Big Bore Research Articles

A metal artifact reduction scheme for accurate iterative dual-energy CT algorithms

CT images have been used to generate radiation therapy treatment plans for more than two decades. Dual-energy CT (DECT) has shown high accuracy in estimating electronic density or proton stopping power maps for the treatment. However, the presence of metal implants introduces severe striking artifacts in the reconstructed results, affecting the diagnostic accuracy and treatment performance. In order to reduce the metal artifact in DECT, we introduce a metal artifact reduction scheme for iterative DECT algorithms. The corrupt data is substituted with an estimation in each iteration. We utilize normalized metal artifact reduction (NMAR) composed with image domain decomposition to initialize the algorithm and speed up the convergence. A joint statistic DECT algorithm, dual-energy alternating minimization (DEAM), with the proposed scheme is tested on experimental and clinical data acquired on a Philips Big Bore scanner. We compared DEAM with the proposed method to the original DEAM and vendor reconstructions with and without O-MAR. The visualization and quantitative analysis show that DEAM with the proposed method has the best performance in reducing striking artifacts caused by metallic objects.

Deriving the Effective Atomic Number with a Dual-Energy Image Set Acquired by the Big Bore CT Simulator

Background: This study aims to determine the effective atomic number (Zeff) from dual-energy image sets obtained using a conventional computed tomography (CT) simulator. The estimated Zeff can be used for deriving the stopping power and material decomposition of CT images, thereby improving dose calculations in radiation therapy. Materials and Methods: An electron-density phantom was scanned using Philips Brilliance CT Big Bore at 80 and 140 kVp. The estimated Zeff values were compared with those obtained using the calibration phantom by applying the Rutherford, Schneider, and Joshi methods. The fitting parameters were optimized using the nonlinear least squares regression algorithm. The fitting curve and mass attenuation data were obtained from the National Institute of Standards and Technology. The fitting parameters obtained from stopping power and material decomposition of CT images, were validated by estimating the residual errors between the reference and calculated Zeff values. Next, the calculation accuracy of Zeff was evaluated by comparing the calculated values with the reference Zeff values of insert plugs. The exposure levels of patients under additional CT scanning at 80, 120, and 140 kVp were evaluated by measuring the weighted CT dose index (CTDIw). Results and Discussion: The residual errors of the fitting parameters were lower than 2%. The best and worst Zeff values were obtained using the Schneider and Joshi methods, respectively. The maximum differences between the reference and calculated values were 11.3% (for lung during inhalation), 4.7% (for adipose tissue), and 9.8% (for lung during inhalation) when applying the Rutherford, Schneider, and Joshi methods, respectively. Under dual-energy scanning (80 and 140 kVp), the patient exposure level was approximately twice that in general single- energy scanning (120 kVp). Conclusion: Zeff was calculated from two image sets scanned by conventional single-energy CT simulator. The results obtained using three different methods were compared. The Zeff calculation based on single-energy exhibited appropriate feasibility.

Characterization and comparison of imaging contrast enhancement with PEG-functionalized gold nanoparticles in kV cone beam computed tomography and computed tomography imaging

The purpose of this study is to define a simplified method to accurately predict and characterize kV cone beam computed tomography (kV CBCT) and computed tomography (CT) image contrast enhancement from gold nanoparticles (GNPs). Parameters of the kV CBCT of a Varian Novalis Tx linear accelerator and of a GE LightSpeed 4 Big Bore CT machine were modeled using the MCNP 6.2 Monte Carlo code. A 0.25 × 0.25 cm2 source, defined with a 100 kVp energy spectrum with appropriate filtration, was implemented in the MCNP6.2 model for kV CBCT, which also contained x- and y-blades and a full bowtie filter. A 1 cm3 cube of GNP solution (modeled as a mass percentage of gold in water) was placed 100 cm below the source. For the CT-simulator model, a source was defined with energy spectra for 80 and 140 kVp x‐rays with appropriate filtration and angular spectrum. A 1 cm3 GNP solution was modeled as before and a detector was placed 40 cm below that. Attenuation coefficients of four GNP solutions were computed and Hounsfield unit (HU) values were calculated. The computed HU values were compared against experimentally measured values obtained by scanning batches of GNPs of various sizes and concentrations using a GE LightSpeed 4 Big Bore CT scanner at 80 kVp and 140 kVp energies, as well as the kV CBCT capability of a Varian Novalis Tx linear accelerator. HU analysis was carried out using Velocity Medical Solutions clinical CT image analysis software. The MCNP calculated HU values matched the measured values to within ± 5%. Image contrast enhancement analysis showed a total increase in HU of up to 223. The sample having the highest gold mass percentage tested showed the greatest increase in HU number compared to water.

