Volume 33, Issue 6, June 2006
Index of content:
- General Poster Discussion: Exhibit Hall F
SU‐FF‐I‐01: Determination of Subjective Similarity for Pairs of Lesions On Mammograms: Comparison of Ranking Scores in 2AFC Versus Absolute Ratings for Masses and Microcalcifications33(2006); http://dx.doi.org/10.1118/1.2240239View Description Hide Description
Purpose: We previously obtained the subjective similarity ratings for pairs of lesions on mammograms for quantitative evaluation of similar images. Our purpose in this study was to investigate whether the absolute similarity ratings can be determined reliably by comparison with ranking scores obtained in a 2‐alternative forced‐choice (2AFC) method. Method and Materials: We selected 8 pairs of masses and 8 pairs of clustered microcalcifications based on radiologists' average similarity ratings; similarity ratings for the two sets of 8 pairs were approximately evenly distributed. In the first study, each pair was compared one by one to all other 7 pairs in each group of masses and microcalcifications. In the second study, we combined four pairs of masses and four pairs of microcalcifications to compare the similarity of a pair of masses with that of microcalcifications. Seven radiologists and 3 senior residents were asked to choose one pair that was more similar than another pair with the 2AFC method. The cases were presented in randomized order. The number of times that a pair was selected as more similar was counted as the subjective ranking score. The average scores were compared with the average similarity ratings determined previously. Results: The average ranking scores from the first study were highly correlated (0.93 and 0.98 for masses and calcifications, respectively) with absolute similarity ratings. When mass pairs were compared with calcification pairs, the correlations between ranking scores and absolute ratings were also very high (0.92 and 0.96). In both studies, observers were very consistent in selecting more similar pairs. Conclusion: The result indicates that absolute similarity ratings determined previously are reliable and useful for selection of similar images. The concept of similarity is robust and meaningful even when mass pairs are compared with microcalcification pairs.
33(2006); http://dx.doi.org/10.1118/1.2240240View Description Hide Description
Purpose: To estimate entrance exposure levels during on‐board kV imaging (Version 1.2) on Trilogy (Varian Medical Systems). Methods and Materials: The patient was simulated by phantom using 40 cm and 20 cm of 40 × 40cm2 water equivalent slabs. Exposure measurements were acquired for 80 and 90 cm source‐to‐surface distances using a 150 cc Fluke ionization chamber (96020C) and an Innovision 3050A dosimeter. The measurements were performed for various preset techniques that are commonly used in the clinic such as AP pelvis, Lat Pelvis, AP head, AP thorax, and AP extremity. Exposure rate levels were also measured during pulsed fluoroscopy with the automatic background control option activated. For comparison purposes, the exposure levels on a conventional simulator were also measured. Results: The entrance exposure levels on the on‐board imager vary between 0.13 mSv for an extremity technique to 4.9 mSv for a lateral pelvis technique and were comparable to the conventional simulator measurements. On‐board imaging pulsed fluoroscopy exposure levels were higher than those measured using the continuous fluoroscopy technique on the conventional simulator. Conclusions: Though the exposure and exposure rates are relatively low and inconsequential to the overall course of prescribed therapy, it is important to document exposures received. This documentation is essential for imaging protocols that may exceed normal imaging and localization exposure levels.
SU‐FF‐I‐03: Computation‐Efficient Cone Beam Image Reconstruction for Image‐ Guided Radiation Therapy Applications Using 3D Weighted Filtered Backprojection (CB‐FBP) Algorithm33(2006); http://dx.doi.org/10.1118/1.2240241View Description Hide Description
Purpose: To extend the 3D weighted cone beam filtered backprojection (CB‐FBP) algorithm for diagnosticCTimaging to image‐guidedradiation therapy(IGRT) applications. Method and Materials: 3D isotropic spatial resolution is one of the most attracting features of state‐of‐the‐art volumetric CT for diagnosticsimaging. However, in IGRT treatment planning, the CTimage slice thickness is usually larger than what is determined by detector row width (namely thin image). A straightforward way to generate thicker image is the combination of weighted adjacent thin images in image domain, which is computationally expensive because each thin image has to go through a computation expensive 3D backprojection. Another way is to carry out cross‐row filtering in projection domain, which may cause shading/glaring artifacts and uneven slice thickness as isotropic 3D geometry is distorted. To optimize both image quality (IQ) and computational efficiency, a virtual reconstruction plane (RP) based algorithm is proposed and implemented. By using the 3D weighted CB‐FBP algorithm, a thick image is still a weighted combination of adjacent thin images, but the combination is implemented in projection domain using virtual RPs. To maintain the IQ of thick image, the 3D weighted CB‐FBP algorithm is applied by tracking re‐sampled projection data. The tracking process is to improve computational efficiency further by making use of the projection data corresponding to involved virtual RPs only, while the re‐sampling process is to improve IQ by increasing the in‐plane sampling‐rate in virtual RPs. Results: By using a helical body phantom, spatial resolution phantom and 20 cm water phantom, the performance of the proposed algorithm, such as suppression of artifacts, uniformity of slice thickness and noise characteristics, are experimentally evaluated and verified. Conclusion: The experimental evaluation shows the proposed algorithm is indeed an optimized image reconstruction solution for IGRT applications in terms of both image quality and computational efficiency.
