Targeted 2D/3D registration using ray normalization and a hybrid optimizer

X-ray images are often used to guide minimally invasive procedures in interventional radiology. The use of a preoperatively obtained 3D volume can enhance the visualization needed for guiding catheters and other surgical devices. However, for intraoperative usefulness, the 3D dataset needs to be registered to the 2D x-ray images of the patient. We investigated the effect of targeting subvolumes of interest in the 3D datasets and registering the projections with C-arm x-ray images. We developed an intensity-based 2D/3D rigid-body registration using a Monte Carlo-based hybrid algorithm as the optimizer, using a single view for registration. Pattern intensity (PI) and mutual information (MI) were two metrics tested. We used normalization of the rays to address the problems due to truncation in 3D necessary for targeting. We tested the algorithm on a C-arm x-ray image of a pig's head and a 3D dataset reconstructed from multiple views of the C-arm. PI and MI were comparable in performance. For two subvolumes starting with a set of initial poses from +/-15 mm in x, from +/-3 mm (random), in y and z and +/-4 deg in the three angles, the robustness was 94% for PI and 91% for MI, with accuracy of 2.4 mm (PI) and 2.6 mm (MI), using the hybrid algorithm. The hybrid optimizer, when compared with a standard Powell's direction set method, increased the robustness from 59% (Powell) to 94% (hybrid). Another set of 50 random initial conditions from [+/-20] mm in x,y,z and [+/-10] deg in the three angles, yielded robustness of 84% (hybrid) versus 38% (Powell) using PI as metric, with accuracies 2.1 mm (hybrid) versus 2.0 mm (Powell).     

2D/3D registration of endoscopic ultrasound to CT volume data

This paper describes a computer-aided navigation system using image fusion to support endoscopic interventions such as the accurate collection of biopsy specimens. An endoscope provides the physician with real-time ultrasound (US) and a video image. An image slice that corresponds to the corresponding image from the US scan head is derived from a preoperative computed tomography (CT) or magnetic resonance image volume data set using oblique reformatting and displayed side by side with the US image. The position of the image acquired by the US scan head is determined by a miniaturized electromagnetic tracking system (EMTS) after calibrating the endoscope's scan head. The transformation between the patient coordinate system and the preoperative data set is calculated using a 2D/3D registration. This is achieved by calibrating an intraoperative interventional CT slice with an optical tracking system (OTS) using the same algorithm as for the US calibration. The slice is then used for 2D/3D registration with the coordinate system of the preoperative volume. The fiducial registration error (FRE) for the US calibration was 2.0 mm +/- 0.4 mm; the interventional CT FRE was 0.36 +/- 0.12 mm; and the 2D/3D registration target registration error (TRE) was 1.8 +/- 0.3 mm. The point-to-point registration between the OTS and the EMTS had an FRE of 0.9 +/- 0.4 mm. Finally, we found an overall TRE for the complete system to be 3.9 +/- 0.6 mm.     

利用2D/3D配准技术开展直立位颈椎椎间孔形态学测量的可行性研究

目的　探讨利用2D/3D配准技术开展直立位颈椎椎间孔形态学测量的可行性研究。 方法　采集7位健康成年志愿者颈椎的CT影像,对其进行三维重建获得颈椎三维模型,并采集同一志愿者保持同一姿势时的颈椎直立正侧位X线平片,在图像处理软件Mimics中建立虚拟X线场景,采用2D/3D图像配准还原摄平片时颈椎的位置,并利用Rapidform XOR3软件测量两种位置的椎间孔面积。 结果　利用2D/3D图像配准方法,可获得颈椎直立位的三维位置数据,共测得56个椎间孔卧位及直立位两位置椎间孔面积,卧位椎间孔面积为(50.9±14.2) mm2,直立位椎间孔面积为(83.6±23.5) mm2,差异具有统计学意义（t=-8.107,P<0.05）。 结论 利用2D/3D配准技术可获得颈椎直立位椎间孔的形态学参数,直立位与卧位椎间孔有显著差异,但其准确性需要进一步的研究验证。     

