Andrew N Primak

Cleveland Clinic, Cleveland, Ohio, United States

Are you Andrew N Primak?

Claim your profile

Publications (91)182.75 Total impact

  • C Liptak, A Morgan, F Dong, A Primak, X Li
    [Show abstract] [Hide abstract]
    ABSTRACT: In Monte Carlo simulations of patient dose from a CT scan, accurate knowledge about the x-ray energy spectra is essential. Simulated CT dose index (CTDI₁₀₀) free-in-air is often used together with measured values to calibrate the intensity of the spectra. To simulate CTDI₁₀₀, a computational model of the ion chamber is required. Various chamber modeling methods have been reported. The purpose of this study was to investigate systematically how chamber modeling affects simulated dose response. Both the pencil chamber and the thimble chamber recently recommended by AAPM TG111 were studied. A Monte Carlo program previously validated for a clinical CT system (SOMATOM Definition Flash, Siemens Healthcare) was used. To examine the effect of chamber modeling, a highly detailed model of an actual CTDI₁₀₀ pencil chamber (model 10×5-3CT, Radcal Corporation) was first defined. Five additional models with reduced detail were then created. They differed from the first model in terms of their dimensions, component parts, material composition, and the use of quadric or voxel geometry. Two models of a TG111 thimble chamber were also created, representing a highly detailed and a rather crude rendition of an actual thimble chamber (model 10×0.6-3CT, Radcal Corporation). Single axial scans for CTDI₁₀₀ chamber models and helical scans for TG111 chamber models were simulated free-in-air at 70 and 120 kVp. Simulated dose responses were then compared amongst different chamber models. For the six models of the CTDI₁₀₀ pencil chamber, the coefficient of variation of the simulated CTDI₁₀₀ was 0.9% at 70 kVp and 1.0% at 120 kVp. For the two models of the TG111 thimble chamber, the difference in the simulated equilibrium-dose-pitch product was 1.9% at 70 kVp and 2.1% at 120 kVp. In Monte Carlo simulations of CT ion chambers, detailed modeling of chamber geometry and material composition is not necessary. This research is supported in part by a Faculty Startup Fund from Cleveland State University.
    Medical Physics 06/2015; 42(6):3747. DOI:10.1118/1.4926317 · 3.01 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Standard CT dosimetry phantoms are currently perfectly circular cylinders. Lately elliptical cylinders have been proposed as an alternative because they better represent the shapes of the human body and allow the evaluation of tube current modulation systems. The purpose of this study was to extend the concept of CTDIw to elliptical phantoms of various aspect ratios. Based on published adult and pediatric data, eight body aspect ratios were chosen from the full range between 1 (perfectly circular) and 1.72 (extremely elliptical). For each value, two elliptical cylinders were created digitally to represent adult and pediatric bodies. They had the same cross-sectional areas as the standard 32-cm and 16-cm CTDI phantoms. For each digital phantom, CTDI₁₀₀ at central and peripheral locations were simulated for tube voltages between 70-140 kVp using a Monte Carlo program previously validated for a clinical CT system (SOMATOM Definition Flash, Siemens Healthcare). The simulation also output the average dose over the cross-sectional area, Dxsec, the quantity CTDIw is intended to represent. Values of Dxsec and CTDI₁₀₀ allowed linear systems of equations to be established, from which central and peripheral weighting coefficients were solved. All elliptical phantoms had the same Dxsec as the standard circular phantoms. Regardless of phantom shape, only two weighting coefficients (w1 and w2) were needed: w1 for central CTDI₁ ₀₀ and w2 for the average of the four peripheral CTDI₁₀₀'s. For perfectly circular phantoms, w1 and w2 were 0.37 and 0.63, respectively, agreeing well with the conventional weighting coefficients of 1/3 and 2/3. Over the full range of aspect ratios, w1 increased linearly from 0.37 to 0.46, whereas w2 decreased linearly from 0.63 to 0.54. CTDI₁₀₀ weighting coefficients are linear functions of phantom aspect ratio, allowing the concept of CTDIw to be easily extended to elliptical phantoms of various shapes. This research is supported in part by a Faculty Startup Fund from Cleveland State University.