Chest CT scan features from 302 patients with COVID-19 in Jordan

ObjectiveTo study non-enhanced computer tomographic features of chest imaging in 302 patients with Corona Virus Disease 2019 (COVID-19) in Jordan. Material and methodA retrospective analysis of non-enhanced computer tomographic scans has been performed in the main center for patients diagnosed with COVID-19 in Prince Hamzah Hospital for those scanned from 13th of March 2020 to 13th of May 2020. Included scans were routinely performed during 24-hs of admission apart from having respiratory complaint. CT protocol included non-enhanced 1 mm slice thickness by Philips Brilliance Big Bore scanner (Philips; Amsterdam, Netherlands).All computer tomographic scans were reviewed by two senior radiologists with more than 8 years of experience each and senior registrar.Several factors have been thoroughly studied including patient age, gender, positive versus negative pulmonary findings, laterality of lung involvement, lobar distribution, pattern of pulmonary changes on initial and follow-up scans. ResultsThe total number of patients evaluated was 302. There were 188 men and 114 women studied. Among the totally studied 302 cases; 181 cases (59.9 %) showed no pulmonary changes.Positive findings were present in 121 patients with a total number of 191 computer tomographic scans including initial and follow-up scans. Positive findings were present in 51 female and 70 male patients (age range, 12–87 years; mean age ± standard deviation, 46.1 ± 16.5).Bilateral disease was more frequently encountered presented in 86 cases (71.1 %), while unilateral disease showed two times more predilection for the right lung compared to the left. The incidence of lobar involvement in descending order: right lower (75.2 %), left lower (71.9 %), right upper (62.8 %), left upper (60.3 %) and right middle (50.4 %). The incidence of the affected lobes on the initial scans were as follow: one lobe (24 %), two lobes (10.7 %), three lobes (9.1 %), four lobes (16.5 %) five lobes (36.4 %). In cases with single lobar involvement (24 %); the left upper and right middle lobes showed lowest incidence of involvement accounting for 10.3 % & 13.8 %, respectively; on the other hand, in cases with four lobar involvement (16.5 %); the right middle lobe was most commonly spared in two third of cases (63.2 %).Several initial patterns of the pulmonary changes resulting from Corona Virus Disease 2019 (COVID-19) were present with a descending order; ground-glass pattern (96.7 %), lenticular pattern (32.2 %), Halo sign (15.7 % %), rounded (14.9 %), nodular (10.7 %), ground-glass with consolidation (8.3 %), tree-in-bud (1.7 %) and pleural effusion (1.7 %). Pathologically-enlarged lymph node was not a feature of COVID-19.The total number of patients with positive findings having follow-up scans was 57 including single (45 patients) versus two (12 patients) follow-up scans. Initial follow up scans showed regression and progression of the pulmonary changes in 35 and 22 patients, respectively.A remarkable pattern was seen in almost all regressed cases that showed patchy reticular pattern changes with septal thickening which was referred to “pulmonary synapses” (34 patients) with only one patient showed complete resolution of the parenchymal changes. Patterns seen in progressed cases were lenticular ground-glass (63.6 %) vs patchy ground-glass (36.4 %) patterns. ConclusionComputer tomographic scan of the chest is a principal diagnostic measure for Corona Virus Disease 2019 (COVID-19). The pulmonary changes showed more propensity being bilateral disease and affecting the lower lobes, while the right middle lobe was the least likely involvement. Several pattern of pulmonary changes can be seen on initial scans including ground glass, consolidation, Halo sign, lenticular, nodular, pleural effusion and tree-in-bud patterns. The tree-in-bud is first-time described pattern of COVID-19 in the current article and thought to be an excluding criterion. In follow-up scans; the lenticular and patchy ground-glass patterns were present in cases with disease progression, compared with “pulmonary synapses” pattern encountered in cases with disease regression.

Assessment of Image Quality and Dosimetric Performance of CT Simulators

BackgroundCT simulator for radiation therapy aims to produce high-quality images for dose calculation and delineation of target and organs at risk in the process of treatment planning. Selection of CT imaging protocols that achieve a desired image quality while minimizing patient dose depends on technical CT parameters and their relationship with image quality and radiation dose. For similar imaging protocols using comparable technical CT parameters, there are also variations in image quality metrics between different CT simulator models. Understanding the relationship and variation is important for selecting appropriate imaging protocol and standardizing QC process. Here, we proposed an automated method to determine the relationship between image quality and radiation dose for various CT technical parameters. Material and MethodThe impact of scan parameters on various aspects of image quality and volumetric CT dose index for a Philips Brilliance Big Bore and a Toshiba Aquilion One CT scanners were determined by using commercial phantom and automated image quality analysis software and cylindrical radiation dose phantom. Results and DiscussionBoth scanners had very similar and satisfactory performance based on the diagnostic acceptance criteria recommended by ACR, International Atomic Energy Agency, and American Association of Physicists in Medicine. However, our results showed a compromise between different image quality components such as low-contrast and spatial resolution with the change of scanning parameters and revealed variations between the two scanners on their image quality performance. Measurement using a generic phantom and analysis by automated software was unbiased and efficient. ConclusionThis method provides information that can be used as a baseline for CT scanner image quality and dosimetric QC for different CT scanner models in a given institution or across sites.

