The article aims to determine if a prospective acquisition algorithm can be used to find the ideal set of free-breathing phases for fast-helical model-based 4D-CT. A retrospective five-patient dataset that consisted of 25 repeated free breathing CT scans per patient was used. The sum of the square root amplitude difference between all the breathing phases was defined as an objective function to determine the optimality of sets of breathing phases. The objective function was intended to determine if a specific set of breathing phases would yield a motion model that could accurately predict the motion in all 25 CT scans. Voxel specific motion models were calculated using all combinations of N scans from 25 breathing trajectories, (3  ⩽  N  ⩽  25), and the minimum number of scans required to absolutely characterize the motion model was analyzed. This analysis suggests that the number of scans could potentially be reduced to as few as five scans. When the objective function was large, the resulting motion model provided an excellent approximation to the motion model created using all 25 scans.

Purpose: To model fast-moving heart surface motion as a function of cardiac-phase in order to compensate for the lack of cardiac-gating in evaluating accurate dose to coronary structures. Methods: Ten subjects were prospectively imaged with a breath-hold, cardiac-gated MRI protocol to determine heart surface motion. Radial and planar views of the heart were resampled into a 3-dimensional volume representing one heartbeat. A multi-resolution optical flow deformable image registration algorithm determined tissue displacement during the cardiac-cycle. The surface of the heart was modeled as a thin membrane comprised of voxels perpendicular to a pencil beam scanning (PBS) beam. The membrane’s out-of-plane spatial displacement was modeled as a harmonic function with Lame’s equations. Model accuracy was assessed with the root mean squared error (RMSE). The model was applied to a cohort of six chest wall irradiation patients with PBS plans generated on phase-sorted 4DCT. Respiratory motion was separated from the cardiac motion with a previously published technique. Volumetric dose painting was simulated and dose accumulated to validate plan robustness (target coverage variation accepted within 2%). Maximum and mean heart surface dose assessed the dosimetric impact of heart and coronary artery motion. Results: Average and maximum heart surface displacements were 2.54±0.35mm and 3.6mm from the end-diastole phase to the end-systole cardiac-phase respectively. An average RMSE of 0.11±0.04 showed the model to be accurate. Observed errors were greatest between the circumflex artery and mitral valve level of the heart anatomy. Heart surface displacements correspond to a 3.6±1.0% and 5.1±2.3% dosimetric impact on the maximum and mean heart surface DVH indicators respectively. Conclusion: Although heart surface motion parallel to beam’s direction was substantial, its maximum dosimetric impact was 5.1±2.3%. Since PBS delivers low doses to coronary structures relative to photon radiotherapy, it is unknown whether this variation would be clinically significant for late effects.

Purpose:The ability of pencil beam scanning (PBS) to deliver highly conformal dose distributions may be affected by patient‐ and physics‐related uncertainties. In clinical practice, selection of proton beam angles is determined qualitatively. This study investigates whether an optimal proton PBS beam angle could be quantitatively determined to ensure robust planning for pelvic targets.Methods:PBS beam angles were optimized based on two independent criteria; shortest and most homogeneous path from the patient surface to the distal edge of the target. The beam angle optimization criteria for gantry angles between 90°‐270° were quantified in 10° increments for each ray, calculated as the straight line distance from the surface of the skin to the CTV's distal edge. The goal was to minimize the path length of a proton PBS beam from the patient surface to the distal edge of the CTV, relative to the entry point, while minimizing HU inhomogeneity along the ray. HU homogeneity (i.e. HU variation) was quantitatively defined as the standard deviation of the average intra‐ray HU intensity distribution of the rays comprising a single beam. This method was validated relative to inter‐fraction changes on ten consecutive, locally advanced, rectal cancer patients, who underwent an average 4 verification CTs. The displacement of the 95–98% isodose lines was determined from forward calculated dose distributions on verification CTs.Results:The posterior beam (180°) had the average shortest path length, 132.7±17.2mm, and the most homogenous path, 31.9±4.3HU. The 95–98% isodose lines from all plans verified our path length to within 2.3±1.2% and HU homogeneity to within 1.2±0.5%.Conclusion:The proposed optimization algorithm determined the posterior beam dose distribution as the most robust relative to inter‐fraction variation for large pelvic targets treated with PBS and was validated via verification CT for our patient cohort. Future work will focus on further algorithm development.