    Medical Physics 06/2015; 42(6):3746. DOI:10.1118/1.4926315 · 3.01 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: This review discusses new technological approaches to reduce radiation dose while maintaining image quality in multi-detector-computed tomography (CT) for both adult and pediatric applications. First, the review focuses on the principles of automatic exposure control (AEC) systems for modulation of the tube current according to patient’s size; as well as special AEC adaptations for cardiac CT and organ-based tube current modulation. The selection of the tube potential (kV) is also discussed, with particular emphasis in a new technology which allows an automatic selection of the tube potential with a corresponding adjustment in the tube current, according to patient’s size and diagnostic task. The principles of iterative reconstruction, which is quickly becoming a standard feature in CT scanners, are also presented with particular emphasis on dose reduction and image quality. A full section is devoted to two latest state-of-the-art applications which can be used for radiation dose reduction: virtual non-contrast imaging with dual-energy CT and ultra-low dose CT with added spectral filtration. In the final section, innovations in CT hardware are presented ranging from the X-ray tube to CT detectors, which enable data acquisition at faster speeds and better efficiency to improve the balance between radiation dose and image quality.
    02/2015; 3(2). DOI:10.1007/s40134-014-0084-6
  • Radiologic Clinics of North America 02/2015; 53(3). DOI:10.1016/j.rcl.2014.12.005 · 1.83 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To compare radiation dose surrogates [volume CT dose index (CTDIvol), dose-length product (DLP), size-specific dose estimate (SSDE), and effective dose] and image noise in a cohort of patients undergoing hepatocellular carcinoma screening who underwent both single-energy CT (SECT) and dual-energy CT (DECT).
    Clinical Radiology 10/2014; DOI:10.1016/j.crad.2014.08.021 · 1.66 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Iterative metal artifact reduction (IMAR) is a sinogram inpainting technique that incorporates high-frequency data from standard weighted filtered back projection (WFBP) reconstructions to reduce metal artifact on computed tomography (CT). This study was designed to compare the image quality of IMAR and WFBP in total shoulder arthroplasties (TSA); determine the optimal amount of WFBP high-frequency data needed for IMAR; and compare image quality of the standard 3D technique with that of a faster 2D technique.
    Skeletal Radiology 08/2014; 43(12). DOI:10.1007/s00256-014-1987-2 · 1.74 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: In CT imaging, a desirable quality assurance (QA) dose quantity should account for the dose variability across scan parameters and scanner models. Recently, AAPM TG 111 proposed to use equilibrium dose-pitch product, in place of CT dose index (CTDI100), for scan modes involving table translation. The purpose of this work is to investigate whether this new concept better accounts for the kVp dependence of organ dose than the conventional CTDI concept.
    Medical Physics 06/2014; 41(6):425-425. DOI:10.1118/1.4889158 · 3.01 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose To compare images acquired with 50% tube exposure with a dual-source computed tomographic (CT) scanner and reconstructed with sinogram-affirmed iterative reconstruction (SAFIRE) with 100% exposure images reconstructed with filtered back projection (FBP) for reader ability to detect stones, reader confidence, and findings outside the urinary tract. Materials and Methods In this HIPAA-compliant, institutional review board-approved study, imaging examinations in 99 patients with urolithiasis were assessed. Data from both tubes were reconstructed with FBP; data from the primary tube only were reconstructed with SAFIRE. Seven readers evaluated randomized studies for calculi in nine regions. Reader confidence was scored by using a five-point scale. Ancillary findings were noted. Nonparametric methods for clustered data were used to estimate the area under the receiver operating characteristic curves with 95% confidence intervals to test for noninferiority of 50% exposure with SAFIRE. Results Calculi were found in 113 locations (pyelocalyceal ureter, 86; proximal ureter, seven; midureter, four; distal ureter, 15; bladder, one) and not found in 752 locations. Mean area under the receiver operating characteristic curve for FBP was 0.879 (range, 0.607-0.967) and for SAFIRE, 0.883 (range, 0.646-0.971; 95% confidence interval: -0.025, 0.031). The SAFIRE images were not significantly inferior to FBP images (P = .001). Reader confidence levels for images with stones were similar with FBP and SAFIRE (P = .963). For the 52 patients who had extraurinary findings, readers reported them correctly in 74.4% (271 of 364) and 72.0% (262 of 364) of cases (P = .215) for FBP and SAFIRE, respectively. For the nine patients with potentially important findings per the reference standard, the detection rates were 44% (28 of 63) and 33% (21 of 63, P = .024), respectively. For the 43 patients with unimportant or likely unimportant findings, the false detection rates were 15% (44 of 301) and 14% (43 of 301, P = .756), respectively. Conclusion The 50% tube exposure CT images reconstructed with SAFIRE were not inferior to 100% exposure images reconstructed with FBP for diagnosis of urolithiasis, without decreases in reader confidence. © RSNA, 2014.