Relevance in Standardizing Hearing Apparatus Contours

The potential for sensorineural hearing loss (SNHL) is an unfortunate reality with patients who receive radiation therapy (RT) to the head and neck (H&N) region. A review of the literature seems to indicate there is a lack in standards for contours associated with SNHL. The range of contoured structures reported by institutions include the tympanic cavity, Eustachian tube, mastoid process, cochlea, vestibule, inter auditory canal (IAC), and various combinations of the aforementioned items. At our institution, we have historically contoured the cochlea, vestibule, and a portion of the inner auditory canal as a single volume for each side (i.e. the bony labyrinth of the inner ear) and used that structure for optimization and dosimetric analysis. This retrospective study examines how much dose the cochlea, vestibule, and IAC individually received when the bony labyrinth of the inner ear was contoured and used as an avoidance structure during planning. Twenty consecutive patients who received IMRT for H&N cancer in 2015 at our institution were selected for this study. A Phillips Brilliance Big Bore 16 slice CT scanner was used for the RT simulation and images were acquired at a slice thickness of 3 mm. Contouring was done on a Phillips Pinnacle treatment planning system (TPS) and planned on an Accuray TomoTherapy TPS. The dose was transferred from the TomoTherapy TPS to the Pinnacle TPS where dose statistics were collected. A paired t-test analysis was performed with 40 data points of each structure (20 entries from each anatomical side) using SPSS software. Mean dose to our planning bony labyrinth organs at risk (OAR) was 12.95 Gy (standard deviation (SD) 9.96; standard error mean (SEM) 1.58). Cochlea mean dose = 13.66 Gy (10.42 SD; 1.65 SEM); Vestibule mean dose 12.05 Gy (9.57 SD, 1.51 SEM); and IAC mean dose = 14.24 Gy (11.35 SD, 1.80 SEM). Statistically significant differences were seen as we compared mean doses between our bony labyrinth to the cochlea (P=.0168) and the vestibule (P = .0001), but not to the IAC (P = .0877). Our results showed significant differences between the mean dose of the bony labyrinth versus the mean doses to the cochlea and vestibule but not with the IAC when the bony labyrinth of the inner ear is contoured as a single structure instead of having separate delineation of its components. Considering that there is ongoing research investigating the potential differences in tolerances for the various hearing structures, it is important to keep in mind that non-standard or ill-defined contouring of these structures may result in variations of dosimetric data between studies. This may lead to false correlations between dose given to OAR’s and clinical effect.

Image quality for radiotherapy CT simulators with different scanner bore size

PurposeWe compare image quality parameters derived from phantom images taken on three commercially available radiotherapy CT simulators. To make an unbiased evaluation, we assured images were obtained with the same surface dose measured using XR-QA2 model GafChromic™ film placed at the imaging phantom surface for all three CT-simulators. MethodsRadiotherapy CT simulators GE LS 16, Philips Brilliance Big Bore, and Toshiba Aquilion LB were compared in terms of spatial resolution, low contrast detectability, image uniformity, and contrast to noise ratio using CATPHAN-504 phantom, scanned with Head and Pelvis protocols. Dose was measured at phantom surface, with CT scans repeated until doses on all scanners were within 2%. ResultsIn terms of spatial resolution, the GE simulator appears slightly better, while Philips CT images are superior in terms of SNR for both scanning protocols. The CNR results show that Philips CT images appear to be better, except for high Z material, while Toshiba appears to fit in between the two simulators. ConclusionsWhile the image quality parameters for three RT CT simulators show comparable results, the scanner bore size is of vital importance in various radiotherapy applications. Since the image quality is a function of a large number of confounding parameters, any loss in image quality due to scanner bore size could be compensated by the appropriate choice of scanning parameters, including the exposure and by balancing between the additional imaging dose to the patient and high image quality required in highly conformal RT techniques.

25. Étude d’un faisceau sans cône égalisateur en conditions stéréotaxiques

Introduction As part of a phase 2 randomized clinical trial that compares Internal Target Volume (ITV) and Mid-Position (Mid-P) conformational planning strategies in locally advanced non-small cell lung cancer treatment, we investigated the impact of the type of CT image, i.e. Maximum Intensity Projection (MIP) image and Mid-P image on the inter-observer delineation variability. Methods We considered 12 patients among the 40 patients included in the clinical trial. For each patient, dataset for the delineation consisted in MIP image and Mid-P image, computed from four-dimensional CT scan (Big Bore, Philips). Primary tumor (IGTVT for MIP image and GTVT for Mid-P image) and pathological lymph nodes for 8 of the 12 patients (IGTVN for MIP image and GTVN for Mid-P image) were delineated on both images by an expert in lung radiation oncology and 2 or 3 radiation oncologists. The expert contours were considered as a reference and were compared to the others using paired Dice coefficient to estimate the overlap, Bidirectional Local Distance (BLD) to evaluate a Hausdorff distance-based between contours and the reference and Root-Mean-Square Deviation (RMSD) to measure volume variations with respect to the reference. Results The Mid-P delineated volumes were smaller than the MIP ones in average: 2.96 ± 2.75 cc and 16.7 ± 20.7 cc for nodes and primary tumors respectively. For the whole cohort, the DICE means and standard deviations were 0.685 ± 0.102, 0.706 ± 0.124, 0.777 ± 0.0926 and 0.784 ± 0.0756 for the GTVN, IGTVN, GTVT and IGTVT respectively. The absolute means BLD between the expert contours and the others were, in the same order, 2.08 ± 0.819 mm, 2.22 ± 1.36 mm, 2.60 ± 1.34 mm and 2.92 ± 1.35 mm. The RMSD were 3.05 ± 1.92 cc, 4.15 ± 4.66 cc, 13.4 ± 13.1 cc and 17.9 ± 18.3 cc respectively. Conclusions For one patient, we excluded an observer because of a non-complete contour. Regarding GTVN and IGTVN, we observed differences in terms of node selections and also delineation interpretations (nodes vs. lymph node areas). To ensure robust data, we only kept nodes selected by the expert. The presence of atelectasis and fibrosis led to large variations of primary tumor delineation (GTVT and IGTVT). The volumes and its variations decreased between MIP and Mid-P contours. Dice coefficients were lower with MIP compared to Mid-P. However, this metric is volume-dependent and tends to be lower for smaller volume. It may not be adapted here. The BLD was not statistically different. Finally, the volume variations were reduced but our results depended on delineation difficulties and did not demonstrate the statistical superiority of the Mid-P strategy in terms of inter-observer delineation variability.