Automated 2D-3D registration of portal images and CT data using line-segment enhancement

In prostate radiotherapy, setup errors with respect to the patient's bony anatomy can be reduced by aligning 2D megavoltage (MV) portal images acquired during treatment to a reference 3D kilovoltage (kV) CT acquired for treatment planning purposes. The purpose of this study was to evaluate a fully automated 2D-3D registration algorithm to quantify setup errors in 3D through the alignment of line-enhanced portal images and digitally reconstructed radiographs computed from the CT. The line-enhanced images were obtained by correlating the images with a filter bank of short line segments, or "sticks" at different orientations. The proposed methods were validated on (1) accurately collected gold-standard data consisting of a 3D kV cone-beam CT scan of an anthropomorphic phantom of the pelvis and 2D MV portal images in the anterior-posterior (AP) view acquired at 15 different poses and (2) a conventional 3D kV CT scan and weekly 2D MV AP portal images of a patient over 8 weeks. The mean (and standard deviation) of the absolute registration error for rotations around the right-lateral (RL), inferior-superior (IS), and posterior-anterior (PA) axes were 0.212 degree (0.214 degree), 0.055 degree (0.033 degree) and 0.041 degree (0.039 degree), respectively. The corresponding registration errors for translations along the RL, IS, and PA axes were 0.161 (0.131) mm, 0.096 (0.033) mm, and 0.612 (0.485) mm. The mean (and standard deviation) of the total registration error was 0.778 (0.543) mm. Registration on the patient images was successful in all eight cases as determined visually. The results indicate that it is feasible to automatically enhance features in MV portal images of the pelvis for use within a completely automated 2D-3D registration framework for the accurate determination of patient setup errors. They also indicate that it is feasible to estimate all six transformation parameters from a 3D CT of the pelvis and a single portal image in the AP view.     

基于单幅X线图像和CT数据的2D/3D配准系统

目的建立基于统一计算架构(CUDA)下以单幅X线图像及CT扫描数据为数据源的2D/3D配准系统,并应用于膝关节在体运动及植入假体稳定性研究。方法首先应用张正友标定法对采集X线图像设备进行标定;其次基于CUDA构架利用光线跟踪算法生成数字影像重建图像,以相关性函数为相似性测度计算2D/3D配准参数;最后以三维激光扫描仪所获得的点云数据进行3D/3D配准,以验证2D/3D配准结果。结果以标本整体位置变换进行配准实验,6自由度平均误差中,位移小于1mm,旋转小于1°。结论此2D/3D配准系统达到了运动检测精度的要求,可以作为研究膝关节运动情况和假体在体稳定性研究的计算平台。     

Automated 2D-3D registration of a radiograph and a cone beam CT using line-segment enhancement

The objective of this study was to develop a fully automated two-dimensional (2D)-three-dimensional (3D) registration framework to quantify setup deviations in prostate radiation therapy from cone beam CT (CBCT) data and a single AP radiograph. A kilovoltage CBCT image and kilovoltage AP radiograph of an anthropomorphic phantom of the pelvis were acquired at 14 accurately known positions. The shifts in the phantom position were subsequently estimated by registering digitally reconstructed radiographs (DRRs) from the 3D CBCT scan to the AP radiographs through the correlation of enhanced linear image features mainly representing bony ridges. Linear features were enhanced by filtering the images with "sticks," short line segments which are varied in orientation to achieve the maximum projection value at every pixel in the image. The mean (and standard deviations) of the absolute errors in estimating translations along the three orthogonal axes in millimeters were 0.134 (0.096) AP(out-of-plane), 0.021 (0.023) ML and 0.020 (0.020) SI. The corresponding errors for rotations in degrees were 0.011 (0.009) AP, 0.029 (0.016) ML (out-of-plane), and 0.030 (0.028) SI (out-of-plane). Preliminary results with megavoltage patient data have also been reported. The results suggest that it may be possible to enhance anatomic features that are common to DRRs from a CBCT image and a single AP radiography of the pelvis for use in a completely automated and accurate 2D-3D registration framework for setup verification in prostate radiotherapy. This technique is theoretically applicable to other rigid bony structures such as the cranial vault or skull base and piecewise rigid structures such as the spine.     

