Analysis of the frequency components of CR portal images and application to multi-objective frequency processing

The image contrast of portal images obtained using computed radiography (CR) is low for radiation therapy, and their noise level is high. In order to improve the image quality of CR portal images, we investigated the relationship between the frequency components of the bone edge and noise area. We used a personal computer, conventional flatbed scanner, and public-domain software (NIH Image). To analyze the frequency components of bone edges, we generated difference images from the original and unsharp mask images of various sizes. In addition, we analyzed the frequency components of noise, after changing the cut-off frequency on the power spectrum to process a low- and high-pass filter. We found that the frequency components of bone edges and noise areas ranged from 0.18 to 0.25 cycle/mm and 1.30 cycles/mm, respectively. We conclude that this process can remove noise from homogeneous areas and preserve bone edges using multi-objective frequency processing.     

3种图像模式在影像引导放射治疗中的差异性研究

目的 检测并分析电子射野影像系统( EPID)、kV级平面摄像(kV planar)、kV级锥形束CT( CBCT)3种图像模式在影像引导放疗(IGRT)中摆位误差的精度.方法 通过在计划系统的模体CT图像上模拟25组摆位偏差,生成25组DRR参考图像和25组三维CT参考图像,分别与模体在原点位置获取的EPID,kV planar和CBCT 3种图像模式进行图像配准和摆位误差的测量,检测并比较3种图像模式在摆位误差测量中的精度.结果 共读取675组残留误差,3种图像及其相应的配准方法均具有较高的精确度,残留误差的平均值在x、y、z方向上均＜1 mm,CBCT的灰度值配准精度＜0.1 mm,kV planar图像在手动配准时其测量摆位误差的精度好于EPID(＜0.65 mm),在自动配准方法时与EPID具有同等的精度.结论 医院IGRT系统的3种图像模式及其相应的配准方法均可以满足临床应用,CBCT为首选,结合图像质量、成像剂量和配准精度等因素,kV planar优于EPID.     

Quantification of the accuracy limits of image registration using peak signal-to-noise ratio

A new method was developed for quantifying the accuracy limits of image registration devices and the distortion of anatomical structures in verification images without image registration. A correlation was found between peak signal-to-noise ratio (PSNR) and the amount of parallel movement (1–10 mm at 1-mm intervals) of a rectangular parallelepiped phantom [correlation coefficient (CC) ?0.91, contribution ratio (CR) 0.83]. Rotating the phantom from 1° to 10° at 1° intervals produced a similar correlation with PSNR (CC ?0.91, CR 0.83). To allow for manual registration, the grid pattern of the Mylar top plate was extracted from 455 pelvic portal images of 21 patients using a band-pass filtering technique. This revealed a different correlation between the original data (CC ?0.62, CR 0.38) and averaged data (CC ?0.96, CR 0.92), but this is considered to have been caused by structural distortion and manual matching errors. Thus, PSNR can be used to evaluate the accuracy limits of image registration and provide a judgment index that can be used in re-planning or re-setup in adaptive radiotherapy.     

Development of a one-stop beam verification system using electronic portal imaging devices for routine quality assurance

In this study, a computer-based system for routine quality assurance (QA) of a linear accelerator (linac) was developed by using the dosimetric properties of an amorphous silicon electronic portal imaging device (EPID). An acrylic template phantom was designed such that it could be placed on the EPID and be aligned with the light field of the collimator. After irradiation, portal images obtained from the EPID were transferred in DICOM format to a computer and analyzed using a program we developed. The symmetry, flatness, field size, and congruence of the light and radiation fields of the photon beams from the linac were verified simultaneously. To validate the QA system, the ion chamber and film (X-Omat V2; Kodak, New York, NY) measurements were compared with the EPID measurements obtained in this study. The EPID measurements agreed with the film measurements. Parameters for beams with energies of 6 MV and 15 MV were obtained daily for 1 month using this system. It was found that our QA tool using EPID could substitute for the film test, which is a time-consuming method for routine QA assessment.     