33(2006); http://dx.doi.org/10.1118/1.2240242View Description Hide Description
Purpose: Gain drifts and nonlinearities in amorphous silicon flat‐panel x‐ray detectors can produce ring artifacts in reconstructed cone‐beam computed tomography(CBCT)images. We have found that the magnitude of these artifacts can exceed 50 HU in clinical situations, and that the intensity of a given ring may not be uniform throughout an image. In some cases (e.g. half‐fan pelvic scans), discrete arcs may be produced. The goal of this study was to develop a post‐processing algorithm to efficiently suppress such variable‐intensity rings in axial slices. Method and Materials: Our approach builds upon the work of Sijbers and Postnov who showed that constant‐intensity rings can be estimated via radial median filtering of the input image after its transformation to polar coordinates. To characterize variable‐intensity rings and arcs, we developed a 2‐D estimation technique that uses a combination of row‐based (radial) and column‐based (angular) filters operating in the polar domain. The 2‐D estimates were transformed back to Cartesian space for subtraction from the original image. The new algorithm was implemented in C++ and tested on clinical and phantom CBCTimages acquired using a Varian 4030CB detector.Results: Correction times (3.2GHz Intel Pentium4 processor), including coordinate transformations, averaged 55 msec/slice for 512×512 matrix sizes. Rings and arcs were reduced in intensity by more than an order of magnitude to levels well below the background noise intensity. By subtracting ring estimates in Cartesian space, the polar matrix size could be reduced without sacrificing spatial resolution in the final image. This permitted for a 4× reduction in execution time compared to the original Sijbers‐Postnov approach where subtraction occurs in polar space. Conclusion: The Sijbers‐Postnov algorithm ring suppression algorithm was modified to provide improved image quality and fast execution times suitable for clinical implementation. Conflict of Interest: Funding provided by Varian Medical Systems.
SU‐FF‐I‐05: Hounsfield Units Calibration with Adaptive Compensation of Beam Hardening for a Dose Limited Breast CT System33(2006); http://dx.doi.org/10.1118/1.2240243View Description Hide Description
Purpose: Hounsfield Units calibration with adaptive compensation of beam hardening for a dose limited breast CT system. Method and Materials: Following a complete cone beam CT scan of the target object (a human breast), geometrical parameters of the object, including the mass center location and maximum radius from the mass center were calculated promptly. These parameters were used to compute X‐ray projection images of a water cylinder based upon photon attenuation in the cylinder and detector response. Projection images of the target object were corrected prior to reconstruction by logarithmic subtraction of the water cylinder projection images.CTimages relative to the Hounsfield Units (HU) scale were produced after cone beam reconstruction of the corrected projection images. Custom built water phantoms were tested with various inserts, including polyethylene, polystyrene, PMMA, nylon, polycarbonate and Teflon, with a density range from 0.92g/cm3 to 2.2g/cm3. Results: A wide range of breast diameters (10cm–18cm) and compositions (0%–100% glandular) were evaluated, and reconstructed and scaled HU values demonstrated excellent uniformity, linearity and consistency. Typical HU values were within 5% of theoretical values. The proposed method was applied to clinical breast scans. The “cupping” artifact caused by beam hardening in the original image was corrected as expected. Conclusion: Conventional methods of Hounsfield Units conversion are based on the scan of few fix‐sized water phantoms and lack the flexibility to compensate for the beam hardening from objects with various sizes. The introduced noise from water phantom scans is also not negligible, especially for a dose limited breast CT system with much lower mAs than whole body CT systems. The proposed method can compensate for beam hardening over a wide range of breast diameters and compositions without increasing noise contents in the image.