    Radiology 05/2014; 272(3):132381. DOI:10.1148/radiol.14132381 · 6.21 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose To measure the effect of reduced radiation exposure on low-contrast low-attenuation liver lesion detection in an anthropomorphic abdominal phantom by using filtered back projection (FBP) and sinogram-affirmed iterative reconstruction. Materials and Methods Eighteen radiologists blinded to phantom and study design interpreted randomized image data sets that contained 36 spherical simulated liver lesions of three sizes and three attenuation differences (5-mm diameter: 12, 18, and 24 HU less than the 90-HU background attenuation of the simulated liver insert; 10- and 15-mm diameter: 6, 12, and 18 HU less than the 90-HU background attenuation) scanned with four discrete exposure settings and reconstructed by using FBP and sinogram-affirmed iterative reconstruction. Response assessment included region-level lesion presence or absence on a five-point diagnostic confidence scale. Statistical evaluation included multireader multicase receiver operating characteristic curve analysis, with nonparametric methods and noninferiority analysis at a margin of -0.10. Results Pooled accuracy at 75% exposure for both FBP and sinogram-affirmed iterative reconstruction was noninferior to 100% exposure (P = .002 and P < .001, respectively). Subsequent exposure reductions resulted in a significant decrease in accuracy. When the smallest (5-mm-diameter) lesions were excluded from analysis, sinogram-affirmed iterative reconstruction was superior to FBP at 100% exposure (P = .011), and sinogram-affirmed iterative reconstruction at 25% and 50% exposure reduction was noninferior to FBP at 100% exposure (P ≤ .013). Reader confidence was greater with sinogram-affirmed iterative reconstruction than with FBP for 10- and 15-mm lesions (2.94 vs 2.76 and 3.62 vs 3.52, respectively). Conclusion In this low-contrast low-attenuation liver lesion model, a 25% exposure reduction maintained noninferior diagnostic accuracy. However, detection was inferior with each subsequent exposure reduction, regardless of reconstruction method. Sinogram-affirmed iterative reconstruction and FBP performed equally well at modest exposure reduction (25%-50%). Readers had higher confidence levels with sinogram-affirmed iterative reconstruction for the 10- and 15-mm lesions. © RSNA, 2014 Online supplemental material is available for this article.
    Radiology 03/2014; 272(1):131928. DOI:10.1148/radiol.14131928 · 6.21 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: PURPOSE Iterative metal artifact reduction (IMAR) is a new sinogram inpainting technique to reduce CT metal artifact which adds high frequency data to improve visualization close to metal edges. Our purpose was to compare the image quality and accuracy of attenuation values near hardware of IMAR and standard filtered back projection (FBP) in patients with shoulder arthroplasties (SA). METHOD AND MATERIALS 8 patients (6 male, avg age 60) with 9 SAs were scanned on a FLASH CT (Siemens) with a standard protocol (140 kVp, 300 eff mAs, 0.6mm collimation, eff pitch 0.35-0.8). Images were reconstructed on a standalone workstation with a smooth kernel (B30) and 0.6mm slice thickness. 3 IMAR reconstructions with different amounts of high frequency data: IMAR (least), IMAR1.5 (more), IMAR2.5 (most) and FBP were ranked for image quality by 5 readers in a side by side comparison from best=1 to worst=4 for bone, soft tissue, metal-bone interface and overall quality. Accuracy of attenuation near hardware was quantified as the absolute difference (AD) between avg HU within a region of interest (ROI) near hardware and for an ROI containing similar tissues on a slice without hardware. RESULTS IMAR1.5 was ranked best for humeral cortex (avg 1.4), glenoid trabeculae (avg 1.36) and glenoid cortex (avg 1.4). IMAR2.5 was ranked best for humeral trabeculae (avg 1.2). IMAR was ranked the best for deltoid muscle (avg 1.2). IMAR1.5 and 2.5 were ranked best for metal-bone interface (avg 1.3). FBP was ranked worst for all structures (avg 3.38 -3.49). All readers ranked IMAR1.5 and 2.5 over FBP (p<0.05) for overall image quality. 3 readers preferred IMAR1.5 and 2.5 over IMAR (p<0.05). There were no significant differences between IMAR1.5 and 2.5 for any reader. Accuracy of attenuation near hardware was significantly better in all 3 IMAR reconstructions than FBP for bone (AD 86–99 vs 430, p<0.006) and soft tissue (AD 23–29 vs 450, p=0.003). There was no significant difference between the IMAR techniques in bone (p=0.4) or soft tissue (p=0.4). CONCLUSION IMAR, especially with added high frequency data, had superior image quality and more accurate attenuation values near hardware than standard FBP in patients with shoulder arthroplasties. CLINICAL RELEVANCE/APPLICATION IMAR is a promising new CT technique to reduce metal artifact that is fully automatic and computationally inexpensive and has the potential to replace standard FBP in patients with hardware.