基于2D/3D配准技术测量颈椎间孔形态学的精度研究

目的　探讨基于2D/3D配准技术测量离体颈椎标本椎间孔形态学的精度。 方法　获取16具颈椎离体标本的螺旋CT影像,通过三维重建获得相应标本的三维模型。然后改变标本的体位,再次行CT扫描并分别拍摄各具离体标本正侧位X线平片,采用2D/3D图像配准技术还原拍摄平片时颈椎的三维位置。利用Rapidform XOR3软件测量配准前与配准后C2/3至C6/7双侧的椎间孔面积、前后径和上下径。 结果　共测得椎间孔的面积、前后径和上下径各158个,经配对样本t检验,配准后椎间孔的面积、前后径、上下径与配准前相比较,结果无显著性差异(P>0.05);椎间孔面积的准确度为96.77%、精确度为（1.27±1.16）mm2,前后径的准确度为94.35%、精确度为（0.30±0.27）mm,上下径的准确度为96.14%、精确度为（0.32±0.28）mm。 结论　应用2D/3D配准技术测量颈椎间孔的形态学参数具有较高的准确度和精确度。     

The influence of distance on registration in ISO-C-3D navigation: a source of error in ISO-C-3D navigation.

Computer-assisted treatments have become increasingly common. Consequently, there is an increased desire for navigation methods with simplified workflow. Anatomic-based pair-point registration is often mentioned as a source of error. Alternatively, the use of preoperatively implanted markers for registration remains complex. The self-acting registration of Iso-C-3D at the moment of data acquisition can reduce essential errors. The aim of this study was to evaluate the effect of reference placement on accuracy, and to determine the maximum acceptable distance between the reference and a given isocentre. This study demonstrates the interdependence of the reference distance on the region of interest (ROI). The mean error of registration amounts to 0.04 mm (0.04-0.05 mm) up to a distance of 200 mm and beyond 0.25 mm (0.24-0.26) for distances beyond 200 mm. The accuracy was significantly lower (p<0.0001) with a distance more than 200 mm. For optimal accuracy when utilizing navigation for pelvic and long bone surgery, the reference base should not been placed at a distance more than 200 mm from the isocentre of interest.     

Effects of x-ray and CT image enhancements on the robustness and accuracy of a rigid 3D/2D image registration

A rigid body three-dimensional/two-dimensional (3D/2D) registration method has been implemented using mutual information, gradient ascent, and 3D texturemap-based digitally reconstructed radiographs. Nine combinations of commonly used x-ray and computed tomography (CT) image enhancement methods, including window leveling, histogram equalization, and adaptive histogram equalization, were examined to assess their effects on accuracy and robustness of the registration method. From a set of experiments using an anthropomorphic chest phantom, we were able to draw several conclusions. First, the CT and x-ray preprocessing combination with the widest attraction range was the one that linearly stretched the histograms onto the entire display range on both CT and x-ray images. The average attraction ranges of this combination were 71.3 mm and 61.3 deg in the translation and rotation dimensions, respectively, and the average errors were 0.12 deg and 0.47 mm. Second, the combination of the CT image with tissue and bone information and the x-ray images with adaptive histogram equalization also showed subvoxel accuracy, especially the best in the translation dimensions. However, its attraction ranges were the smallest among the examined combinations (on average 36 mm and 19 deg). Last the bone-only information on the CT image did not show convergency property to the correct registration.     