The optimal proton pencil beam scanning (PBS) beam angle was calculated for a population of patients with lower pelvis targets. 10 patients were planned with a single PBS beam in the left lateral, right lateral, and posterior directions. A beam direction was considered to be optimal if it satisfied two metrics: shortest path length and least Hounsfield Unit (HU) variation. To determine these metrics, a ray-trace approach was adopted where the length of a ray represented the path length and the variation of HUs across the ray represented the HU variation. A Kolmogorov-Smirnov test determined the normalcy of the path length and HU variation at the 95% confidence level. Results showed that both the path length and HU variation were normally distributed. The path length was shortest for the posterior beam, and the left lateral beam had the least HU variation. Combining both optimization metrics, the posterior beam was more optimal since i) the path length was much shorter than the lateral beam’s path length, and ii) the HU variation along the posterior beam’s path length was only slightly less homogeneous than the HU variation along the lateral beam’s path length, but statistically similar. This study found that a posterior beam is optimal for lower pelvis targets when a single PBS beam is used.

Purpose:To develop a quantitative decision making metric for automatically detecting irregular breathing using a large patient population that received phase‐sorted 4DCT.Methods:This study employed two patient cohorts. Cohort#1 contained 256 patients who received a phasesorted 4DCT. Cohort#2 contained 86 patients who received three weekly phase‐sorted 4DCT scans. A previously published technique used a single abdominal surrogate to calculate the ratio of extreme inhalation tidal volume to normal inhalation tidal volume, referred to as the κ metric. Since a single surrogate is standard for phase‐sorted 4DCT in radiation oncology clinical practice, tidal volume was not quantified. Without tidal volume, the absolute κ metric could not be determined, so a relative κ (κrel) metric was defined based on the measured surrogate amplitude instead of tidal volume. Receiver operator characteristic (ROC) curves were used to quantitatively determine the optimal cutoff value (jk) and efficiency cutoff value (τk) of κrel to automatically identify irregular breathing that would reduce the image quality of phase‐sorted 4DCT. Discriminatory accuracy (area under the ROC curve) of κrel was calculated by a trapezoidal numeric integration technique.Results:The discriminatory accuracy of ?rel was found to be 0.746. The key values of jk and tk were calculated to be 1.45 and 1.72 respectively. For values of ?rel such that jk≤κrel≤τk, the decision to reacquire the 4DCT would be at the discretion of the physician. This accounted for only 11.9% of the patients in this study. The magnitude of κrel held consistent over 3 weeks for 73% of the patients in cohort#3.Conclusion:The decision making metric, ?rel, was shown to be an accurate classifier of irregular breathing patients in a large patient population. This work provided an automatic quantitative decision making metric to quickly and accurately assess the extent to which irregular breathing is occurring during phase‐sorted 4DCT.