24. Does Mid-P CT image decrease inter-observer variability compared to MIP CT image?

Introduction As part of a phase 2 randomized clinical trial that compares Internal Target Volume (ITV) and Mid-Position (Mid-P) conformational planning strategies in locally advanced non-small cell lung cancer treatment, we investigated the impact of the type of CT image, i.e. Maximum Intensity Projection (MIP) image and Mid-P image on the inter-observer delineation variability. Methods We considered 12 patients among the 40 patients included in the clinical trial. For each patient, dataset for the delineation consisted in MIP image and Mid-P image, computed from four-dimensional CT scan (Big Bore, Philips). Primary tumor (IGTVT for MIP image and GTVT for Mid-P image) and pathological lymph nodes for 8 of the 12 patients (IGTVN for MIP image and GTVN for Mid-P image) were delineated on both images by an expert in lung radiation oncology and 2 or 3 radiation oncologists. The expert contours were considered as a reference and were compared to the others using paired Dice coefficient to estimate the overlap, Bidirectional Local Distance (BLD) to evaluate a Hausdorff distance-based between contours and the reference and Root-Mean-Square Deviation (RMSD) to measure volume variations with respect to the reference. Results The Mid-P delineated volumes were smaller than the MIP ones in average: 2.96 ± 2.75 cc and 16.7 ± 20.7 cc for nodes and primary tumors respectively. For the whole cohort, the DICE means and standard deviations were 0.685 ± 0.102, 0.706 ± 0.124, 0.777 ± 0.0926 and 0.784 ± 0.0756 for the GTVN, IGTVN, GTVT and IGTVT respectively. The absolute means BLD between the expert contours and the others were, in the same order, 2.08 ± 0.819 mm, 2.22 ± 1.36 mm, 2.60 ± 1.34 mm and 2.92 ± 1.35 mm. The RMSD were 3.05 ± 1.92 cc, 4.15 ± 4.66 cc, 13.4 ± 13.1 cc and 17.9 ± 18.3 cc respectively. Conclusions For one patient, we excluded an observer because of a non-complete contour. Regarding GTVN and IGTVN, we observed differences in terms of node selections and also delineation interpretations (nodes vs. lymph node areas). To ensure robust data, we only kept nodes selected by the expert. The presence of atelectasis and fibrosis led to large variations of primary tumor delineation (GTVT and IGTVT). The volumes and its variations decreased between MIP and Mid-P contours. Dice coefficients were lower with MIP compared to Mid-P. However, this metric is volume-dependent and tends to be lower for smaller volume. It may not be adapted here. The BLD was not statistically different. Finally, the volume variations were reduced but our results depended on delineation difficulties and did not demonstrate the statistical superiority of the Mid-P strategy in terms of inter-observer delineation variability.

3. Feasibility of prostate treatment plan based on MRI images

Introduction Computerized Tomography (CT) scan provides Hounsfield Unit information (HU) and so Electronic Density (ED) which is necessary for planned dose treatment calculation in radiotherapy (RT). For prostate cancer and for delineation issues, MRI provides excellent soft tissue contrast compared to CT images and it is increasingly used for target delineation. Our project was first to develop our own algorithm conversion and image transformation between MRI signal intensity into HU and then to evaluate the algorithm on prostate clinical case as previous studies have obtained encouraging results [1] . Methods First, we set up MRI parameters (GE Signa 1.5 T, Lava Flex sequence) so that signal information matches CT-scan ones (Philips Big Bore). We selected 4 pelvic patients that were installed on a flat table to have a close installation to that of CT-scan. Using Matlab R2016a, we developed our own algorithm of conversion and image transformation between MRI signal intensity and HU using 3DSlicer 4.5 for specific density segmentation. MRI intensity signals for muscles and soft tissue were similar: we needed two conversion functions (soft tissues and bones). Images produced thanks to conversion functions were compared to real CT-scan images used for RT. We compared the depth dose profile of a 6MV field over 5 × 5 cm2 targeting the prostate and a 6 MV Volumetric Modulated Arc (VMAT). VMAT dose distributions were compared in terms of Gamma Index criteria. Results Preliminary results gave distinct intervals function definition for different structures, which was similar to the Finish-Helsinki university hospital study [1] , [2] . We identified spongy bone and cortical bone (bone function), fat, muscle, urine and prostate (soft tissue). Those distinct intervals testified the accuracy of our image conversion to produce the pseudo-CT. Dose differences were less than 5% outside the build-up region between depth dose profiles of a 6MV beam. For the VMAT treatment plan, 99% of points passed the gamma criterion of 3% dose and 3 mm distance to agreement. Finally, the pseudo-CT produced using our model showed a small effect on dose differences compared to real CT images. Conclusions This study revealed promising results to the possibility of using specific sequences of MRI for dose calculation in RT. Taking into account the potential limitation of our work, further investigation could be necessary. It could be particularly interesting to increase the number of patients, to improve the acquisition conditions and to integrate the different uncertainties of the MRI in the model. Finally, we plan to study the potential and feasibility of this model for other localization such as lung and head&neck.