    Radiological Society of North America 2013 Scientific Assembly and Annual Meeting; 12/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: PURPOSE To assess the effect of reduced radiation exposure and reconstruction method on detection of lesions that are low-contrast, low-attenuation (LCLA) relative to the background liver METHOD AND MATERIALS Semi-anthropomorphic phantom containing custom inserts with 36 spherical liver lesions of 3 sizes and attenuations (10 and 15-mm at 6, 12 and 18HU, and 5-mm at 12, 18 and 24HU below 90HU simulated liver) was scanned at 120kVp, 0.6-mm collimation, 200 (CTDIvol 13.49), 150, 100 and 50mAs on a 128-slice MDCT scanner (Definition Flash, Siemens). Lesions were distributed non-uniformly to reduce memory bias. Images were reconstructed at 3-mm thickness using filtered back projection (FBP) and sinogram-affirmed iterative reconstruction (SAFIRE, S3). A randomized dataset containing 256-images was generated for each reader (12 images with one lesion, 12 with two lesions and 8 without lesions, for each dose and reconstruction method). Eighteen Radiologists blinded to phantom and study design independently reported region-level lesion presence or absence on a 5-point diagnostic confidence scale. Statistical evaluation included multi-reader, multi-case (MRMC) ROC analysis using nonparametric methods with the area under the ROC curve (AUC) considered accuracy. RESULTS Pooled AUC decreased with each 25% reduction from 100% dose: 0.848, 0.842, 0.792 and 0.743 for FBP; and 0.862, 0.855, 0.785 and 0.735 for SAFIRE. At a given dose, improvement in AUC with SAFIRE was, however, not statistically significant. For both FBP and SAFIRE, accuracy at 75% dose was statistically equivalent to 100% dose FBP (p =0.002 and <0.001 respectively). When lesions with the lowest attenuation difference (6HU) were excluded, AUC at 50% SAFIRE (0.800) was not statistically equivalent to 100% FBP (0.846) (p=0.1) although AUC at 50% FBP (0.809) was found equivalent (p=0.029). CONCLUSION In this LCLA liver lesion model, a 25% dose reduction did not reduce detection of the lesions studied. However, detection was inferior with each subsequent dose reduction regardless of reconstruction method. For lesions with attenuation differences larger than or equal to 12HU, lesion detection was not reduced even at 50% dose with FBP. CLINICAL RELEVANCE/APPLICATION Estimates of loss of accuracy at reduced doses and limits of iterative reconstruction should be known especially for low contrast, low attenuation liver lesions to enable dose optimization in practice
    Radiological Society of North America 2013 Scientific Assembly and Annual Meeting; 12/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: We aimed to test the hypothesis that three-dimensional (3D) volume-based scoring of computed tomography (CT) images of the paranasal sinuses was superior to Lund-Mackay CT scoring of disease severity in chronic rhinosinusitis (CRS). We determined correlation between changes in CT scores (using each scoring system) with changes in other measures of disease severity (symptoms, endoscopic scoring, and quality of life) in patients with CRS treated with triamcinolone. The study group comprised 48 adult subjects with CRS. Baseline symptoms and quality of life were assessed. Endoscopy and CT scans were performed. Patients received a single systemic dose of intramuscular triamcinolone and were reevaluated 1 month later. Strengths of the correlations between changes in CT scores and changes in CRS signs and symptoms and quality of life were determined. We observed some variability in degree of improvement for the different symptom, endoscopic, and quality-of-life parameters after treatment. Improvement of parameters was significantly correlated with improvement in CT disease score using both CT scoring methods. However, volumetric CT scoring had greater correlation with these parameters than Lund-Mackay scoring. Volumetric scoring exhibited higher degree of correlation than Lund-Mackay scoring when comparing improvement in CT score with improvement in score for symptoms, endoscopic exam, and quality of life in this group of patients who received beneficial medical treatment for CRS.
