后通过Narrow Band 算法完成轮廓线的抽取。该算法应用于医学CT/ MRI影像以及显微图像的目标分割中,取得了较好的实验结果,证明该方法非常适合于对具有分支、突触以及拓扑结构变化的目标进行快速精确分割。文中给出了算法实现的基本流程、相关参数的选取准则和部分实验结果。实验发现算法中涉及的参数对提取的轮廓线的精确和光滑程度有较大影响。     

基于CT图像的汽车碰撞假人上衣模型点云数据提取研究

CT辅助逆向工程技术是在不破坏实物原型基础上,利用CT扫描获得物体内外表面有序点云数据,是现代产品设计与建模的重要手段。对于具有薄壁结构的产品,传统方法在利用其CT断层图像提取点云数据时,会存在轮廓线交错、断点等问题,影响着点云的提取精度。本文针对汽车碰撞假人上衣这种薄壁件表面点云数据提取困难的问题,提出一种适用于复杂薄壁件内外点云提取的新方法。实验证明,该方法能够得到高精度的复杂薄壁件内外表面点云数据,为后续的CAD建模提供基础。     

基于CT图像重建腰椎活动节段三维有限元模型及其应用

腰椎节段三维有限元模型既要求在视觉上与实际相似 ,又要求易于网格划分和数学求解。本文基于中国人体腰椎L4 - 5活动节段的系列二维截面CT图像 ,采用轮廓线的描述方式提取平面数据 ,通过将平面数据坐标进行三维化处理 ,获得腰椎节段的三维体数据 ,最后直接利用ANSYS有限元分析软件 ,采用自底向上的建模方式重建腰椎节段的三维有限元模型 ,并利用该模型研究了L4 - 5腰椎间盘的生物力学行为。结果表明 ,该模型具有很好的视觉效果和数学求解能力。     

不锈钢粉末注射坯缺陷在工业断层扫描图像中的特征分析

利用工业断层扫描技术对不锈钢粉末注射坯进行了检测。针对注射样中孔隙较小,不易在CT扫描图片上发现的问题,在Matlab软件上开发出一套图像计算方法,完成了对注射样扫描图像中孔隙缺陷的提取。然后对缺陷出现切面的二维重建图像进行了图像处理,得到了孔隙在这些切面上的二维形貌。最后利用三维重建技术将缺陷区域的二值图像进行重建,得到了孔隙的空间形貌。结合孔隙在注射坯中出现的位置及其三维形貌对注射工艺提出了相应的修正。     

基于ICT断层扫描图像的逆向工程功能设计

本文开发设计了一套基于ICT断层切描图像的逆向工程软件平台。运用图像图形处理中的边缘检测、阈值分割、边界细化、矢量化及格式转换等算法与技术,实现工件由三维断层ICT数据到快速成型机STL接口及CAD图纸再编辑的功能。所设计的软件功能齐全,层次清晰,具有较高实用性。     

胸部常见放疗反应有哪些？

近年来,肺癌、纵膈肿瘤等胸部肿瘤的发病率逐年提高。随着放射物理、计算机技术和医学影像技术的发展,肿瘤的放射治疗技术迅猛发展,许多先进的放疗技术逐渐应用于临床,如三维适形放疗技术、三维适形调强放疗技术、影像引导的放疗技术和断层放疗技术等。这些前沿技术的临床应用使得放疗模式发生了质的转变,即由过去的二维照射模式转换成了三维立体定向的精确放疗模式。     

一种三维规则数据场中复杂组织快速分割方法

提出了一种三维规则数据场中复杂组织快速分割方法：首先熵阈值二值化三维医学图像，然后用三维形态学腐蚀操作，断开复杂组织与其它组织间的弱连接，并对复杂组织进行连通标记；接着提取出腐蚀后的复杂组织模板，并对此模板进行三维形态学扩张操作，恢复先前三维形态学腐蚀操作消除的部分；最后采用改进的快速三维种子填充算法精确地分割出复杂组织。复杂组织分割实验结果表明了该方法的有效性。     