The point spread function revisited: image restoration using 2-D deconvolution

Purpose : The quality of an image can be significantly improved by digital deconvolution with the (two-dimensional) point spread function (PSF) of the imaging system. We investigated the significance of this improvement for a projection radiograph of vertebral bone, using commercially available software.Methods : A magnified image of the PSF of a GE Advantx RFX system was obtained directly from a pinhole radiograph of the X-ray source and digitized. Images of vertebral bone obtained using similar technique factors were deconvolved with the PSF images, with due regard for magnification effects and using Wiener filtering to avoid amplifying the effects of noise.Results : The spatial resolution of these restored images was significantly better than the original images, and they were less noisy. A significant improvement in image quality could also be obtained by high-pass filtering using a Butterworth filter and a cut-off frequency matching that of the PSF.Conclusions : The quality of a degraded image can be significantly restored by deconvolution with the PSF as long as magnified images are obtained and digitized, and Wiener filtering is used. Alternatively, high-pass filtering can deliver some improvement and is even simpler to implement.     

基于电子射野影像系统建立医用直线加速器晨检工具的临床测试

目的 利用电子射野影像系统（EPID）建立加速器快速晨检工具,并进行评估。方法 用Synergy加速器EPID测得10 cm×10 cm开野及楔形野图像,由Matlab提取并分析图像参数,实现快速晨检。对EPID剂量-机械重复性、灰度值与MU值线性、输出量及射野大小测量准确性测试,用电离室、EPID分别测量Synergy输出量随MU变化情况,用DailyQA3及研究中所开发的工具对Synergy进行2个月的监测。结果 EPID剂量重复性测试稳定,射野大小和中心测试精度分别为0.5和1.0 mm,平坦度和对称性测试精度均为0.17%;机械精度测试结果与剂量重复性测试一致。EPID对加速器输出剂量响应线性相关（R²>0.999）。EPID对输出剂量和射野大小的探测灵敏度较高。EPID与DailyQA3所有临床测试结果均在允许限值内,且两种结果一致。结论 EPID剂量-机械稳定性及响应线性均良好,输出量及射野大小监测结果均准确,研究中建立的晨检工具准确可靠。     

Robust rigid registration of retinal angiograms through optimization

Retinal fundus photographs are employed as standard diagnostic tools in ophthalmology. Serial photographs of the flow of fluorescein and indocyanine green (ICG) dye are used to determine the areas of the retinal lesions. For objective measurements of features, the registration of the images is a necessity. In this paper, we employ optimization techniques for registration with the help of 2-parameter translational motion model of retinal angiograms, based on non-linear pre-processing (Wiener filtering and morphological gradient) and computation of the similarity criteria for the alignment of the two gradient images for any given rigid transformation. The optimization methods are effectively employed to minimize the similarity criterion.

The presence of noise, the variations in the background and the temporal variation of the fluorescence level pose serious problems in obtaining a robust registration of the retinal images. Moreover, local search strategies are not robust in the case of ICG angiograms, even if one uses a multiresolution approach.

The present work makes a systematic comparison of different optimization techniques, namely the minimization method derived from the optical flow formulation, the Nelder-Mead local search and the HCIAC ant colony metaheuristic, each optimizing a similarity criterion for the gradient images. The impact of the resolution and median filtering of gradient image is studied and the robustness of the approaches is tested through experimental studies, performed on macular fluorescein and ICG angiographies.

Our proposed optimization techniques have shown interesting results especially for high resolution difficult registration problems. Moreover, this approach seems promising for affine (6-parameter motion model) or elastical registrations.     