    International Forum of Allergy and Rhinology 12/2013; 3(12). DOI:10.1002/alr.21219 · 2.37 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: PURPOSE To assess the effect of CT dose reduction on the detection of urolithiasis. METHOD AND MATERIALS 99 patients with 192 kidneys (6 solitary) were imaged to follow urolithiasis on a dual energy scanner [Definition Flash (Siemens Healthcare)] in dual-source mode using 120 kVp, 128x0.6 collimation & pitch 0.9. Dose modulation used with weight-based reference mAs. Data from both tubes was reconstructed with standard filtered back projection (100% FBP). Data from primary tube (50% total dose) was reconstructed using sonogram-affirmed iterative reconstruction i31 (50% IR). 7 readers (2 senior & 2 junior staff, 2 imaging fellows, 1 urology fellow) evaluated 100% FBP and 50% IR images in a randomized fashion for presence or absence of calculi in 9 regions (pyelocalyceal, proximal, mid, distal ureter, & bladder). Largest axial stone size on magnified bone windows per region was measured and categorized as ≤1,2-3, 4-5, 6-7, ≥8mm. Confidence scored on 5 point scale. Presence or absence of ancillary findings (hydronephrosis, stranding) or alternative diagnosis to explain flank pain was noted. Findings unrelated to history were scored using the CT colonography extracolonic reporting system. Truth was determined by 2 senior uroradiologists in consensus with access to medical record and other imaging. Nonparametric methods for clustered data were used to estimate the ROC curves and their areas for each reader. A 95% CI was constructed for the difference in the mean ROC areas. RESULTS 113 locations had stones & 752 did not (86 pyelocalyceal, 7 proximal, 4 mid, 15 distal ureter). Mean ROC area for FBP was 0.879 (range 0.607-0.967) and 50% IR was 0.883 (0.646-0.971). For one reader, ROC area with 50% IR was significantly better. The p-value for the hypothesis of non-inferiority was 0.001, indicating that 50% dose IR was not inferior. The 95% CI for the difference in ROC areas between 100% FBP vs. 50% IR is [-0.025, +0.031]. There was hydronephrosis or stranding in 23, an alternate diagnosis to explain pain in 1, clinically unimportant incidental findings in 37, likely unimportant findings in 5, and potentially significant findings in 9 patients. CONCLUSION 50% CT dose reconstructed with IR was equivalent to standard dose reconstructed with conventional FBP to detect urolithiasis. CLINICAL RELEVANCE/APPLICATION 50% dose reduction does not alter urolithiasis identification efficacy.