SU‐FF‐I‐06: A Portable Test Platform for Image Acquisition and Calibration for Cone Beam Computed Tomography (CBCT) and Region of Interest CBCT (ROI‐CBCT) On a Commercial X‐Ray C‐Arm System33(2006); http://dx.doi.org/10.1118/1.2240244View Description Hide Description
Purpose: We have developed a unique portable test platform (PTP) which enables CBCT for specimens and phantoms on standard commercial clinical x‐ray systems. This PTP can be used to acquire ROI‐CBCT projection images, where a lower resolution, lower dose image peripheral to a high resolution ROI is acquired. This is achieved either by acquiring an image using an Image Intensifier (II) with an ROI filter in the x‐ray beam or by combining images acquired separately with low and high resolution x‐ray detectors.Method and Materials: The CBCTimages are acquired as the object rotates on the computer‐controlled rotary table of the PTP. For ROI‐CBCT, a micro‐angiography (MA) detector or an ROI filter is mounted on the PTP. The PTP also provides for relative X, Y, Z adjustments. After coarse alignment adjustments of the PTP, fine translational and angular adjustments are made based on fluoroscopic imaging of a cylindrical calibration phantom. Results: The PTP allows quick assembly of the parts required for CBCT or ROI‐CBCT reconstruction, reduces initial setup time to < 45 min, and provides for setup reproducibility. The system can be aligned to within one pixel (43 micron for the MA detector), with angular alignments of pitch and roll of the object better than 0.7° and 0.1° respectively. Conclusion:, The PTP allows fast and reliable set‐up and alignment of CBCT specimens, for standard and for ROI‐CBCT applications. The PTP may enable wider use of CBCT and ROI‐CBCT for specimens and phantoms without a costly dedicated system.
(Partial support from NIH Grants R01‐NS43924, R01‐EB02873, R01‐HL52567, R01‐EB02916, and Toshiba Medical Systems Corporation).
33(2006); http://dx.doi.org/10.1118/1.2240245View Description Hide Description
Purpose: To quantify the anatomical and imaging characteristics of a novel anthropomorphic lung phantom constructed using plastination. Method and Materials: The pig's thorax was scanned in‐vivo at known partial pressures on a clinical CT (Siemens Sensation 16, 120kVP, 100mAs, recon 0.54×0.54×0.75mm∧3). The lungs were extracted, inflated, and fixed by intra‐tracheal perfusion of 10% formalin while the pulmonary vessels were injected with Silastic E RTV silicone. The specimen was dehydrated (remove and replace tissue fluid with an organic solvent) in cold acetone and the lungs were impregnated with a curable silicone polymer via slow decreasing pressure. Finally, the polymer was polymerized using a curing agent. The plastinated phantom was then scanned (120kVp, 200mAs, 0.43×0.43×0.75mm∧3). Anatomical features, volume measurements, and CT values were compared using in‐vivo and phantom clinical CT reconstructions. Results: The plastinated phantom is stable on the timescale of years and retains major anatomical features of the in‐vivo lung. The phantom airway volume was 66% of the in‐vivo measurement at inspiration but equal to the measurement at expiration. Vessel and lung volume comparisons were complicated by incompletely filled vessels and air pockets inside the phantom; nevertheless, lung volume measurements differed by less than 15%. Mean CT values of the cardiac tissue in the phantom (168 +/− 46) were 132 HU higher than in‐vivo (36 +/− 87). Mean CT values of the pulmonary tissue were nearly equivalent for both datasets, attributed to an 11% decrease in the apparent tissue density due to over‐inflation during plastination. Conclusion: This work shows that the novel plastinated lung phantom retains the anatomical and imaging characteristics of an in‐vivo lung. This accurate and complex lung phantom has many uses including imaging system comparisons, providing a known, stable reproducible complex background for visibility studies and will be used for our own studies in lungtomosynthesis optimization.