基于VoxelMorph无监督缺失图像配准的无标记射束方向观肿瘤跟踪算法

目的 基于机器学习提出可应用于低图像质量、多叶准直器(MLC)遮挡和非刚性变形兆伏级(MV)图像的无标记射束方向观(BEV)肿瘤放疗跟踪算法。方法 采用窗口模板匹配法和Voxelmorph端到端无监督网络,处理MV图像中的配准问题。使用动态胸部模体,验证肿瘤跟踪算法的准确性。将模体质量保证(QA)计划在加速器上手动设置治疗偏移后执行,收集治疗过程中的682幅电子射野影像系统(EPID)图像作为固定图像;同时采集计划系统中对应射野角度的数字影像重建(DRR)图作为浮动图像,进行靶区跟踪研究。收集21例肺部肿瘤放疗的533对EPID和DRR图像进行肿瘤跟踪研究,提供治疗过程中肿瘤位置变化定量结果。图像相似度用于算法的第三方验证。结果 算法可应对不同程度(10%~80%)的图像缺失,且对数据缺失图像的非刚性配准表现较好。模体验证中86.8%的跟踪误差<3 mm,<2 mm的比例约80%作用。配准后标准化互信息(NMI)由1.18±0.02提高到1.20±0.02(t=-6.78,P=0.001)。临床病例肿瘤运动以平移为主,平均位移3.78 mm,最大位移可达7.46 mm。配准结果显示存在非刚性形变,配准后NMI由1.21±0.03增至到1.22±0.03(t=-2.91,P=0.001)。结论 肿瘤跟踪算法跟踪精度可靠且鲁棒性好,可用于无创、实时、无额外设备和辐射剂量的肿瘤跟踪。     

The performance of a fluoroscopic electronic portal imaging device modified for portability

Advances in external beam therapy technology have made routine, efficient conformal therapy a reality. With it comes the increasing need for online treatment verification, which is only achievable at present through the use of electronic portal imaging devices (EPIDs). For a large radiotherapy centre, the provision of one EPID per treatment machine proves extremely expensive. This paper details modifications to the design of a commercial fluoroscopic EPID (the SRI-100) to produce a portable system, capable of providing quick, high quality imaging on more than one treatment machine. We describe the necessary hardware and software changes made to the system, as well as the variety of mechanical and quality control checks performed for testing the stability and quality of the imaging. The modified system has been found to be both electronically and mechanically robust, with associated image quality, scaling, distortion and movement similar to other EPIDs in the department. Although the modification was designed specifically to allow for the acquisition of images from multiple treatment machines, it may also enable the operation of the EPID for other uses such as total body irradiation (TBI) treatment verification and a further range of quality control procedures on the linear accelerator itself.     

改进式的二维图像配准算法在头颈部肿瘤中的应用

目的 研究摆位误差导致的二维图像投影变化及其对图像配准的影响,提出一种改进的互信息配准算法。方法 借助仿真头部体模,分别模拟旋转误差和平移误差,通过互信息的变化来反映投影形变。以3mm平移误差和3°旋转误差为界,模拟10例较小摆位误差和10例较大摆位误差,拍摄正侧位射野图像,分别使用目前加速器自带的传统互信息配准方法和本研究改进的互信息配准方法获取摆位误差,并与实际摆位误差相比较,以判断本研究提出的改进配准方法的优劣。结果 对于摆位误差较小的实例,加速器自带的传统互信息配准方法的x、y、z轴平均平移误差分别为0.3、0.4和0.3 mm,x、z轴平均旋转误差均为0.4°,平均耗时28.7 s。本研究改进的互信息配准方法的平均误差为0.4、0.3和0.3 mm,x、z轴平均旋转误差分别为0.5°和0.4°,平均耗时31.1 s。对于摆位误差较大的实例,加速器自带的传统互信息配准方法的平均误差分别为0.9、0.7和0.8 mm,x、z轴平均旋转误差为0.9°和0.8°,平均耗时29.9 s,本研究的改进互信息方法平均误差分别为0.5、0.4和0.5 mm,x、z轴平均旋转误差分别为0.6°和0.5°,平均耗时33.2 s。结论 对于较小的摆位误差,两种方法都具有较高的配准精度,但对于较大的摆位误差,本研究改进互信息配准方法较加速器自带的互信息配准方法具有显着的精度优势,并且配准耗时也在临床可以接受的范围内。     

Evaluation of median filtering after reconstruction with maximum likelihood expectation maximization (ML-EM) by real space and frequency space