The purpose of this study was to assess the feasibility of proton pencil beam scanning (PBS) for the treatment of mediastinal lymphoma. A group of 7 patients of varying tumor size (100–800 cc) were planned using a PBS anterior field. We investigated 17 fractions of 1.8 Gy(RBE) to deliver 30.6 Gy(RBE) to the internal target volume (ITV). Spots with σ ranging from 4 mm to 8 mm were used for all patients, while larger spots (σ = 6–16 mm) were employed for patients with motion perpendicular to the beam (⩾5 mm), based on initial 4-dimensional computed tomography (4D CT) motion evaluation. We considered volumetric repainting such that the same field would be delivered twice in each fraction. The ratio of extreme inhalation amplitude and regular tidal inhalation amplitude (free-breathing variability) was quantified as an indicator of potential irregular breathing during the scanning. Four-dimensional dose was calculated on the 4D CT scans based on the respiratory trace and beam delivery sequence, implemented by partitioning the spots into separate plans on each 4D CT phase. Four starting phases (end of inhalation, end of exhalation, middle of inhalation and middle of exhalation) were sampled for each painting and 4 energy switching times (0.5 s, 1 s, 3 s and 5 s) were tested, which resulted in 896 dose distributions for the analyzed cohort. Plan robustness was measured for the target and critical structures in terms of the percent difference between ‘delivered’ dose (4D-evaluated) and planned dose (calculated on average CT). It was found that none of the patients exhibited highly variable or chaotic breathing patterns. For all patients, the ITV D98% was degraded by <2% (standard deviations ∼ 0.1%) when averaged over the whole treatment course. For six out of seven patients, the average degradation of ITV D98% per fraction was within 5% . For one patient with motion perpendicular to the beam (⩾5 mm), the degradation of ITV D98% per fraction was up to 15%, which was mitigated to 2% by employing larger spots and repainting. Deviation of mean lung dose was at most 0.2 Gy(RBE) (less than 1% of prescribed dose, 30.6 Gy(RBE)), while the deviation of heart maximum dose and cord maximum dose could exceed 5% of the prescribed dose. No significant difference in either target coverage or normal tissue dose was observed for different energy switching times compared via two-sided Wilcoxon signed-rank tests (p < 0.05). This feasibility study demonstrates that, for mediastinal lymphoma, the impact of the interplay effect on the PBS plan robustness is minimal when volumetric repainting and/or larger spots are employed.

Purpose: To investigate, in a treatment plan design and robustness study, whether proton pencil beam scanning (PBS) has the potential to offer advantages, relative to inter-fraction uncertainties, over photon volumetric modulated arc therapy (VMAT) in a locally advanced rectal cancer patient population.Methods and Materials: Ten patients received a planning CT scan, followed by an average of 4 weekly offline CT verification CT scans, which were rigidly co-registered to the planning CT. Clinical PBS plans were generated on the planning CT, using a single-field uniform-dose technique with single-posterior and parallel-opposed (LAT) fields geometries. The VMAT plans were generated on the planning CT using 2 6-MV, 220 degrees coplanar arcs. Clinical plans were forward-calculated on verification CTs to assess robustness relative to anatomic changes. Setup errors were assessed by forward-calculating clinical plans with a +/- 5-mm (left-right, anterior-posterior, superior-inferior) isocenter shift on the planning CT. Differences in clinical target volume and organ at risk dose-volume histogram (DHV) indicators between plans were tested for significance using an appropriate Wilcoxon test (P<.05).Results: Dosimetrically, PBS plans were statistically different from VMAT plans, showing greater organ at risk sparing. However, the bladder was statistically identical among LAT and VMAT plans. The clinical target volume coverage was statistically identical among all plans. The robustness test found that all DVH indicators for PBS and VMAT plans were robust, except the LAT's genitalia (V5, V35). The verification CT plans showed that all DVH indicators were robust.Conclusions: Pencil beam scanning plans were found to be as robust as VMAT plans relative to interfractional changes during treatment when posterior beam angles and appropriate range margins are used. Pencil beam scanning dosimetric gains in the bowel (V15, V20) over VMAT suggest that using PBS to treat rectal cancer may reduce radiation treatment-related toxicity. (C) 2016 Elsevier Inc. All rights reserved.

The dosimetric impact of accurately delineating the left anterior descending artery (LAD) was investigated for routine treatment planning deep inspiration breath-hold (DIBH) CT where cardiac motion was not accounted for. The LAD was contoured in the routine DIBH CT images by an expert radiologist in a population of 10 patients with cancer of the left breast. The motion blurring of the LAD in the DIBH CT images was extracted from the contoured LAD to create a corrected LAD volume. This was used to compare the maximum and mean LAD dose over 3D conformal radiotherapy (3DCRT), uniform scattering (US) proton, and pencil beam scanning (PBS) proton plans. Using a corrected LAD volume reduced the maximum dose LAD DVH indicator by 2% (3DCRT), 4% (US), and 25% (PBS). A corrected LAD volume increased the mean dose LAD DVH indicator by 25% (3DCRT), 61% (US), and 35% (PBS). In terms of absolute dose the impact of contouring on LAD is higher for photon therapy due to the higher doses delivered using this modality. Overall, the results demonstrate that the LAD volume could potentially be the source of inconsistencies in correlation between dose and radiation-induced cardiac toxicity when uncompensated motion is present in the treatment planning images.