Automatic localization of vertebral levels in x-ray fluoroscopy using 3D-2D registration: a tool to reduce wrong-site surgery

Surgical targeting of the incorrect vertebral level (wrong-level surgery) is among the more common wrong-site surgical errors, attributed primarily to the lack of uniquely identifiable radiographic landmarks in the mid-thoracic spine. The conventional localization method involves manual counting of vertebral bodies under fluoroscopy, is prone to human error and carries additional time and dose. We propose an image registration and visualization system (referred to as LevelCheck), for decision support in spine surgery by automatically labeling vertebral levels in fluoroscopy using a GPU-accelerated, intensity-based 3D-2D (namely CT-to-fluoroscopy) registration. A gradient information (GI) similarity metric and a CMA-ES optimizer were chosen due to their robustness and inherent suitability for parallelization. Simulation studies involved ten patient CT datasets from which 50?000 simulated fluoroscopic images were generated from C-arm poses selected to approximate the C-arm operator and positioning variability. Physical experiments used an anthropomorphic chest phantom imaged under real fluoroscopy. The registration accuracy was evaluated as the mean projection distance (mPD) between the estimated and true center of vertebral levels. Trials were defined as successful if the estimated position was within the projection of the vertebral body (namely mPD <5?mm). Simulation studies showed a success rate of 99.998% (1 failure in 50?000 trials) and computation time of 4.7?s on a midrange GPU. Analysis of failure modes identified cases of false local optima in the search space arising from longitudinal periodicity in vertebral structures. Physical experiments demonstrated the robustness of the algorithm against quantum noise and x-ray scatter. The ability to automatically localize target anatomy in fluoroscopy in near-real-time could be valuable in reducing the occurrence of wrong-site surgery while helping to reduce radiation exposure. The method is applicable beyond the specific case of vertebral labeling, since any structure defined in pre-operative (or intra-operative) CT or cone-beam CT can be automatically registered to the fluoroscopic scene.     

2D-3D registration of coronary angiograms for cardiac procedure planning and guidance

We present a completely automated 2D-3D registration technique that accurately maps a patient-specific heart model, created from preoperative images, to the patient's orientation in the operating room. This mapping is based on the registration of preoperatively acquired 3D vascular data with intraoperatively acquired angiograms. Registration using both single and dual-plane angiograms is explored using simulated but realistic datasets that were created from clinical images. Heart deformations and cardiac phase mismatches are taken into account in our validation using a digital 4D human heart model. In an ideal situation where the pre- and intraoperative images were acquired at identical time points within the cardiac cycle, the single-plane and the dual-plane registrations resulted in 3D root-mean-square (rms) errors of 1.60 +/- 0.21 and 0.53 +/- 0.08 mm, respectively. When a 10% timing offset was added between the pre- and the intraoperative acquisitions, the single-plane registration approach resulted in inaccurate registrations in the out-of-plane axis, whereas the dual-plane registration exhibited a 98% success rate with a 3D rms error of 1.33 +/- 0.28 mm. When all potential sources of error were included, namely, the anatomical background, timing offset, and typical errors in the vascular tree reconstruction, the dual-plane registration performed at 94% with an accuracy of 2.19 +/- 0.77 mm.     

Mammogram synthesis using a 3D simulation. II. Evaluation of synthetic mammogram texture

We have evaluated a method for synthesizing mammograms by comparing the texture of clinical and synthetic mammograms. The synthesis algorithm is based upon simulations of breast tissue and the mammographic imaging process. Mammogram texture was synthesized by projections of simulated adipose tissue compartments. It was hypothesized that the synthetic and clinical texture have similar properties, assuming that the mammogram texture reflects the 3D tissue distribution. The size of the projected compartments was computed by mathematical morphology. The texture energy and fractal dimension were also computed and analyzed in terms of the distribution of texture features within four different tissue regions in clinical and synthetic mammograms. Comparison of the cumulative distributions of the mean features computed from 95 mammograms showed that the synthetic images simulate the mean features of the texture of clinical mammograms. Correlation of clinical and synthetic texture feature histograms, averaged over all images, showed that the synthetic images can simulate the range of features seen over a large group of mammograms. The best agreement with clinical texture was achieved for simulated compartments with radii of 4-13.3 mm in predominantly adipose tissue regions, and radii of 2.7-5.33 and 1.3-2.7 mm in retroareolar and dense fibroglandular tissue regions, respectively.     