10. Assessment and optimization of PET contours recovery for respiratory-gated radiotherapy

Introduction The development of medical imaging, and more specifically functional imaging, open new perspectives for radiotherapy treatments. The joint use of the dosimetric CT and the PET improve the determination of target volume by combining functional and morphological data. The purpose of this study is to evaluate the recovery process of the PET contours for respiratory-gated radiotherapy according to two methods: the Replanning algorithm (RP) and the module Integrated Registration (IR) from General Electric (GE) Medical System. Methods The 4D acquisitions are achieved on a dosimetric CT Brillance Big Bore® (Philips Medical System) associated to an abdominal belt Pneumo chest Bellows (Lafayette Instrument) and on a PET/CT Discovery 710 (GE Medical System) coupled to the respiratory gating system RPM (Real-Time Position Management™, Varian Medical System). The Quasar™ Programmable Respiratory Motion Phantom (Modus Medical Device Inc) is used with a lung density insert. The lesions are simulated by solid spheres of 1 and 3 cm diameter and another empty one of 2 cm. The acquisitions have been performed with breathing amplitude of 2 and 4 cm and breathing frequency of 10, 15 and 20 bpm. The influence of the acquisition parameters, reconstruction parameters and variability of the patient parameters (amplitude, frequency, lesion size) are being assessed. On the CT images, the spheres are segmented with a threshold allowing to obtain theoretical volume of the sphere on a 3D acquisition. On the PET images, the spheres are segmented with a 42% threshold of the SUVmax. The PET/CT contours obtained with the RP algorithm and the IR module have been compared to the contours delineated on the dosimetric CT with Eclipse v13.5, using the Dice’s coefficient. Results The influence of the acquisition and reconstruction parameters studied with the RP algorithm show Dice’s coefficients ranging from 0.93 to 0.96. Dice’s coefficients variating between 0.85 and 0.94 are obtained for the study of variability of patients parameters. With regard to the two retrieval methods of PET contours, they show similar Dice’s coefficients. For an amplitude of 2 cm and a respiratory frequency of 15 bpm, the Dice’s coefficient is 0.84 with RP algorithm and 0.85 with the IR method. Conclusions For pulmonary lesions, the two retrieval methods of RP or IR contours appear similar. In the field of radiotherapy, a new way to use the replanning algorithm seems possible and allows for the simplification of the recovery of the PET contours. A clinical study is in the process of being evaluated. The objective is to validate this method in lung lesions cases.

The Mobius AIRO mobile CT for image-guided proton therapy: Characterization & commissioning.

PurposeThe purpose of this study was to characterize the Mobius AIRO Mobile CT System for localization and image‐guided proton therapy. This is the first known application of the AIRO for proton therapy.MethodsFive CT images of a Catphan®504 phantom were acquired on the AIRO Mobile CT System, Varian EDGE radiosurgery system cone beam CT (CBCT), Philips Brilliance Big Bore 16 slice CT simulator, and Siemens SOMATOM Definition AS 20 slice CT simulator. DoseLAB software v.6.6 was utilized for image quality analysis. Modulation transfer function, scaling discrepancy, geometric distortion, spatial resolution, overall uniformity, minimum uniformity, contrast, high CNR, and maximum HU deviation were acquired. Low CNR was acquired manually using the CTP515 module. Localization accuracy and CT Dose Index were measured and compared to reported values on each imaging device. For treatment delivery systems (Edge and Mevion), the localization accuracy of the 3D imaging systems were compared to 2D imaging systems on each system.ResultsThe AIRO spatial resolution was 0.21 lp mm−1 compared with 0.40 lp mm−1 for the Philips CT Simulator, 0.37 lp mm−1 for the Edge CBCT, and 0.35 lp mm−1 for the Siemens CT Simulator. AIRO/Siemens and AIRO/Philips differences exceeded 100% for scaling discrepancy (191.2% and 145.8%). The AIRO exhibited higher dose (>27 mGy) than the Philips CT Simulator. Localization accuracy (based on the MIMI phantom) was 0.6° and 0.5 mm. Localization accuracy (based on Stereophan) demonstrated maximum AIRO‐kV/kV shift differences of 0.1 mm in the x‐direction, 0.1 mm in the y‐direction, and 0.2 mm in the z‐direction.ConclusionsThe localization accuracy of AIRO was determined to be within 0.6° and 0.5 mm despite its slightly lower image quality overall compared to other CT imaging systems at our institution. Based on our study, the Mobile AIRO CT system can be utilized accurately and reliably for image‐guided proton therapy.