Maximum likelihood expectation maximization (ML-EM) image quality is sensitive to the number of iterations, because a large number of iterations leads to images with checkerboard noise. The use of median filtering in the reconstruction process allows both noise reduction and edge preservation. We examined the value of median filtering after reconstruction with ML-EM by comparing filtered back projection (FBP) with a ramp filter or ML-EM without filtering. SPECT images were obtained with a dual-head gamma camera. The acquisition time was changed from 10 to 200 (seconds/frame) to examine the effect of the count statistics on the quality of the reconstructed images. First, images were reconstructed with ML-EM by changing the number of iterations from 1 to 150 in each study. Additionally, median filtering was applied following reconstruction with ML-EM. The quality of the reconstructed images was evaluated in terms of normalized mean square error (NMSE) values and two-dimensional power spectrum analysis. Median filtering after reconstruction by the ML-EM method provided stable NMSE values even when the number of iterations was increased. The signal element of the image was close to the reference image for any repetition number of iterations. Median filtering after reconstruction with ML-EM was useful in reducing noise, with a similar resolution achieved by reconstruction with FBP and a ramp filter. Especially in images with poor count statistics, median filtering after reconstruction with ML-EM is effective as a simple, widely available method.     

Registration accuracy of<Superscript> 153</Superscript>Gd transmission images of the brain

Purpose The aim of the study was to determine the accuracy of non-rigid nine-parameter image registrations based on ¹⁵³Gd transmission computed tomography (TCT) images as compared with those based on ^99mTc-ethyl cysteinate dimer (ECD) images and to assess whether normalised mutual information (NMI) or count difference (CD) should be used.Methods TCT and ECD data were acquired in 25 randomly selected patients. Emission images were registered to an ECD template with a CD cost function. The same registration parameters were applied to the transmission images to create a TCT template. All TCT images were registered to the TCT template and the same registration parameters were applied to the ECD images. The procedure was repeated with NMI as cost function. Accuracy of both ECD-based and TCT-based registrations was assessed by comparing the normalisation parameter values and regional activities in the spatially normalised ECD images, using a mixed-model analysis of variance (ANOVA). Scheffé post hoc tests were performed.Results No significant differences were found between ECD/CD, ECD/NMI and TCT/CD, suggesting that ECD registration can be done with either CD or NMI, and that TCT registration using CD is equally as accurate as ECD registration. The accuracy of TCT registration with NMI was lower, with discrepancies occurring in the frontal inferior region and the cerebellum. The analysis of normalisation parameters indicated that z-scaling is underestimated and yz-rotation overestimated with TCT/NMI registration.Conclusion We conclude that ECD registrations with CD or NMI are as accurate as TCT registrations with CD and that TCT registrations with NMI should be avoided.     

Lung compensator design using an electronic portal imaging device.

Using a liquid filled electronic portal imaging device (EPID) installed on a linear accelerator and a composite chest phantom, exit dose measurements were carried out to establish an empirical relationship between the pixel values of the imaging detector and the corresponding equivalent thickness of the overlying phantom material. Results for 6 and 10 MV photons show that the relationship depends on the so-called input/output characteristics of the imaging device for a particular photon energy. For a chest irradiation, an EPID image obtained under treatment geometry provides the pixel value information that is used to calculate the tissue deficit over the lung region. The compensators are made of lead whose thickness is calculated from the established empirical relationship to replace the tissue deficit over lungs. The effectiveness of the method is demonstrated with thermoluminescent dosimetry (TLD) for 6 and 10 MV beams. With compensators in place, the dose uniformity was found to be within +/- 5%.     

An empirical calibration method for an a-Si portal imaging device: applications in pretreatment verification of IMRT

PURPOSE: A new calibration method for an amorphoussilicon (a-Si) electronic portal imaging device (EPID) used for dose measurements in pretreatment verification (field-related) of intensity-modulated radiation therapy (IMRT) with sliding-window technique. The method is independent of data contained in the multileaf collimator (MLC) leaf-motion files and of any calculations made by the treatment planning system (TPS). MATERIALS AND METHODS: Sensitivity of the EPID is dependent on radiation energy. For fluence-modulated fields, different dose/reading calibration factors are associated with each pixel of the image acquired by calculating equivalent areas representing the exact ratio between primary and scatter components. The dose measured in the detector plane was compared with that calculated with TPS by using gamma-analysis. Each calibration factor was compared with that calculated by considering the individual contributions of primary and secondary radiation obtained using the convolution method with analytical kernel for homogeneous media. RESULTS: In 837/854 (98%) of the clinical fields analysed, the proportion of irradiated area in which the gamma-index was <1.0 exceeded 95%. The overall average gamma-index was 0.39. There was good agreement between the dose/reading calibration factors obtained with the empirical algorithm and with the convolution method. CONCLUSIONS: The proposed calibration method is suitable for routine clinical pretreatment verification in IMRT.     