Classification of a large microarray data set: algorithm comparison and analysis of drug signatures

A large gene expression database has been produced that characterizes the gene expression and physiological effects of hundreds of approved and withdrawn drugs, toxicants, and biochemical standards in various organs of live rats. In order to derive useful biological knowledge from this large database, a variety of supervised classification algorithms were compared using a 597-microarray subset of the data. Our studies show that several types of linear classifiers based on Support Vector Machines (SVMs) and Logistic Regression can be used to derive readily interpretable drug signatures with high classification performance. Both methods can be tuned to produce classifiers of drug treatments in the form of short, weighted gene lists which upon analysis reveal that some of the signature genes have a positive contribution (act as "rewards" for the class-of-interest) while others have a negative contribution (act as "penalties") to the classification decision. The combination of reward and penalty genes enhances performance by keeping the number of false positive treatments low. The results of these algorithms are combined with feature selection techniques that further reduce the length of the drug signatures, an important step towards the development of useful diagnostic biomarkers and low-cost assays. Multiple signatures with no genes in common can be generated for the same classification end-point. Comparison of these gene lists identifies biological processes characteristic of a given class.     

Comparison of basis functions for 3D PET reconstruction using a Monte Carlo system matrix

In emission tomography, iterative statistical methods are accepted as the reconstruction algorithms that achieve the best image quality. The accuracy of these methods relies partly on the quality of the system response matrix (SRM) that characterizes the scanner. The more physical phenomena included in the SRM, the higher the SRM quality, and therefore higher image quality is obtained from the reconstruction process. High-resolution small animal scanners contain as many as 103-10? small crystal pairs, while the field of view (FOV) is divided into hundreds of thousands of small voxels. These two characteristics have a significant impact on the number of elements to be calculated in the SRM. Monte Carlo (MC) methods have gained popularity as a way of calculating the SRM, due to the increased accuracy achievable, at the cost of introducing some statistical noise and long simulation times. In the work presented here the SRM is calculated using MC methods exploiting the cylindrical symmetries of the scanner, significantly reducing the simulation time necessary to calculate a high statistical quality SRM and the storage space necessary. The use of cylindrical symmetries makes polar voxels a convenient basis function. Alternatively, spherically symmetric basis functions result in improved noise properties compared to cubic and polar basis functions. The quality of reconstructed images using polar voxels, spherically symmetric basis functions on a polar grid, cubic voxels and post-reconstruction filtered polar and cubic voxels is compared from a noise and spatial resolution perspective. This study demonstrates that polar voxels perform as well as cubic voxels, reducing the simulation time necessary to calculate the SRM and the disk space necessary to store it. Results showed that spherically symmetric functions outperform polar and cubic basis functions in terms of noise properties, at the cost of slightly degraded spatial resolution, larger SRM file size and longer reconstruction times. However, we demonstrate that post-reconstruction smoothing, usually applied in emission imaging to reduce the level of noise, can produce a spatial resolution degradation of ~50%, while spherically symmetric basis functions produce a degradation of only ~6%, compared to polar and cubic voxels, at the same noise level. Therefore, the image quality trade-off obtained with blobs is higher than that obtained with cubic or polar voxels.     