光学投影断层成像在离体样本中的应用研究

目的利用光学投影断层成像系统对小鼠离体心脏和肝脏进行三维成像,进而通过考察三维成像效果研究光学投影断层成像的应用价值。方法对小鼠进行心脏灌注,分别取出心脏、肝脏组织,并对组织进行琼脂包埋、脱水和透明化处理,得到心脏和肝脏样本。利用光学投影断层成像系统采集心脏和肝脏的透射数据,并通过滤波反投影的方法得到心脏和肝脏的三维图像。分析得到的三维图像,研究光学投影断层成像的应用价值。结果通过上述实验方法得到心脏和肝脏样本,利用光学投影断层成像系统得到具有较高清晰度和空间分辨率的样本三维结构图像。结论利用光学投影断层成像技术可获得具有高清晰度的生物组织器官的三维结构图像,在组织分辨率水平上获得小动物器官的数据信息,为尺度在1~10 mm的生物样本提供了有力的研究手段,必将大大推动生物科学基础研究的发展。     

A comparison of two radiological path length algorithms

Most radiation therapy dose calculation methods require the determination of the effective path length of the primary radiation from the radiation source to the point at which the dose is calculated. This usually involves representing the patient anatomy as a set of polygons (contours) as approximations to plane curves. Several algorithms are known for determining the length of a segment or segments on a ray through a planar contour, that are interior to the contour. We have implemented two of these algorithms in a test program to benchmark their relative efficiency. One algorithm uses a linear search over all the contour segments, and the other method represents the contour as a binary tree of "strips," of successively increasing resolution. In general, the tree search should give times proportional to log(n) where n is the number of contour segments, and the linear search time should be proportional to n. Thus, one might expect the tree search to run faster once the number of segments reaches some sufficiently large value. We found that this value is a number of contour points far in excess of that typical for contours representing radiation therapy patient anatomy. Therefore, for this application the linear search method is more efficient.     

Three-dimensional movement of a liver tumor detected by high-speed magnetic resonance imaging

Objective: Three-dimensional (3D) movement of a spherical liver tumor during respiration was investigated with magnetic resonance imaging (MRI) using a high-speed sequence.

Methods: A marker was placed on the surface of the patient as a reference of distance. Repetition time (TR) was 7.7 ms, echo time (TE) was 4.2 ms, flip angle was 20°, section thickness was 8 mm, and a 256×128 matrix was used. The acquisition time was 1.0 s followed by an interval of 0.5 s. The 20 tumor contours extracted during 30 s were superimposed on sagittal and coronal MR images.

Results: The maximum value of tumor edge location was 3.9 cm in the cranio–caudal direction, 2.3 cm in the ventro–dorsal direction, and 3.1 cm in the lateral direction. The mean length of tumor displacement observed was 2.1 cm in the cranio–caudal direction, 0.8 cm in the ventro–dorsal and 0.9 cm in the left–right direction, respectively. The locus of the center of the tumor contour in the sagittal cross section was inclined at 23° and in the coronal cross section was inclined at 18° to the cranio–caudal axis of body.

Conclusion: In conclusion, 3D movement of a spherical liver tumor was detected using rapid MRI sequential examinations. Magnetic resonance imaging has a potential to improve the accuracy of the planning target volume of a liver tumor.     

CT and PET lung image registration and fusion in radiotherapy treatment planning using the chamfer-matching method

PURPOSE: We present a validation study of CT and PET lung image registration and fusion based on the chamfer-matching method. METHODS AND MATERIALS: The contours of the lung surfaces from CT and PET transmission images were automatically segmented by the thresholding technique. The chamfer-matching technique was then used to register the extracted lung surfaces. Arithmetic means of distance between the two data sets of the pleural surfaces were used as the cost function. Matching was then achieved by iteratively minimizing the cost function through three-dimensional (3D) translation and rotation with an optimization method. RESULTS: Both anatomic thoracic phantom images and clinical patient images were used to evaluate the performance of our registration system. Quantitative analysis from five patients indicates that the registration error in translation was 2-3 mm in the transverse plane, 3-4 mm in the longitudinal direction, and about 1.5 degree in rotation. Typical computing time for chamfer matching is about 1 min. The total time required to register a set of CT and PET lung images, including contour extraction, was generally less than 30 min. CONCLUSION: We have implemented and validated the chamfer-matching method for CT and PET lung image registration and fusion. Our preliminary results show that the chamfer-matching method for CT and PET images in the lung area is feasible. The described registration system has been used to facilitate target definition and treatment planning in radiotherapy.     