SU‐FF‐I‐08: Results of An Optical Fiber‐Based Dosimetry System for Use in Computed Tomography Characterization33(2006); http://dx.doi.org/10.1118/1.2240246View Description Hide Description
Purpose: Modern multi‐detector CT and cone‐beam CT offer wide beams, making the concept of CT dose index (CDTI) no longer valid for CTdosimetry. A real‐time OSLdosimetrysystem has been developed and is evaluated for CTdosimetry in this study. Comparisons with a pencil ionization chamber were made. Methods and Materials: The system utilizes the optically stimulated luminescence(OSL) of KBr:Eu. The size of the KBr:Eu single crystal dosimeter equals approximately 1 mm3. The dosimeter was affixed to the terminal end of a plastic fiber cable and placed in the center hole of a plastic cylindrical phantom. The distal end of the fiber cable was attached to OSL reader, containing a 658 nm red laser, and photo‐multiplying tube (PMT), and associated optics/ electronics. CT slices of 1 s duration were performed over a range of energies (80–140 kVp) and tube currents (60–350 mA), as well as slice thickness (5 and 10 mm) using a GE LightSpeed Ultra scanner. Gantry tilt dependence was investigated over a range of 40.5° (22° superior to 18.5° inferior). OSL data was obtained before, during, and after the scan at the rate of 10Hz. Results: Performance was determined in part by normalizing both the initial OSL intensity and the background‐subtracted integral OSL to exposure reported previously by an ionization chamber. Good correlation between exposure and OSL data was found. Initial intensity and background‐subtracted OSL normalized to exposure show coefficients of variation of ∼%5 or less. Significant deviation was observed between the ∼10 OSL measurements taken for each slice, presumably as a result of absorption of x‐rays by the patient table. Conclusions: Initial tests have shown that this OSLdosimetrysystem possesses great potential for faster CT characterization. This system may prove a valuable alternative to CTDI.
SU‐FF‐I‐09: Minimizing Scatter Artifact in Cone‐Beam CT Reconstruction Using Both Kilovoltage and Megavoltage Beams33(2006); http://dx.doi.org/10.1118/1.2240247View Description Hide Description
Purpose: To study the effect of imaging orientations on image quality of aggregated conebeam CT and to develop a strategy to minimize scatter artifact. Method and Materials: Orthogonal kilovoltage (kV) and megavoltage (MV) beam projections were acquired and used to reconstruct kV/MV aggregated CTimages. With a $15$‐degree fan‐beam angle, only $105$‐degree gantry rotation s needed to obtain a minimal scan coverage. Besides its short scanning time, the new scan technique has the advantage of getting high soft‐tissue contrast from kV beams and low scatter artifacts from MV beams. The kV and MV projection images were obtained by using a kV on‐board imager and MV portal imager mounted orthogonally on a Varian‐21EX LINAC. A linear model was established to fit the kV and MV attenuation values over the 15‐degree overlap region. MV projections were then converted into kV‐equivalent ones. Aggregated CTimages were reconstructed from both kV and converted MV projections jointly, using the filter‐back projection method. Results: Using 8 different kV/MV orientations, we reconstructed 16 aggregated CTimages for two phantoms, eight for each one. Each image was registered to its corresponding simulator‐CT image from a GE LightSpeed‐RT simulator. The reconstructed images show good contrast for both bones and soft‐tissue. However some of them have relatively severe scatter artifacts. We calculated the normalized mutual information (NMI) between the aggregated CT and simulator CTimages. The NMI exhibits a sinusoidal oscillation when plotted against the gantry start angle. The two maximums correspond to the two imaging orientations with kV (or MV) beams through the thinner (or thicker) side views of the phantoms, indicating that scatter from kV beams has more impact on image quality than MV beams. Conclusion: A strategy has been developed to determine the optimal imaging orientation for aggregated kV/MV CBCT.
33(2006); http://dx.doi.org/10.1118/1.2240248View Description Hide Description
Purpose: Compensating filters have been considered for use in flat‐panel based cone‐beam computed tomography(CBCT) for the primary purpose of reducing the range of exposure reaching the detector. The use of compensators can also have other benefits, including reducing the magnitude of x‐ray scatter reaching the detector, as well as reducing the x‐ray scatter induced dose within the patient. The magnitude of these effects is characterized for a set of compensators constructed for clinical cone‐beam CTimaging geometries. Method and Materials: A set of copper compensators has been constructed with a range of modulation (1:1, 4:1, 8:1) for compensation of a cylindrical phantom (16 cm diameter). Investigations were performed on a bench‐top flat‐panel CBCTsystem that matches the imaging geometry (source‐to‐axis distance of 100 cm, magnification of 1.4 to 1.85) of clinical cone‐beam CTsystems used in radiation therapy. The influence of the compensators on scatter‐to‐primary ratio (SPR) at the detector (Paxscan 4030A, Varian Medical Systems) and dose within the phantom (NE 2571 0.6cc ion chamber) was measured. The influence of the compensators on reconstructedCBCTimage quality (uniformity, accuracy) was assessed for cylindrical water baths. Results: Depending upon phantom size (16 or 32 cm) and imaging geometry compensation reduced x‐ray scatter at the detector by up to a factor of two. Dose at the center of phantom was reduced by 25 to 35% for the same primary fluence along the beam midline. The reduction of scatter was correlated with a reduction in cupping artifacts, however, spectral hardening by the compensator introduced other non‐uniformities in the water bath reconstructions.Conclusion: The implementation of simple compensators in CBCT has great potential for reducing x‐ray scatter at the detector and patient dose. Further investigations will focus on determining the optimal compensation scheme for patient and task specific CBCTimaging.