    Radiological Society of North America 2013 Scientific Assembly and Annual Meeting; 12/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: PURPOSE To compare the dose and noise level between single energy (SE) and dual energy (DE) multi-detector computed tomography (MDCT) examinations in patients undergoing screening for Hepatocellular Carcinoma (HCC). METHOD AND MATERIALS IRB-approved, HIPPA-compliant prospective study of 59 adult subjects (mean age 59.5yrs) undergoing HCC screening with 3-phase CT (unenhanced, arterial and portal-venous phases), who were each examined on both SE (Sensation 64, Siemens Healthcare) and DE CT scanners (Flash, Siemens Healthcare) on different dates. SE scans were performed using 120kVp and weight-based mAs (mAs=patient’s weight), and DE scans at 100kVP and 140kVp, with mAs adjusted to match the estimated CTDIvol of a weight-based mAs SE scan. The CTDIvol and DLP of each phase were recorded. Maximum anteroposterior and transverse dimensions measured from CT radiographs were used to calculate the effective diameter (ED) and size-specific dose estimate (SSDE). Regions of interest (ROI) were drawn in liver, retroperitoneal (RP) fat, IVC, and aorta and Hounsfield unit values with Standard Deviation (SD) recorded. Paired t-tests were used to compare BMI, weight, and ED at the time of the two imaging studies. Distributions of outcome variables (dose and noise) were examined using Q-Q plots and Shapiro tests. RESULTS BMI and weight of the subjects were highly correlated with the ED (r=0.75 and 0.87) and did not differ significantly between the two scans. CTDIvol and SSDE were significantly lower for all the phases on DE scans compared to SE scans (p-values<0.001 for all the comparisons). On the DE scans, the mean CTDIvol and SSDE were 8.6 and 8.2mGy, respectively, for each unenhanced and arterial phase, and 8.8 and 8.5mGy for the portal venous phase. On the SE scans, the mean CTDIvol and SSDE were 12.6 and 12.7 mGy for each unenhanced and arterial phase, and 13.5 and 13.7mGy for the portal venous phase The mean noise was not significantly different for DE and SE scans for both the liver and RP fat (p-values 1.0), but was significantly lower for the DE than the SE scan for the IVC (22 vs. 23.2; p=0.025) and aorta (25.2 vs. 26.2;p=0.008). CONCLUSION Dose with the MDCT DE scanning protocol was significantly lower when compared to SE examinations, with either similar or lower noise levels. CLINICAL RELEVANCE/APPLICATION DE scanning protocols can be an alternative to decrease dose in patients undergoing HCC screening who require repetitive imaging.
    Radiological Society of North America 2013 Scientific Assembly and Annual Meeting; 12/2013
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Accurate assessment of air density used to quantitatively characterize amount and distribution of emphysema in chronic obstructive pulmonary disease (COPD) subjects has remained challenging. Hounsfield units (HU) within tracheal air can be considerably less negative than -1000 HU. This study has sought to characterize the effects of improved scatter correction used in dual-source pulmonary computed tomography (CT). Dual-source dual-energy (DSDE) and single-source (SS) scans taken at multiple energy levels and scan settings were acquired for quantitative comparison using anesthetized ovine (n = 6), swine (n = 13), and a lung phantom. Data were evaluated for the lung, inferior vena cava, and tracheal segments. To minimize the effect of cross-scatter, the phantom scans in the DSDE mode were obtained by reducing the current of one of the tubes to near zero. A significant shift in mean HU values in the tracheal regions of animals and the phantom is observed, with values consistently closer to -1000 HU in DSDE mode. HU values associated with SS mode demonstrated a positive shift of up to 32 HU. In vivo tracheal air measurements demonstrated considerable variability with SS scanning, whereas these values were more consistent with DSDE imaging. Scatter effects in the lung parenchyma differed from adjacent tracheal measures. Data suggest that the scatter correction introduced into the dual-energy mode of imaging has served to provide more accurate CT lung density measures sought to quantitatively assess the presence and distribution of emphysema in COPD subjects. Data further suggest that CT images, acquired without adequate scatter correction, cannot be corrected by linear algorithms given the variability in tracheal air HU values and the independent scatter effects on lung parenchyma.
    Academic radiology 11/2013; 20(11):1334-1343. DOI:10.1016/j.acra.2013.04.018 · 2.08 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: Low‐contrast lesion detection is critical for abdomen CT. However, there is no standardized, objective measure of the detectability of low‐contrast objects. The purpose of this study is to investigate a new measure, the ability of a circular lesion to maintain its shape or circularity, as a potential indicator of lesion detectability. Methods: An anthropomorphic phantom with embedded spherical liver lesions was custom‐designed to assess the effectiveness of iterative reconstruction for improving the low contrast lesion detection. This phantom consists of 36 spherical lesions with 3 different sizes (5mm, 10mm, and 15mm). Each size has three different contrast levels. All spherical lesions were embedded non‐uniformly to minimize the memory bias for a concurrent human observer study. Phantom was scanned with a MDCT (Siemens Definition Flash) at multiple dose levels and images were reconstructed with both filtered back‐projection (FBP) and Iterative Reconstruction (SAFIRE). An adaptive Wiener filter was applied during the lesion detection. The circularity was calculated based on the ratio of the area to (perimeter)2 of each detected lesion. Results: The circularity was normalized to 1 when the detected lesion has a perfectly circular shape and the boundary is connected. At 50mAs, the circularity for a 10mm spherical lesion with 18HU contrast was 0.618 with FBP vs. 0.650 with SAFIRE; While at the same mAs, for a 15mm spherical lesion with 12HU contrast, the circularity was 0.666 with FBP vs. 0.763 for SAFIRE . At 200mAs, the circularity for a 10mm spherical lesion was 0.811with FBP vs. 0.870 with SAFIRE ; for a 15mm lesion, the circularity was 0.818 with FBP vs. 0.863 with SAFIRE. Conclusion: Circularity decreased with decreasing lesion size, contrast differential and dose. Iterative reconstruction techniques improved upon lesion circularity when compared to FBP at the same dose level, at similar lesion size and density differences. Dr. Herts and Dr. Baker are currently funded by Siemens Healthcare for research in low dose abdominal CT.