In a previous work, a complete model has been presented for the thermal property characterization of suspended wires using the 3 omega technique. As validation of the model and comparison of the effects of measurement configuration, several samples are measured in a vacuum chamber (<0.001 Pa) using voltage and current sources for sample excitation. In particular, the characteristics of each excitation type are shown along with the effect of cancellation of the first harmonic (1 omega) signal. Without cancellation, large uncertainties occur in thermal conductivity and thermal diffusivity measurement while volumetric heat capacity cannot be directly measured with acceptable accuracy. Using methods to cancel the 1w voltage greatly improves measurement precision with the ability to independently measure the three aforementioned thermal properties. Voltage source excitation results in the most precise measurements. However, for samples with resistance (<similar to 0.5 k Omega), susceptible to the influence of other components of the circuit, an extrapolation procedure using measurements of varied Wheatstone bridge resistances is necessary to obtain thermal conductivity and heat capacity. Detailed uncertainty analysis of each measurement configuration shows that the voltage source provides the best overall measurement uncertainties (similar to 8.5% and 3.9% for thermal conductivity and diffusivity respectively). (C) 2014 Elsevier Ltd. All rights reserved.

Purpose:To assess the feasibility of routine treatment planning 4DCT and deep inspiration breath‐hold (DIBH) to accurately contour the left anterior descending artery (LAD), a primary indicator of cardiac toxicity, for radiotherapy treatment planning of breast cancer.Methods:Ten subjects were imaged with a cardiac‐gated MRI protocol to determine the displacement of a ROI that included the LAD. The subjects performed a series of breath‐hold maneuvers to obtain short‐axis and radial views, which were resampled to create a 3D‐volume. Tissue motion was determined using a multi‐resolution 3D optical flow deformable image registration algorithm. The ROI motion was then used as a spatial boundary to characterize the blurring motion of the LAD in ten patients during clinical 4DCT and DIBH protocols. A radiologist contoured the LAD. Coronary motion‐induced blurring artifacts were quantified by applying an unsharp filter to accentuate the LAD despite motion‐blurring. The 4DCT maximum inhalation and exhalation respiratory phases were co‐registered to determine the LAD displacement during tidal respiration, as visualized in 4DCT.Results:The average 90th percentile heart motion for the ROI was 0.7±0.1mm(LR), 1.3±0.6mm(SI), 0.6±0.2mm(AP) in the cardiac‐gated MRI cohort. The average relative increase in the number of voxels comprising the LAD contour was 69.4±4.5% for the DIBH. During tidal respiration, the average relative increase in the LAD contour was 69.3±5.9% and 67.9±4.6% for inhalation and exhalation respiratory phases respectively. The average 90th percentile LAD motion was 4.8±1.1mm(LR), 0.9±0.4mm(SI), 1.9±0.6mm(AP) for the 4DCT cohort, in the absence of cardiac‐gating.Conclusion:Uncompensated coronary motion was the dominant form of motion blurring present in the CT images due to the high frequency of the cardiac cycle relative to the respiratory cycle. The 4D and DIBH CT contour delineation of the LAD was consistently overestimated without cardiac‐gating, which could have led to inaccurate dose volume histogram indicators in clinical practice.