33(2006); http://dx.doi.org/10.1118/1.2240249View Description Hide Description
Purpose: MV CBCT is an image guidance modality which yields a 3D dataset representative of the patient anatomy in treatment position. During the acquisition, a series of low‐dose projection images are generated by exposing a high detection efficiency flat panel to short bursts of the linear accelerator beam. In order to avoid image artifacts, the projection images need to be “gain” corrected for variations in pixel intensity unrelated to traversed patient anatomy. These variations arise from differences in individual pixel sensitivities, as well as from the spectral and the intensity non‐uniformity of the incident beam. In this work, we propose and validate two treatment site‐specific gain correction (GC) strategies. Method and Materials: The first GC approach employs an open‐field in‐air image of the beam. Each subsequently acquired image is then divided by this GC image (GCI) to remove the effects of differing pixel gains as well as of incident beam non‐uniformity. The second approach acquires a GCI using a flattened beam. A tray with a 1/2 inch uniform lead plate is inserted in the accessory tray holder prior the gain image acquisition. The presence of the lead plate in the beam path results in spectral and spatial homogeneity of the beam incident on the flat panel imager.Results: The clinical scope of the two correction approaches was investigated by clinical MV CBCT acquisitions. In‐air gain calibration is well suited for head and neck imaging since the underlying spatial and spectral properties of the beam remain unaltered. However, for prostate cases the second calibration approach significantly reduces image artifacts related to the coupled phenomena of beam hardening and profile flattening. Conclusion: Site‐specific acquisition protocols that employ GCIs generated under different conditions result in substantial MV CBCTimage quality improvement. Conflict of Interest: Supported by Siemens.
SU‐FF‐I‐12: Validation of Geant4's Predictions On X‐Ray Scatter and Glandular Dose in Pendant‐Geometry Cone‐Beam Breast CT33(2006); http://dx.doi.org/10.1118/1.2240250View Description Hide Description
Purpose: The Geant4 toolkit is a freely available, widely supported base package for the simulation of particles through matter. This study aimed to test Geant4's accuracy by comparing its predictions for glandular dose and x‐ray scatter in pendant‐geometry cone‐beam breast CT against previously published experimental and simulated data. Method and Materials: We performed Monte Carlo simulations using the Geant4 package [Agostinelli et al, Nucl Instrum Meth A 506: 250–303, 2003] to recreate the conditions of three previously published papers on breast CT dose and scatter. Geant4's scatter simulations are compared against experimental data [Kwan et al, Med Phys 32(9): 2967–2975, 2005], while the dose results are compared against Monte Carlo simulations based on other codes [Boone et al, Med Phys 31(2): 226–235, 2004; Thacker et al, Phys Med Biol 49: 5433–5444, 2004]. The compared scatter results include scatter‐to‐primary ratio profiles for breasts of different sizes, glandularity and incident x‐ray spectra. For the dose comparisons, we compared Geant4's monochromatic results with the monochromatic results reported by the two previously published papers. Results: Geant4 matches the reported experimental SPR profiles to an accuracy of 1.6–16.7% (μ=9.2%, σ=5.4%). The sources for observed deviations include inexact re‐creation of the experimental setup and lack of specific information on the x‐ray spectra used in the experiments. The dose results agree with Boone's published results within 0.7–12.3% (μ=5.3%, σ=4.2%) and with Thacker's to within 1.2–22.3% (μ=10.8%, σ=8.8%). Conclusion: The data comparison suggests that Geant4 can be used to predict x‐ray scatter and dose deposition in low energy experiments such as dedicated breast CT. Given the availability, support and flexibility of the Geant4 toolkit, the use of this package for simulation of breast CT studies can be very useful to researchers in the field. Research supported in part by: NIH‐NIBIB Grant RO1‐EB002123 and the Georgia Cancer Coalition.