    Medical Physics 06/2013; 40(6):133. DOI:10.1118/1.4814147 · 3.01 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: Contrast‐to‐Noise ratio (CNR) is one measure to assess image quality and lesion detectability for CT exams. The purpose of this study is to calculate the change in CNR between standard filtered back‐projection (FBP) and sinogram‐affirmed iterative reconstruction (SAFIRE) by using an anthropomorphic phantom with spherical simulated liver lesions. Methods: An anthropomorphic phantom with embedded spherical liver lesions was custom‐designed to assess low contrast lesion detection at different CT doses and reconstruction techniques. This phantom consists of 36 spherical lesions with 3 different sizes (5mm, 10mm, and 15mm). Each size has three different contrast levels. All spherical lesions were embedded non‐uniformly to minimize the memory bias for a concurrent human observer study. Phantom was scanned with a MDCT (Siemens Definition Flash) at 4 dose levels and images were reconstructed with both FBP and SAFIRE using strength setting 3. CNR was computed as the ratio of the difference of the mean CT number at the center of the lesion and the liver background to the standard deviation of the CT numbers from the liver background. CNR was evaluated for the each lesion size at the center slice that intersected the middle of the lesion to minimize the error from the partial volume effect. Results: Averaged across all lesion sizes and the contrast levels (vs. the liver background), the CNR improvement from FBP to SAFIRE (strength 3) was 34% ±10% (mean ± standard deviation) at 50mAs,(p<0.0001), 33%±11% at 100mAs (p<0.0002), 36% ± 9% at 150mAs (p<0.0001) and 36% ± 7% at 200mAs (p<0.0003). Noise reduction from SAFIRE contributed almost entirely to the CNR improvement. Conclusion: Iterative reconstruction (SAFIRE) improved CNR of spherical low‐contrast liver lesions, but almost entirely due to noise reduction, with little improvement in lesion contrast. The overall CNR improvement was uniform for each dose level tested. Dr. Herts and Dr. Baker are funded by Siemens Healthcare for research in the field of abdominal CT imaging.
    Medical Physics 06/2013; 40(6):137. DOI:10.1118/1.4814162 · 3.01 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: OBJECTIVE. The purpose of this article is to determine the decrease in volume CT dose index (CTDIvol) and dose-length product (DLP) achieved by switching from fixed quality reference tube current protocols with automatic tube current modulation to protocols adjusting the quality reference tube current, slice collimation, and peak kilovoltage according to patient weight. MATERIALS AND METHODS. All adult patients who underwent CT examinations of the abdomen or abdomen and pelvis during 2010 using weight-based protocols who also underwent a CT examination in 2008 or 2009 using fixed quality reference tube current protocols were identified from the radiology information system. Protocol pages were electronically retrieved, and the CT model, examination date, scan protocol, CTDIvol, and DLP were extracted from the DICOM header or by optical character recognition. There were 15,779 scans with dose records for 2700 patients. Changes in CTDIvol and DLP were compared only between examinations of the same patient and same CT system model for examinations performed in 2008 or 2009 and those performed in 2010. The final analysis consisted of 1117 comparisons in 1057 patients, and 1209 comparisons in 988 patients for CTDIvol and DLP, respectively. RESULTS. The change to a weight-based protocol resulted in a statistically significant reduction in CTDIvol and DLP on three MDCT system models (p < 0.001). The largest average CTDIvol decrease was 13.9%, and the largest average DLP decrease was 16.1% on a 64-MDCT system. Both the CTDIvol and DLP decreased the most for patients who weighed less than 250 lb (112.5 kg). CONCLUSION. Adjusting the CT protocol by selecting parameters according to patient weight is a viable method for reducing CT radiation dose. The largest reductions occurred in the patients weighing less than 250 lb.