Impact of JPEG lossy image compression on quantitative digital subtraction radiography

OBJECTIVES: The aim of the study was to evaluate the impact of JPEG lossy image compression on the estimation of alveolar bone gain by quantitative digital subtraction radiography (DSR). METHODS: Nine dry domestic pig mandible posterior segments were radiographed three times ('Baseline', 'No change', and 'Gain') with standardized projection geometry. Bone gain was simulated by adding artificial bone chips (1, 4, and 15 mg). Images were either compressed before or after registration. No change areas in compressed and subtracted 'No change-Baseline' images and bone gain volumes in compressed and subtracted 'Gain-Baseline' images were calculated and compared to the corresponding measurements performed on original subtracted images. RESULTS: Measurements of no change areas ('No change-Baseline') were only slightly affected by compressions down to JPEG 50 (J50) applied either before or after registration. Simulated gain of alveolar bone ('Gain-Baseline') was underestimated when compression before registration was performed. The underestimation was bigger when small bone chips of 1 mg were measured and when higher compression rates were used. Bone chips of 4 and 15 mg were only slightly underestimated when using J90, J70, and J50 compressions before registration. CONCLUSIONS: Lossy JPEG compression does not affect the measurements of no change areas by DSR. Images undergoing subtraction should be registered before compression and if so, J90 compression with a compression ratio of 1:7 can be used to detect and measure 4 mg and larger bone gain.     

Clinical practice and evaluation of electronic portal imaging device for VMAT quality assurance

Volumetric-modulated arc therapy (VMAT) is a novel extension of the intensity-modulated radiation therapy (IMRT) technique, which has brought challenges to dose verification. To perform VMAT pretreatment quality assurance, an electronic portal imaging device (EPID) can be applied. This study's aim was to evaluate EPID performance for VMAT dose verification. First, dosimetric characteristics of EPID were investigated. Then 10 selected VMAT dose plans were measured by EPID with the rotational method. The overall variation of EPID dosimetric characteristics was within 1.4% for VMAT. The film system serving as a conventional tool for verification showed good agreement both with EPID measurements ([94.1 ± 1.5]% with 3 mm/3% criteria) and treatment planning system (TPS) calculations ([97.4 ± 2.8]% with 3 mm/3% criteria). In addition, EPID measurements for VMAT presented good agreement with TPS calculations ([99.1 ± 0.6]% with 3 mm/3% criteria). The EPID system performed the robustness of potential error findings in TPS calculations and the delivery system. This study demonstrated that an EPID system can be used as a reliable and efficient quality assurance tool for VMAT dose verification.     

Quantitative comparison of automatic and interactive methods for MRI-SPECT image registration of the brain based on 3-dimensional calculation of error.

A wide range of techniques for registration of medical images has been devised in recent years. The aim of this study is to quantify the overall spatial registration error of 3 different methods for image registration: interactive matching, surface matching, and uniformity index matching as described by Woods. METHODS: MRI and ethylcysteinate dimer-SPECT images of the brain were registered for 15 patients. The matching error was assessed by determining intra- and interobserver variability of registrations. Quantification of the registration error was based on the mean spatial distance of 5000 voxels between 2 image positions. The mean position after repeated registrations in each patient was used as the gold standard. To evaluate the coherence of the 3 different registration methods, intermethod variability was determined. RESULTS: Interactive matching showed an intraobserver/interobserver variability of 1.5+/-0.3 mm/1.6+/-0.3 mm (mean +/- SD). The time demand for this method was 11+/-5 min. Surface matching revealed a variability of 2.6+/-1.1 mm/3.8+/-1.0 mm and a time demand of 26+/-12 min. Reproducibility of Woods' algorithm was 2.2+/-0.8 mm with a time demand of 9+/-3 min. In 4 of the 15 cases, Woods' method failed. The mean deviation between all 3 methods was 2.3+/-0.8 mm. CONCLUSION: With a suitable user interface, interactive matching had the lowest registration error. The influence of subjectivity was shown to be negligible. Therefore, interactive matching is our preferred technique for image fusion of the brain.     