3-D Visualization of Medical Images with Arbitrary Sections

1　ＩＮＴＲＯＤＵＣＴＩＯＮ　　Ｔｈｒｅｅｄｉｍｅｎｓｉｏｎａｌ(3Ｄ)ｖｉｓｕａｌｉｚａｔｉｏｎｏｆｍｅｄｉｃａｌｉｍａｇｅｓ,ｗｈｉｃｈｉｓｆｏｃｕｓｅｄｏｎｔｈｅｄｉｓｐｌａｙａｎｄｉｎｔｅｒｐｒｅｔａｔｉｏｎｏｆａｎａｔｏｍｉｃａｌｓｔｒｕｃｔｕｒｅｓ,ｐｌａｙｓａｎｉｍｐｏｒｔａｎｔｒｏｌｅｉｎｔｈｅｍｅｄｉｃａｌｄｅｖｅｌｏｐｍｅｎｔ.Ｉｔｉｓｎｏｔｏｎｌｙａｐｐｌｉｅｄｔｏａｉｄｄｉａｇｎｏｓｉｓ,ｂｕｔａｌｓｏｔｈｅｋｅｙｔｅｃｈｎｏｌｏｇｙｏｆｍａｎｙｍｏｄｅｒｎｍｅｄｉｃａｌｐｒｏｊ…     

Distance deviation measure of contouring variability

Background

Several methods that are currently used for contouring analysis have problems providing reliable and/or meaningful results. In this paper a solution to these problems is proposed in a form of a novel measure, which was developed based on requirements defined for contouring studies.

Materials and methods

The proposed distance deviation measure can be understood as an extension of the closest point measures in such a way that it does not measure only distances between points on contours but rather analyse deviation of distances to both/all contours from each image point/voxel. The obtained result is information rich, reliable and provided in a form of an image, enabling detailed topographic analysis. In addition to image representation, results can be further processed into angular representation for compact topographic analysis or into overall scalar estimates for quick assessment of contour disagreement.

Results

Distance deviation method is demonstrated on a multi observer contouring example with complex contour shapes, i.e., with pronounced extremes and void interior. The results are presented using the three proposed methods.

Conclusions

The proposed method can detect and measure contour variation irrespective of contour complexity and number of contour segments, while the obtained results are easy to interpret. It can be used in various situations, regarding the presence of reference contour or multiple test contours.     

Automatic contouring of brachial plexus using a multi-atlas approach for lung cancer radiation therapy

PurposeTo demonstrate a multi-atlas segmentation approach to facilitating accurate and consistent delineation of low-contrast brachial plexuses on computed tomographic images for lung cancer radiation therapy.Methods and MaterialsWe retrospectively identified 90 lung cancer patients with treatment volumes near the brachial plexus. Ten representative patients were selected to form an atlas group, and their brachial plexuses were delineated manually. We used deformable image registration to map each atlas brachial plexus to the remaining 80 patients. In each patient, a composite contour was created from 10 individual segmentations using the simultaneous truth and performance level estimation algorithm. This auto-delineated contour was reviewed and modified appropriately for each patient. We also performed 10 leave-one-out tests using the 10 atlases to validate the segmentation accuracy and demonstrate the contouring consistency using multi-atlas segmentation.ResultsThe multi-atlas segmentation took less than 2 minutes to complete. Contour modification took 5 minutes compared with 20 minutes for manual contouring from scratch. The multi-atlas segmentation from the 10 leave-one-out tests had a mean 3-dimensional (3D) volume overlap of 59.2% ± 8.2% and a mean 3D surface distance of 2.4 mm ± 0.5 mm. The distances between the individual and average contours in the 10 leave-one-out tests demonstrated much better contouring consistency for modified contours than for manual contours. The auto-segmented contours did not require substantial modification, demonstrated by the good agreement between the modified and auto-segmented contours in the 80 patients. Dose volume histograms of auto-segmented and modified contours were also in good agreement, showing that editing auto-segmented contours is clinically acceptable in view of the dosimetric impact.ConclusionsMulti-atlas segmentation greatly reduced contouring time and improved contouring consistency. Editing auto-segmented contours to delineate the brachial plexus proved to be a better clinical practice than manually contouring from scratch.     