33(2006); http://dx.doi.org/10.1118/1.2240251View Description Hide Description
Purpose: Megavoltage Cone‐Beam CT (MVCBCT) has recently been introduced in the clinic to improve patient alignment prior to dose delivery. The objective of this research was to evaluate the dose delivered to patients for MVCBCT acquisition. We also studied the possibility of making simple plan modifications to compensate for the dose delivered by daily MVCBCT imaging.Method and Materials: Because MVCBCT uses the treatment beam, conventional CT scans (pelvis and head and neck patients) were imported in a treatment planning system (Phillips, Pinnacle) to simulate an MVCBCT acquisition. To validate the dose obtained from Pinnacle, a simple water‐equivalent cylindrical phantom with spaces for MOSFETs and an ion chamber was used to measure the actual dose delivered during MVCBCT. Results: The MVCBCT dose delivered to the phantom, calculated from Pinnacle, was within 3% to all the MOSFET measurement points. The difference between Pinnacle and the ion chamber was 0.2%. For a typical MVCBCT (arc: 270° to 110°) the delivered dose forms an anterior‐posterior gradient. Head and neck patients receive dose ranging from 0.7 to 1.2 cGy per MVCBCT monitor unit (MU). The range is 0.6 to 1.2 cGy per MVCBCT MU for pelvis patients. The total dose for daily positioning using MVCBCT can be reduced and made uniform by alternating between two opposed imaging arcs. Dose‐volume histograms of a compensated plan for a pelvis patient imaged with 10 MU MVCBCTs for 40 fractions show no additional dose to the target and small increases at low doses. Conclusion: Given that clinical MVCBCTs are currently performed at doses ranging from 2–15 MU, simple plan modifications, such as reducing the total number of MU, can be used to nearly eliminate the dose used for daily positioning. Results for other body sites will also be presented. Conflict of Interest: Research sponsored by Siemens OCS.
33(2006); http://dx.doi.org/10.1118/1.2240252View Description Hide Description
Purpose: Separating primary and scatteredradiation in cone‐beam Computed Tomography to improve image quality. Method and Materials: Earlier generations of CT uses narrow beams or fan beams of x‐rays, which suffer little from scattered x‐ray photons. Newer generations of CTsystems, however, use cone‐beamed x‐ray beams are associated with large amount of scatterradiations that tend to blur the reconstructed images based on the radiation absorbed by the detectors. In the low‐energy range (∼100 kV) of x‐ray photons involved in diagnostic imaging, the scatter‐to‐primary (SPR) is on the order of 1, as compared to megavoltage x‐rays, where typical SPR is on the order of 0.1. Swindell and Evans (Med. Phys., vol. 23, p. 63, 1996) had computed that the central axis SPR is almost linear with beam area, and is also almost linear with depth in water for a 6MV beam. Three methods are proposed to separate primary and scatteredradiation in cone‐beam CT based on the ideas in the field of radiation therapy: (1) using a relationship between SPR and distance to the radiation source, in conjunction with a two‐layer detector array; (2) using a pencil beam to sample primary dose; and (3) using Monte Carlo simulations to reconstruct primary and scatteredradiation.Results: Simulations suggest that separate primary image and scatteredimages can be obtained, with the primary image at an improved image quality as compared with the image from the total radiation.Conclusion: Separate primary and scatteredimages can be obtained through mathematical means and a simple upgrade of detector array to achieve better images and more information.
SU‐FF‐I‐15: Effects, Detection and Removal of Zingers From Scattered X‐Rays in CCD Based Cone Beam CT33(2006); http://dx.doi.org/10.1118/1.2240253View Description Hide Description
Purpose: Zingers are tiny spurious white dots that appear randomly in CCDimages. In order to improve the quality of CCD based cone beam CT technique, a new technique for the detection and removal of zingers is described and evaluated. Method and Materials: A bench top CCD based cone beam CT system was used to measure and investigate the presence of zingers. The cause and effects of zingers were studied. A new technique was developed to detect and correct the zingers. With this technique, the statistical behavior of pixel values in a projection image was first analyzed to identify candidates for zingers. Pixel values at the detected zinger locations were then compared in two consecutive projection views to eliminate false detections. To investigate and evaluate this technique, zingers were simulated by increasing the pixel values at randomly selected locations in projection data computed for a modified Shepp‐Logan phantom. The simulated data were then detected and corrected for zingers and used for reconstruction. The resulting reconstructed image was compared with the imagereconstructed from zinger free data and with imagesreconstructed from data corrected using three other zinger removal techniques. Results: Our measurement indicated that zingers may have resulted from scattered x‐rays. They were found to generate visible artifacts and degrade the quality of reconstructed images. It was shown that zingers detection by comparing two identically acquired projections could be highly effective but impractical in CTimaging. Detection by comparing two consecutive projection views was equally effective but may be subject image blurring. Detection by analyzingsignal fluctuations could result in a large number of faulty detections. The proposed new detection technique was found to be practical and effective without resulting in image blurring or faulty detections.