P2.05-042 Development of Thoracic Magnetic Resonance Imaging (MRI) for Radiotherapy Planning: Topic: RT Techniques

The ability to accurately visualize and delineate tumor and surrounding normal tissue is an essential component of radical radiotherapy treatment planning and in non-thoracic sites has been improved by integrating MRI. This early work investigated and optimized different MRI sequences for potential use in thoracic radiotherapy planning. 15 patients with primary lung cancer were scanned using a 1.5 Tesla scanner (Magnetom Aera; Siemens) and radiotherapy planning scanner (Philips Brilliance CT Big Bore, Philips Medical Systems). An identical patient immobilisation board was used (Extended Wing Board; Oncology Systems Limited) for both scans. Multiple MRI sequences were investigated and optimised to give similarity to contrast CT scans (3- 4 mm slice thickness, whole thorax coverage and axial imaging). After reviewing the entire anatomic structure, ability to visualize the primary tumor, lungs, heart and oesophagus were scored using a 5-point system (1, unacceptable; 2, poor; 3, acceptable; 4 good; 5 excellent) and compared to CT (two tailed t-test). Respiratory triggered T2w SPACE (n=12) and T2w TSE (n=3) sequences suggest improved visualization of primary tumor (mean score 4.2 and 4.0) in comparison to CT (3.9), p> 0.05. T2w TSE, T1w TSE (n=3) and T1w 2-point DIXON (n=5, breath hold) sequences may enhance oesophageal (4.3, 3.7 and 3.6) and cardiac (3.3, 4 and 3.4) visualisation, compared with CT (3.0), p>0.05). CT (n=15) was the optimal imaging modality for viewing lungs (p< 0.05). T1w Cartesian VIBE (n=12, breath hold) provided no clear benefit over CT and diffusion-weighted single-shot-planar images (n=8) remain problematic due to image distortion. This preliminary study demonstrates the potential for MRI to improve the visualization of thoracic primary tumors, oesophagus and cardiac anatomy, all of which can be challenging to see on CT imaging, particularly in patients with collapse, consolidation or mediastinal tumor invasion.

Evaluation of positioning uncertainty for frameless radiosurgery for three IGRT methods with a head phantom

Introduction A BRAINLAB system for frameless radiosurgery is commissioned obtaining patient positioning accuracy with the STEEV head phantom (CIRS). Purpose Firstly the position accuracy for frameless radiosurgery has been compared with three different IGRT methods: EXACTRAC (BRANLAB), OBI-2D (VARIAN) and CBCT (VARIAN) with EXACTRAC registration. Secondly the uncertainty of image registration between MRI (PHILIPS INGENIA SYSTEMS 3.0T) and CT (PHILIPS BIG BORE) has been evaluated with IPLANRT image 4.1.1 software. Materials and methods Phantom positioning was done with automatic registration and with equivalent X-ray parameters for the three IGRT methods. Measurement was done 5 times, every of these the phantom with mask was replaced. Position accuracy was evaluated with the tool “detect Winston-Lutz pointer” from EXACTRAC. CT and MRI images of a multimodality fillable sphere insert in the phantom were registered. The registration was evaluated with IPLANRT image software finding the center of the sphere in each modality. Results For the three IGRT techniques the accuracy obtained was (0.466 ± 0.071)mm for EXACTRAC, (0.744 ± 0.209)mm for OBI-2D and (0.750 ± 0.386)mm for CBCT registered with EXACTRAC. For MRI was obtained (0.300 ± 0.1)mm. The total uncertainty for EXACTRACT IGRT system is (0.554 ± 0.374)mm (k = 2). Conclusions • The positioning system EXACTRAC have less uncertainty than both OBI-2D and CBCT registered with EXACTRAC for cranial SRS. • The total uncertainty for the EXACTRAC including IGRT and also image registration between CT and MRI is (0.554 ± 0.374)mm.

Selection of high-rate gas well completion designs applying multi-criteria decision making and hierarchy methods

Selecting the appropriate well completion design, equipment and installation techniques are frequent decisions required of oil and gas operators. Making selections from alternative completion designs can lead to enhanced well productivity, lower capital costs, less completion time and lower intervention/maintenance costs. However, typically there is a tradeoff of these benefits; with certain designs leading to lower costs and others leading to higher productivity or less risk of operational risks and costs. The decision often becomes a strategic one; imposing identified priorities of the operator upon the technical performance and cost assumptions made for each alternative. Multi-criteria decision-making methods (MCDM) offer a means of evaluating such decisions in a consistent and systematic way taking into account available information, assumptions and strategic preferences. MCDM is applied to three high-rate-gas-well-completion alternatives, i.e., mono bore (MB), big bore (BB) and optimized big bore (OBB), in order to aid the selection among them given certain strategic priorities. The MCDM techniques applied to gas-well-completion decisions are analytic hierarchy process (AHP) and technique for order of preference by similarity to ideal solution (TOPSIS). The criteria considered by these techniques include: time to run the completion, capital and operating costs involved, gas production rate achieved, and risk exposure associated with each well completion technique. The MCDM applied are particularly effective at providing insight to decisions when sensitivity analysis is conducted based upon different weights (i.e., strategic priorities) applied to the key technical attributes of the well completion alternatives.