PURPOSE To improve the quality of mega-voltage orthogonal scout images (MV topograms) for a fast and low-dose alternative technique for patient localization on the TomoTherapy HiART system. METHODS Digitally reconstructed radiographs (DRR) of anthropomorphic head and pelvis phantoms were synthesized from kVCT under TomoTherapy geometry (kV-DRR). Lateral (LAT) and anterior-posterior (AP) aligned topograms were acquired with couch speeds of 1cm/s, 2cm/s, and 3cm/s. The phantoms were rigidly translated in all spatial directions with known offsets in increments of 5mm, 10mm, and 15mm to simulate daily positioning errors. The contrast of the MV topograms was automatically adjusted based on the image intensity characteristics. A low-pass fast Fourier transform filter removed high-frequency noise and a Weiner filter reduced stochastic noise caused by scattered radiation to the detector array. An intensity-based image registration algorithm was used to register the MV topograms to a corresponding kV-DRR by minimizing the mean square error between corresponding pixel intensities. The registration accuracy was assessed by comparing the normalized cross correlation coefficients (NCC) between the registered topograms and the kV-DRR. The applied phantom offsets were determined by registering the MV topograms with the kV-DRR and recovering the spatial translation of the MV topograms. RESULTS The automatic registration technique provided millimeter accuracy and was robust for the deformed MV topograms for three tested couch speeds. The lowest average NCC for all AP and LAT MV topograms was 0.96 for the head phantom and 0.93 for the pelvis phantom. The offsets were recovered to within 1.6mm and 6.5mm for the processed and the original MV topograms respectively. CONCLUSION Automatic registration of the processed MV topograms to a corresponding kV-DRR recovered simulated daily positioning errors that were accurate to the order of a millimeter. These results suggest the clinical use of MV topograms as a promising alternative to MVCT patient alignment.

The thermal conductivity and diffusivity of the dragline silk of the Nephila (N.) clavipes spider has been characterized by one research group to be 151-416 W m(-1) K-1 and 6.4-12.3 x 10(-5) m(2) s(-1), respectively, for samples with low to high strains (zero to 19.7%). Thermal diffusivity of the dragline silk of a different spider species, Araneus diadematus, has been determined by another research group as 2 x 10(-7) m(2)s(-1) for un-stretched silk. To improve measurement reliability and repeatability and resolve the orders of magnitude discrepancy between the two different measurements, this paper measured 13 un-stretched dragline silk samples of the N. clavipes spider with different lengths using the same electrothermal technique as the first group but with a much higher vacuum level and an improved heat transfer model. The measured thermal conductivity is 1.2 W m(-1) K-1 and thermal diffusivity is 6 x 10(-7) m(2) s(-1). The measured thermal diffusivity of the N. clavipes spider silk is in the same order of magnitude as that of the diadematus spider but is 1/100-1/200 of the value by the first group. The measured thermal conductivity is 1/150-1/400 of that measured in literature. The discrepancy between this research and the first group may reside in the vacuum level and the improved heat transfer analysis. The difference in thermal diffusivity measurement between the current research and the results of the second group may be because of different species. (C) 2014 Elsevier Ltd. All rights reserved.