The Electronic Portal Imaging System Siemens Beamview Plus® Versus the Conventional Verification Films CEA-TVS® and DuPont CQL-7®A Critical Appraisal of Visual Image Quality

AIM: The aim of this study was the validation of the visual image quality of electronic portal imaging devices (EPID) and conventional verification films from the point of view of the end-viewers of portal films, the radiotherapists. MATERIAL AND METHODS: The verification image was represented in two different forms, viz. an electronic portal image employing Siemens Beamview Plus (on a computer monitor) and two different portal films using the conventional verification films CEA-TVS and DuPont CQL-7 (on a negatoscope). A total of 270 image sets (simulation film and portal image) were evaluated by each radiotherapist, evaluation extending to 90 sets of each type of verification film. Each set was evaluated by three specialists in radiotherapy examining subjective visual image quality whereby the following aspects served as evaluation criteria: contrast, artifacts, determination of actual radiation field edge position, anatomical structures and main structural feature for the determination of treatment field position. In addition, the anatomical structures employed for visual feature correlation between reference and portal films were classified according to their importance. RESULTS: In general the electronic portal image was rated significantly "visible" or better. Only the evaluation of artifacts showed an appreciable disadvantage for electronic portal imaging caused by physical artifacts due to radiographic technique and data processing aspects peculiar to the Siemens Beamview Plus 1.1. and also caused by different image processing tools reducing physical artifacts and enhancing the visibility of anatomical structures and likewise of anatomical artifacts (e.g. intestinal gas). By calculating the Spearman correlation coefficient to detect a possible relationship between the different criteria of subjective visual image quality, the research demonstrated that artifacts when limited to a tolerable proportion had no significant impact on the other criteria. CONCLUSIONS: As data of EPIDS are digital, images can be postprocessed and enhanced in a wide variety of ways. Using this tool the electronic portal imaging device provides images that, in terms of visual image quality, are at least comparable to the two evaluated types of radiographic films and also have the added advantage that such images are stored and can be transferred electronically being presupposition for digital patient documentation.     

Motion corrected free-breathing delayed-enhancement imaging of myocardial infarction using nonrigid registration

PURPOSE: To develop and test an automatic free-breathing, delayed enhancement imaging method with improved image signal-to-noise ratio (SNR). MATERIALS AND METHODS: The proposed approach uses free-breathing, inversion-recovery single-shot fast imaging with steady precession (FISP) delayed-enhancement with respiratory motion compensation based on nonrigid image registration. Motion-corrected averaging is used to enhance SNR. RESULTS: Fully automatic, nonrigid registration was compared to previously validated rigid body registration that required user interaction. The performance was measured using the variance of edge positions in intensity profiles through the myocardial infarction (MI) enhanced region and through the right ventricular (RV) wall. Measured variation of the MI edge was 1.16 +/- 0.71 mm (N = 6 patients; mean +/- SD) for rigid body and 1.08 +/- 0.76 mm for nonrigid registration (no significant difference). On the other hand, significant improvement (P < 0.005) was found in the measurements at the RV edge where the SD was 2.06 +/- 0.56 mm for rigid body and 0.59 +/- 0.22 mm for nonrigid registration. CONCLUSION: The proposed approach achieves delayed enhancement images with high resolution and SNR without requiring a breathhold. Motion correction of free-breathing delayed-enhancement imaging using nonrigid image registration may be implemented in a fully automatic fashion and performs uniformly well across the full field of view (FOV).     

Development of automatic verification system for portal image in radiotherapy

The current method of verification for external beam radiation therapy visually compares a simulation image with a portal image. However, because this method depends largely upon the observer's experience, it sometimes results in inter-observer differences. In this study, we developed software to measure atomatically the quantitative difference between the simulation image and portal image using an image-analysis method. The feasibility of this software was evaluated on a rectangular field in the pelvic region. We took 12 simulation images of a pelvic phantom, setting 4 different field shapes on each of 3 isocenters. We then obtained 84 portal images setting 7 known distances from each of the 12 simulation images. Using this software, the direction of shift was detected correctly, and the distance of shift was detected correctly to within less than 3 mm. We consider that this software could be a useful method of verification.