SU‐G‐IeP1‐06: Estimating Relative Tissue Density From Quantitative MR Images: A Novel Perspective for MRI‐Only Heterogeneity Corrected Dose Calculation

Purpose:To explore the feasibility of extracting the relative density from quantitative MRI measurements as well as estimate a correlation between the extracted measures and CT Hounsfield units.Methods:MRI has the ability to separate water and fat signals, producing two separate images for each component. By performing appropriate corrections on the separated images, quantitative measurement of water and fat mass density can be estimated. This work aims to test this hypothesis on 1.5T.Peanut oil was used as fat‐representative, while agar as water‐representative. Gadolinium Chloride III and Sodium Chloride were added to the agar solution to adjust the relaxation times and the medium conductivity, respectively. Peanut oil was added to the agar solution with different percentages: 0%, 3%, 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% and 100%. The phantom was scanned on 1.5T GE Optima 450W with the body coil using a multigradient echo sequences. Water/fat separation were performed while correcting for main field (B0) inhomogeneity and T2* relaxation time. B1+ inhomogeneities were ignored. The phantom was subsequently scanned on a Philips Brilliance CT Big Bore. MR‐corrected fat signal from all vials were normalized to 100% fat signal. CT Hounsfield values were then compared to those obtained from the normalized MR‐corrected fat values as well as to the phantom for validation.Results:Good agreement were found between CT HU and the MR‐extracted fat values (R2 = 0.98). CT HU also showed excellent agreement with the prepared fat fractions (R2=0.99). Vials with 70%, 80%, and 90% fat percentages showed inhomogeneous distributions, however their results were included for completion.Conclusion:Quantitative MRI water/fat imaging can be potentially used to extract the relative tissue density. Further in‐vivo validation are required.

SU‐F‐J‐185: Detection of Low Contrast Objects: Optimization of CT Simulation Reconstruction Protocol Parameters

Purpose:CT simulation has become an integral component of modern RT planning and therefore must be continually optimized. We evaluated the detection of small and low‐contrast regions in images obtained during CT simulation.Methods:A CT phantom containing a contrast detail modulus for detection of low‐contrast structures was used to optimize the CT reconstruction protocol for abdomen. The parameters (A) Pitch, (B) Reconstruction Filter, and (C) Rotation Time type were varied for assessment of image quality.Three factors, three levels, and nine experiments were identified. According to the Taguchi approach an L9 orthogonal array was selected. The reconstruction parameters of the CT scanner (Brilliance Big Bore), Pitch, Reconstruction Filter type, and Rotation Time, were iteratively scanned according to the orthogonal array. A Catphan 604 CT phantom was used to characterize low‐contrast resolution (CPT730 module). All CT scan images were analyzed by IMAGE‐OWL software. The objective of the study was to identify parameters that maximize the low‐contrast resolution of the images. The ANOVA and F‐tests were used to analyze results using JMP statistical software.Results:The optimal settings and predicted optimal values for low‐contrast resolution were determined. The ANOVA was used to determine the optimum combination of process parameters more accurately by investigating the relative importance of each process parameter. We determined that Pitch (61.4%) had the most significant influence on low contrast resolution, followed by the Reconstruction Filter type (29.7%). The optimal setting level is A1‐B1‐C3, 0.68 pitch, smooth filter, and gantry rotation time 1.5sec., respectively. Additional measurements were made to confirm the prediction error model is justified and the results are validated.Conclusion:These experiments have several implications for CT imaging, especially for clinical detection of small, low‐contrast lesions in liver or pancreas. In the phantom model of this study, optimal Contrast Detail Values were determined to be: 1% contrast, 2mm; 0.5% contrast, 4mm; 0.3% contrast, 7mm.

A Dirichlet process mixture model for automatic18F-FDG PET image segmentation: Validation study on phantoms and on lung and esophageal lesions