Orthogonal scout images (topograms) can be a fast and accurate alternative to existing megavoltage computed tomography (MVCT) for patient alignment on TomoTherapy HiART system. Patient localization with the on-board MVCT delivers non-negligible dose and has poor temporal resolution compared to kilovoltage computed tomography (kVCT). A potential limiting factor in translating MV topograms into clinical use has been the low quality of the images. The goal of this study was to improve the image quality of MV topograms for better patient localization. Digitally reconstructed radiographs (DRR) of anthropomorphic head and pelvis phantoms were synthesized from kVCT under Tomotherapy geometry. Lateral (LAT) and anterior-posterior (AP) aligned topograms were acquired with a couch speed of 1 cm/s. The phantoms were rigidly translated in all spatial directions with known offsets in increments of 5 mm, 10 mm, and 15 mm to simulate daily positioning errors. The contrast of the MV topograms was automatically adjusted based on the image intensity characteristics. A low-pass frequency filter in Fourier space and a Weiner filter were implemented to reduce the stochastic noise caused by scattered radiation to the detector array. An intensity-based image registration algorithm was used to register the MV topograms to a corresponding kVDRR by minimizing the mean square error between corresponding pixel intensity. The registration accuracy was assessed by comparing the normalized cross correlation coefficients (NCC) between the registered topograms and the kVDRR. The applied phantom offsets were determined by registering the MV topograms with the kVDRR and recovering the spatial translation of the MV topograms. The NCC coefficients for the head phantom were 0.96±0.01 (AP) and 0.95±0.01 (LAT) for filtered topogram registration and 0.76±0.01 (AP) and 0.86±0.01 (LAT) for unfiltered topogram registration. The NCC coefficients for the pelvis phantom were 0.94±0.02 (AP) and 0.95±0.01 (LAT) for filtered topogram registration and 0.89±0.01 (AP) and 0.74±0.01 (LAT) for unfiltered topogram registration. The automatic registration technique provided sub-millimeter accuracy. The results can be seen in Table. The automatic registration of the filtered MV topograms to a corresponding kVDRR and the recovered simulated daily positioning errors were accurate to the order of a millimeter. These results suggest the clinical use of MV topograms as a promising alternative to MVCT patient alignment.Scientific Abstract 3607; TableAccuracy of recovered daily patient positing shifts using MV topogramsAP (mm)RL (mm)SI (mm)Head0.69±0.930.73±0.690.67±0.83 (AP Topo)1.33±1.11 (LAT Topo)Pelvis0.44±0.500.46±0.540.69±0.75 (AP Topo)0.83±0.77 (LAT Topo) Open table in a new tab

The transient electrothermal technique is a powerful tool to obtain thermal properties of fine fibers. However, the technique suffers from several inherent pitfalls, which affect measurement accuracy, especially with application to coated, nonconductive samples. In this paper, measurement challenges are described and quantified for several associated parameters and physics including: sample length, time of Joule-heating initiation, sample resistance including measurement uncertainty as well as evolving resistance for coated samples, coating influence, lateral surface heat losses, vacuum level, and variable heat generation. Several methods to overcome these challenges to ensure good measurement accuracy are provided. These methods are applied to the measurement of thermal conductivity and thermal diffusivity of gold-coated glass fibers (nonconductor). The resulting measured thermal conductivity of 1.35 Wm(-1) K-1, and thermal diffusivity of 7.6 x 10(-7) m(2) s(-1) compare well to literature values. Additionally, an analytic formula is developed along with limiting conditions for simplified application, which accounts for neglected heat losses. The result is a factor that can be applied to correct a more straightforward heat model of the sample, which neglects heat losses. To further validate the method and quantify measurement variability, a detailed uncertainty analysis is performed using methods based on the Taylor series method for propagation of uncertainty and Monte Carlo simulation. The resulting measurement uncertainty is found to be similar to 7% for thermal conductivity and similar to 4% for thermal diffusivity.

The 3 omega method is one of few reliable measurement techniques for thermal characterization of micro to nanoscale suspended wires or fibers and has been applied for measurements of carbon nanotubes and silicon nanowires. However, the models described in the past were either complicated for analysis or simplified from a more complete solution. In addition, the past models cannot be implemented directly when using a more reliable measurement configuration with a Wheatstone bridge. In this work, a simpler, explicit model, is developed to describe the heat transfer process through a suspended wire for measurement of its thermal properties, Generic trends and values of the 3 omega harmonic voltage amplitude and phase responses clearly indicate the frequency limits for thermal conductivity and heat capacity determination and ideal conditions for thermal diffusivity estimation. Based on a sensitivity analysis, these limits are confirmed and appropriate frequency ranges for thermal conductivity and diffusivity are recommended. Radiation influence on the measurement results is quantified and correlated to a dimensionless radiation parameter. Two methods are presented to determine sample thermal properties independent of lateral heat losses and validated by numerical experiments using COMSOL. Uncertainty analysis was also derived by Taylor series expansion with calculated parameter sensitivities. (C) 2014 Elsevier Ltd. All rights reserved.