Feasibility of geometrical verification of patient set-up using body contours and computed tomography data.

BACKGROUND AND PURPOSE: Body contours can potentially be used for patient set-up verification in external-beam radiotherapy and might enable more accurate set-up of patients prior to irradiation. The aim of this study is to test the feasibility of patient set-up verification using a body contour scanner. MATERIAL AND METHODS: Body contour scans of 33 lung cancer and 21 head-and-neck cancer patients were acquired on a simulator. We assume that this dataset is representative for the patient set-up on an accelerator. Shortly before acquisition of the body contour scan, a pair of orthogonal simulator images was taken as a reference. Both the body contour scan and the simulator images were matched in 3D to the planning computed tomography scan. Movement of skin with respect to bone was quantified based on an analysis of variance method. RESULTS: Set-up errors determined with body-contours agreed reasonably well with those determined with simulator images. For the lung cancer patients, the average set-up errors (mm)+/-1 standard deviation (SD) for the left-right, cranio-caudal and anterior-posterior directions were 1.2+/-2.9, -0.8+/-5.0 and -2.3+/-3.1 using body contours, compared to -0.8+/-3.2, -1.0+/-4.1 and -1.2+/-2.4 using simulator images. For the head-and-neck cancer patients, the set-up errors were 0.5+/-1.8, 0.5+/-2.7 and -2.2+/-1.8 using body contours compared to -0.4+/-1.2, 0.1+/-2.1, -0.1+/-1.8 using simulator images. The SD of the set-up errors obtained from analysis of the body contours were not significantly different from those obtained from analysis of the simulator images. Movement of the skin with respect to bone (1 SD) was estimated at 2.3 mm for lung cancer patients and 1.7 mm for head-and-neck cancer patients. CONCLUSION: Measurement of patient set-up using a body-contouring device is possible. The accuracy, however, is limited by the movement of the skin with respect to the bone. In situations where the error in the patient set-up is relatively large, it is possible to reduce these errors using a computer-aided set-up technique based on contour information.     

External contour acquisition system for radiotherapy: an original solution]

A contour acquisition system has been designed in radiotherapy at the Sagamie Hospital complex (Chicoutimi, Québec) to measure the external contours of the patients who do not need a CT exam. This measuring system can produce transversal, sagittal or coronal patient contours in the treatment position. The absolute accuracy of the system is +/- 1 mm. The contours produced by this equipment can be transferred electronically or on paper to the planning system.     

Automated contour mapping with a regional deformable model

PURPOSE: To develop a regional narrow-band algorithm to auto-propagate the contour surface of a region of interest (ROI) from one phase to other phases of four-dimensional computed tomography (4D-CT). METHODS AND MATERIALS: The ROI contours were manually delineated on a selected phase of 4D-CT. A narrow band encompassing the ROI boundary was created on the image and used as a compact representation of the ROI surface. A BSpline deformable registration was performed to map the band to other phases. A Mattes mutual information was used as the metric function, and the limited memory Broyden-Fletcher-Goldfarb-Shanno algorithm was used to optimize the function. After registration the deformation field was extracted and used to transform the manual contours to other phases. Bidirectional contour mapping was introduced to evaluate the proposed technique. The new algorithm was tested on synthetic images and applied to 4D-CT images of 4 thoracic patients and a head-and-neck Cone-beam CT case. RESULTS: Application of the algorithm to synthetic images and Cone-beam CT images indicates that an accuracy of 1.0 mm is achievable and that 4D-CT images show a spatial accuracy better than 1.5 mm for ROI mappings between adjacent phases, and 3 mm in opposite-phase mapping. Compared with whole image-based calculations, the computation was an order of magnitude more efficient, in addition to the much-reduced computer memory consumption. CONCLUSIONS: A narrow-band model is an efficient way for contour mapping and should find widespread application in future 4D treatment planning.