The aim of this study was to implement a Dirichlet process mixture (DPM) model for automatic tumor edge identification on (18)F-fluorodeoxyglucose positron emission tomography ((18)F-FDG PET) images by optimizing the parameters on which the algorithm depends, to validate it experimentally, and to test its robustness. The DPM model belongs to the class of the Bayesian nonparametric models and uses the Dirichlet process prior for flexible nonparametric mixture modeling, without any preliminary choice of the number of mixture components. The DPM algorithm implemented in the statistical software package R was used in this work. The contouring accuracy was evaluated on several image data sets: on an IEC phantom (spherical inserts with diameter in the range 10-37 mm) acquired by a Philips Gemini Big Bore PET-CT scanner, using 9 different target-to-background ratios (TBRs) from 2.5 to 70; on a digital phantom simulating spherical/uniform lesions and tumors, irregular in shape and activity; and on 20 clinical cases (10 lung and 10 esophageal cancer patients). The influence of the DPM parameters on contour generation was studied in two steps. In the first one, only the IEC spheres having diameters of 22 and 37 mm and a sphere of the digital phantom (41.6 mm diameter) were studied by varying the main parameters until the diameter of the spheres was obtained within 0.2% of the true value. In the second step, the results obtained for this training set were applied to the entire data set to determine DPM based volumes of all available lesions. These volumes were compared to those obtained by applying already known algorithms (Gaussian mixture model and gradient-based) and to true values, when available. Only one parameter was found able to significantly influence segmentation accuracy (ANOVA test). This parameter was linearly connected to the uptake variance of the tested region of interest (ROI). In the first step of the study, a calibration curve was determined to automatically generate the optimal parameter from the variance of the ROI. This "calibration curve" was then applied to contour the whole data set. The accuracy (mean discrepancy between DPM model-based contours and reference contours) of volume estimation was below (1 ± 7)% on the whole data set (1 SD). The overlap between true and automatically segmented contours, measured by the Dice similarity coefficient, was 0.93 with a SD of 0.03. The proposed DPM model was able to accurately reproduce known volumes of FDG concentration, with high overlap between segmented and true volumes. For all the analyzed inserts of the IEC phantom, the algorithm proved to be robust to variations in radius and in TBR. The main advantage of this algorithm was that no setting of DPM parameters was required in advance, since the proper setting of the only parameter that could significantly influence the segmentation results was automatically related to the uptake variance of the chosen ROI. Furthermore, the algorithm did not need any preliminary choice of the optimum number of classes to describe the ROIs within PET images and no assumption about the shape of the lesion and the uptake heterogeneity of the tracer was required.

A measurement-based X-ray source model characterization for CT dosimetry computations.

The purpose of this study was to show that the nominal peak tube voltage potential (kVp) and measured half‐value layer (HVL) can be used to generate energy spectra and fluence profiles for characterizing a computed tomography (CT) X‐ray source, and to validate the source model and an in‐house kV X‐ray dose computation algorithm (kVDoseCalc) for computing machine‐ and patient‐specific CT dose. Spatial variation of the X‐ray source spectra of a Philips Brilliance and a GE Optima Big Bore CT scanner were found by measuring the HVL along the direction of the internal bow‐tie filter axes. Third‐party software, Spektr, and the nominal kVp settings were used to generate the energy spectra. Beam fluence was calculated by dividing the integral product of the spectra and the in‐air NIST mass‐energy attenuation coefficients by in‐air dose measurements along the filter axis. The authors found the optimal number of photons to seed in kVDoseCalc to achieve dose convergence. The Philips Brilliance beams were modeled for 90, 120, and 140 kVp tube settings. The GE Optima beams were modeled for 80, 100, 120, and 140 kVp tube settings. Relative doses measured using a Capintec Farmer‐type ionization chamber (0.65 cc) placed in a cylindrical polymethyl methacrylate (PMMA) phantom and irradiated by the Philips Brilliance, were compared to those computed with kVDoseCalc. Relative doses in an anthropomorphic thorax phantom (E2E SBRT Phantom) irradiated by the GE Optima were measured using a (0.015 cc) PTW Freiburg ionization chamber and compared to computations from kVDoseCalc. The number of photons required to reduce the average statistical uncertainty in dose to <0.3% was 2×105. The average percent difference between calculation and measurement over all 12 PMMA phantom positions was found to be 1.44%, 1.47%, and 1.41% for 90, 120, and 140 kVp, respectively. The maximum percent difference between calculation and measurement for all energies, measurement positions, and phantoms was less than 3.50%. Thirty‐five out of a total of 36 simulation conditions were within the experimental uncertainties associated with measurement reproducibility and chamber volume effects for the PMMA phantom. The agreement between calculation and measurement was within experimental uncertainty for 19 out of 20 simulation conditions at five points of interest in the anthropomorphic thorax phantom for the four beam energies modeled. The source model and characterization technique based on HVL measurements and nominal kVp can be used to accurately compute CT dose. This accuracy provides experimental validation of kVDoseCalc for computing CT dose.PACS numbers: 87.57.Q‐, 87.57.uq, 87.10.Rt

Helical 4D CT pitch management for the Brilliance CT Big Bore in clinical practice.

In external beam radiotherapy treatment planning for patients with thoracic malignancies, respiratory‐correlated CT (4D CT) is used to obtain high quality studies in the presence of respiratory motion. When helical 4D CT scans are acquired with a Brilliance CT Big Bore, the pitch must meet two conditions. It must be low enough to avoid motion artifacts, and high enough to cover the entire scan length within 120 s to prevent overheating of the X‐ray tube. We developed a nomogram that can be used to obtain a suitable pitch satisfying both requirements. We also assessed the effects on the image quality of a pitch that exceeds the maximum pitch, and of a field of view (FOV) reduction. It was shown that, for AV G and MIP reconstructions, the manufacturer's maximum pitch equation yields an underestimation due to its FOV term.PACS number: 87.57.Q‐, 87.57.cp