The transient electrothermal technique has been shown to be an important method for the thermal characterization of fine fibers. However, this study shows that the accuracy of this method hinges on the proper consideration of several sample parameters including the initial electrical resistance and the emissivity of the sample. Such consideration is verified through measurement of platinum samples with various lengths. A multi-level heating scheme is proposed to mitigate the high sensitivity to the initial sample electrical resistance resulting in a demonstrated precision uncertainty of \({<}3~\%\). Using a recently expanded thermal model accounting for heretofore neglected heat losses, the emissivity of the sample may be measured simultaneously with thermal conductivity. The influence of the vacuum level is demonstrated by the results of measurements at two vacuum levels \({<}0.001\hbox { Pa}\) (diaphragm plus turbo pumps) and \({\sim } 2\hbox { Pa}\) (regular rotary vane pump). Neglecting these considerations can produce seemingly viable measurement results having significant bias error. Based on the characteristics of the expanded model, a new approach is presented that takes advantage of the simplicity of the reduced model, which neglects lateral heat losses from the sample, to obtain sample thermal properties that are independent of heat loss effects. Lastly, several experimental results validate the new model.

Purpose: To improve image quality in low‐dose 4D‐CT using a selective averaging algorithm that combines images acquired under free breathing and registered to a single breathing phase. Methods: Five patients were imaged with a low‐pitch helical protocol on a 64‐slice scanner during free patient breathing, as part of an IRB‐approved research protocol. 25 low dose scans were performed in order to image the lungs at varying breathing phases to generate a breathing motion model. The first scan was registered to the subsequent 24 using a b‐spline registration algorithm to produce 25 representations of a single patient scan geometry. These images were averaged using k‐means clustering (k=2 clusters) in conjunction with the arithmetic mean. A gradient mapping algorithm assigned the appropriate mean value to each voxel to produce a composite image. This algorithm calculates the image gradient at each voxel and assigns a cluster mean when the gradient magnitude is above a threshold, otherwise assigning the arithmetic mean. The composite image was compared with the first image in the set as well as the arithmetic mean of all 25 scans. Image noise was calculated in an axial region of the liver. Sharpness was measured as the sum of gradient magnitudes across the image. Results: By combining gradient selective k‐means clustering with the arithmetic mean, image noise was reduced by 78% while maintaining sharpness within 14.5% of the reference image. This improves on the arithmetic mean in sharpness by 14% with less than a 1% increase in noise. Conclusion: Results indicate that 4D‐CT image quality can be improved by co‐registration followed by a gradient selective averaging algorithm. This method greatly reduces image noise while maintaining sharpness that is lost in a simple averaging algorithm. This work supported in part by NIH R01CA096679

Purpose: To develop a new motion modeling technique to account for internal lung tissue displacement due to cardiac motion during free breathing CT scans. Methods: Five patients were imaged 25 successive times under free breathing conditions in alternating directions with a 64‐slice CT scanner using a low dose fast helical protocol. A pneumatic bellows around the abdomen was used to as a breathing surrogate and a 3‐lead ECG monitor simultaneously measured heart rate. The lungs were segmented from each image and deformable registration was used to register the first image to the subsequent 24 segmented images. The registered voxel locations were fitted to a linear motion model, relating the internal lung tissue deformation to the tidal volume v, airflow f and cardiac phase h. The time dependence of cardiac induced tissue displacement was characterized by a periodic function synchronized to the cardiac cycle, optimized to reduce model error, and was patient specific. The magnitude of cardiac induced motion was evaluated by comparing the discrepancies between fitted and measured motion with and without the cardiac term applied. Results: The magnitude of cardiac induced lung tissue displacement was determined to be up to 2.5mm in the lung for regions close to the myocardium. The addition of the cardiac term reduced the total number of voxels with mean errors above 1mm by 33%. Conclusion: Application of the cardiac term in the motion model reduces large errors in motion modeling in regions close to the myocardium. This work supported in part by NIH R01CA096679