This work was supported in part by a research grant CA104759 from NIHNCI.
33(2006); http://dx.doi.org/10.1118/1.2240254View Description Hide Description
In this study, we investigate the feasibility of using VOI projection data acquired at high resolution in conjunction with full width projection data acquired at low resolution to reconstruct cone beam CTimages for the VOI. To simulate cone beam CT with dual resolution image acquisition, flat panel images of a mastectomy specimen, acquired in the non‐binning mode, were converted into low resolution full width projection data. High resolution VOI projection data were directly extracted from the original data. To prepare for reconstruction, the low resolution projection data were first interpolated, re‐sampled to fill in the truncated space outside the VOI. The dual resolution full width projection data, consisting of true high resolution data in the VOI and interpolated data outside the VOI, were then used to reconstruct the 3‐D image for the VOI. Reconstructed images obtained with dual resolution projection data were compared with those obtained with low resolution data and those obtained with high resolution data for the visibility of small calcifications. We have successfully demonstrated the use of dual resolution projection data for VOI cone beam CTimaging. While the low resolution full width projection data did not allow smaller calcifications to be seen in the reconstructed images, addition of high resolution projection data for the VOI only could make them visible. The use of interpolated low resolution projection data to pad the truncated space outside the VOI did not affect the spatial resolution of reconstructed images inside the VOI. With the dual resolution technique, it would be possible to selectively image a VOI at very high resolution without requiring excessively long acquisition and reconstruction or unnecessarily overexposing the patient outside the VOI.
(This work was supported in part by a research grant CA104759 from the NCI and a research grant EB‐00117 from the NIBIB).
SU‐FF‐I‐17: Reduction of Set‐Up Error Using Cone Beam‐CT in Patients Undergoing Partial Breast Irradiation33(2006); http://dx.doi.org/10.1118/1.2240255View Description Hide Description
Purpose: Partial breast irradiation with multiple external beams requires accurate alignment of the target volume and treatment isocenter. On‐board plannar kV/MV imaging can verify set‐up based on boney landmarks. On‐board cone beam CT(CBCT) may provide additional soft‐tissue‐based information to further improve set‐up accuracy. In this study, we assess the utility of CBCT in patients receiving partial breast radiation therapy, who have been aligned with on‐board plannar kV/MV imaging.Method and Materials: Patients undergoing partial breast irradiation were imaged using an on‐board‐imager attached to the gantry of a Varian 21EX machine. Each patient was aligned to skin marks in the treatment position. Orthogonal kV images were acquired and registered to digitally reconstructedradiographs from the planning CT. Subsequently, a CBCTimage data set was acquired and compared to the planning CT using both boney anatomy and soft tissue information which yielded estimations of residual setup error. Results: 10 patients were studied under an IRB protocol. Each patient had an average acquisition of 9 pre‐RT CBCTimages. After 2D kV image registration based on boney anatomy, the average soft tissue residual error, based on the CBCT vs. planning CT, was 4, 2, and 3 mm in the Ant/Post, Sup/Inf, and Rt/Lt directions respectively. The ranges were 0–15, 0–9, and 0–7 mm, respectively. In one case, 30% of the planning tumor volume (PTV) would have been outside of the therapeutic isodose volume based on the pre‐RT CBCTimages.Conclusion:CBCT is a practical tool for image registration while visualizing the soft tissue of the tumor bed. It provides additional anatomic information and soft‐tissue detail beyond that provided by planar radiographs. Such increased accuracy can significantly improve dosimetric coverage of the planning target volume. Conflict of Interest: Partially supported by a Varian research grant.