Determination of the detective quantum efficiency of a digital x-ray detector: comparison of three evaluations using a common image data set

The detective quantum efficiency (DQE) of an x-ray digital imaging detector was determined independently by the three participants of this study, using the same data set consisting of edge and flat field images. The aim was to assess the possible variation in DQE originating from established, but slightly different, data processing methods used by different groups. For the case evaluated in this study differences in DQE of up to +/-15% compared to the mean were found. The differences could be traced back mainly to differences in the modulation transfer function (MTF) and noise power spectrum (NPS) determination. Of special importance is the inclusion of a possible low-frequency drop in MTF and the proper handling of signal offsets for the determination of the NPS. When accounting for these factors the deviation between the evaluations reduced to approximately +/-5%. It is expected that the recently published standard on DQE determination will further reduce variations in the data evaluation and thus in the results of DQE measurements.     

Development and validation of a CT-3D rotational angiography registration method for AVM radiosurgery

In this paper a novel technique is proposed and validated for radiosurgery treatment planning of arteriovenous malformations (AVMs). The technique was developed for frameless radiosurgery by means of the CyberKnife, a nonisocentric, linac-based system which allows highly conformed isodose surfaces to be obtained, while also being valid for other treatment strategies. The technique is based on registration between computed tomography (CT) and three-dimensional rotational angiography (3DRA). Tests were initially performed on the effectiveness of the correction method for distortion offered by the angiographic system. These results determined the registration technique that was ultimately chosen. For CT-3DRA registration, a twelve-parameter affine transformation was selected, based on a mutual information maximization algorithm. The robustness of the algorithm was tested by attempting to register data sets increasingly distant from each other, both in translation and rotation. Registration accuracy was estimated by means of the "full circle consistency test." A registration quality index (expressed in millimeters) based on these results was also defined. A hybrid subtraction between CT and 3DRA is proposed in order to improve 3D reconstruction. Preprocessing improved the ability of the algorithm to find an acceptable solution to the registration process. The robustness tests showed that data sets must be manually prealigned within approximately 15 mm and 20 degrees with respect to all three directions simultaneously. Results of the consistency test showed agreement between the quality index and registration accuracy stated by visual inspection in 20 good and 10 artificially worsened registration processes. The quality index showed values smaller than the maximum voxel size (mean 0.8 mm compared to 2 mm) for all successful registrations, while it resulted in much greater values (mean 20 mm) for unsuccessful registrations. Once registered, the two data sets can be used for CyberKnife treatment planning. Target delineation is performed on 3DRA while dose calculation and DRR generation are performed on CT. In conclusion, a method was developed for using 3DRA images for AVM frameless radiosurgery treatment planning. The method proved to be feasible, robust, and accurate for clinical use. 3DRA can be performed at different times or locations compared to standard, frame based stereotactic angiography. Unlike two-dimensional angiography, 3DRA allows examination of the shape of the AVM and of the surrounding target from any arbitrary point of view during treatment planning. The method can be applied to any case of intermodality registration, is operator-independent, and allows estimation of registration quality. Further research is desirable to improve time resolution in order to distinguish between feeding and draining vessels.     

Immunoregulatory functions of paf-acether. II. Decrease of CD2 and CD3 antigen expression

Paf-acether (platelet-activating factor) is a phospholipid initially described as a potent platelet-aggregating compound. It is produced by numerous cell types and is now considered as an important mediator of cell-cell interactions. The effect of paf-acether on the expression of CD2 and CD3, two human T cell surface glycoproteins, was investigated by indirect immunofluorescence and flow cytometry. Paf-acether partially down-regulated, in a time- and dose-dependent manner, CD2 and CD3 but not HLA class I antigen expression on peripheral human T cells and Jurkat cells. Lysophosphatidylcholine, a phospholipid closely related to paf-acether, had no detectable modulatory effect on CD2 and CD3 expression. In addition to CD2/CD3 modulation, paf-acether markedly inhibited T cell proliferative response not only to phytohemagglutinin or concanavalin A but also to anti-CD3 or a stimulatory combination of anti-CD2 monoclonal antibodies. These data demonstrate for the first time that lipid mediators such as paf-acether might be involved in the regulation of the expression of cell surface glycoproteins that are essential in the execution of T cell function.