33(2006); http://dx.doi.org/10.1118/1.2240256View Description Hide Description
Purpose: To quantify the geometric accuracy of the On Board Imager in both the kV radiographic and cone beam imaging modes. Method and Materials: The Winston‐Lutz test was performed to localize a 5mm tungsten sphere placed within +/− 0.25 mm of the radiation isocenter. The sphere was imaged with half fan cone beam scans, and kV radiographs at the 4 principal gantry angles. The displacement of the sphere from the ‘imaging isocenter’ (the actual position of a point object that the imaging system would find to be at isocenter) was determined for each imaging mode. This test has been repeated 18 times over a period of one year. Results: The average displacement of the sphere from the imaging isocenter using a half fan technique was found to be 0.9 mm Right, 0.9 mm Anterior, and 1.1 mm Inferior, assuming a head first supine orientation. These offsets are incorporated in image‐guided patient setup procedures. Small systematic errors as a function of gantry angle were also measured for the radiographs. A point at the radiation isocenter will appear about 1mm higher in a right lateral image than in a left lateral image. A similar left / right discrepancy exits for anterior and posterior images.Conclusion: The systematic geometric errors of the kV imaging equipment and associated techniques need to be measured and incorporated into the procedure of on‐line image‐guidedpatient treatment. For the On Board Imager, a geometric accuracy of better than 1mm can be achieved.
33(2006); http://dx.doi.org/10.1118/1.2240258View Description Hide Description
Purpose: To study the impact of projection numbers on the quality of tomosynthesisimages using megavoltage beams. Method and Materials: It has been shown that isocentric kilovoltage tomosynthesisimages can be generated with no more than 50 degree scan angle. With only one seventh of scan time and dose of CBCT, this new technology can generate on‐board images with comparable quality as CBCT to guide patient's positioning. In this study, we used megavoltage (MV) beam projections to reconstructtomosynthesisimages. The MV projection images were acquired by using a MV portal imager mounted on a Varian‐21EX LINAC. The high‐resolution mode was used for acquisition and it required less than 1 cGy dose for each projection. The raw projections data needed background and floor image corrections and was then taken to generate tomosynthesisimages by using the FDK cone‐beam reconstruction algorithm. Results: Compared with kV tomosynthesisimages, MV tomosynthesisimages also give good contrast for bones. However, for soft tissue, their contrast is relatively lower. We calculate normalized mutual information (NMI) between MV and kV tomosynthesisimages and studied its dependence on scan angle and angular interval. The experiment shows that with the angular interval fixed, NMI increases as the scan angle grows and with the scan angle fixed, NMI decreases as the angular interval increases. For our study on a spine phantom, 50‐degree scan angle and 1‐degree angular interval gives good trade‐off between images quality and projection numbers. Conclusion: MV beam can be used to generate good quality tomosynthesisimages.
SU‐FF‐I‐20: Digital Tomosynthesis Mammography (DTM) : Dependence of Reconstruction Image Quality On Number and Angular Range of Projection Views33(2006); http://dx.doi.org/10.1118/1.2240259View Description Hide Description
Purpose: We have previously found in a phantom study that a simultaneous algebraic reconstruction technique (SART) could achieve similar image quality to maximum likelihood reconstruction, but with fewer iterations. In this study, we applied SART to patient and breast phantom images and evaluated the reconstructed image quality when the number and the angular range of the projection views (PVs) were varied. Method and Materials: A second generation GE prototype DTM system was used for image acquisition. PVs are acquired from 21 angles in 3° increments over a ±30° range in less than 8‐sec. The DTM system uses an Rh/Rh x‐ray source for all exposures. The total dose of the 21 PVs is set to be about 1.5X of that of a single‐view mammogram at the corresponding breast thickness. The phantom is composed of four 1‐cm‐thick breast‐shaped slabs of heterogeneous or homogeneous mixtures of fibroglandular‐and‐fatty‐tissue‐mimicking material with embedded masses. A 5×6 array of contrast‐detail disk‐shaped holes were drilled on one of the homogeneous slabs. DTMs were reconstructed with all 21 PVs, or subsets of the PVs to simulate different angular ranges and increments. The use of subsets of PVs resulted in a reduction in the total dose of the reconstructedDTM in this study. The image quality of the reconstructedDTMs under different conditions was compared. Results: For both the phantom and patient DTMs,image sharpness and contrast decreased with decreasing numbers of PVs used in the reconstruction. The interplane artifacts increased with decreasing angular range of the PVs. Conclusion: The image quality of DTMs depends on the number of PVs and the angular range used in image acquisition. Further investigation is needed to evaluate trade‐offs between the angular increment and angular range, as well as the effects of reconstruction parameters on image quality when the number of PVs varies.