    American Journal of Roentgenology 06/2013; 200(6):1298-1303. DOI:10.2214/AJR.12.9387 · 2.74 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: PURPOSE The purpose of the study is to determine the total effective dose and cancer risk induced by treatment monitoring and surveillance CT scans in patients diagnosed with lymphoma. METHOD AND MATERIALS All patients diagnosed with lymphoma in 2008 were identified from a cancer center registry (N=111). CT dose data was retrieved in 76 patients (43 men, 33 women; 54 non-hodgkin’s lymphoma, 22 hodgkin’s lymphoma) from the electronic archive using optical character recognition software. Inclusion dates for dose data was from two months prior to treatment initiation through December 2011. Average follow up time was 26.7 months (range 1-40). Cumulative dose length product (DLP) for all head and neck examinations, and for all chest, abdomen, and pelvis examinations was calculated for each patient. Total effective dose was calculated for each patient during and following the treatment course using conversion factors of 0.015 times the DLP for chest, abdomen and pelvis exams, and 0.0031 for head and neck exams. Radiation-induced cancer risk was estimated based on data from BEIR VII. RESULTS Mean effective dose per patient for head and neck examinations was 5.6 mSv during treatment (N=9 patients) and 4.4 mSv after treatment (N=26). Mean effective dose per patient for chest, abdomen, and pelvis examinations was 26.7 mSv during treatment (N=43) and 44.1 mSv after treatment (N=59). Mean total effective dose per patient was 51.4 mSv (range 0.8-171). Mean cancer risk per patient defined as the US population averaged cancer mortality risk was 0.29% (0.01%-0.97%). CONCLUSION Mean cancer risk in lymphoma patients undergoing regular CT scans during the course of treatment as well during the surveillance period is relatively low. Therefore, mortality related to radiation-induced cancer should not be a consideration in determining radiologic follow-up in current lymphoma treatment protocols. CLINICAL RELEVANCE/APPLICATION The risk of radiation induced cancer should not be a consideration when using CT to assess treatment response and subsequent surveillance in lymphoma patients.
    Radiological Society of North America 2012 Scientific Assembly and Annual Meeting; 11/2012
  • [Show abstract] [Hide abstract]
    ABSTRACT: The purpose of this article is to evaluate the effect of sinogram-affirmed iterative reconstruction (SAFIRE) on contrast-to-noise ratio (CNR) compared with filtered back projection (FBP) and to determine whether SAFIRE improves low-contrast object detection or conspicuity in a low-contrast object phantom and in the liver on full- and low-dose examinations. A low-contrast object phantom was scanned at 100%, 70%, 50%, and 30% dose using a single-source made of a dual-source MDCT scanner, with the raw data reconstructed with SAFIRE and FBP. Unenhanced liver CT scans in 22 patients were performed using a dual-source MDCT. The raw data from both tubes (100% dose) were reconstructed using FBP, and data from one tube (50% dose) were reconstructed using both FBP and SAFIRE. CNR was measured in the phantom and in the liver. Noise, contrast, and CNR were compared using paired Student t tests. Six readers assessed sphere detection and conspicuity in the phantom and liver-inferior vena cava conspicuity in the patient data. The phantom and patient data were assessed using multiple-variable logistic regression. The phantom at 70% and 50% doses with SAFIRE had decreased noise and increased CNR compared with the 100% dose with FBP. In the liver, the mean CNR improvement at 50% dose with SAFIRE compared with FBP was 31.4% and 88% at 100% and 50% doses, respectively (p < 0.001). Sphere object detection and conspicuity improved with SAFIRE (p < 0.001). However, smaller spheres were obscured on both FBP and SAFIRE images at lower doses. Liver-vessel conspicuity improved with SAFIRE over 50%-dose FBP in 67.4% of cases (p < 0.001), and versus 100%-dose FBP, improved in 38.6% of cases (p = 0.085). As a predictor for detection, CNR alone had a discriminatory ability (c-index, 0.970) similar to that of the model that analyzed dose, lesion size, attenuation difference, and reconstruction technique (c-index, 0.978). Lower dose scans reconstructed with SAFIRE have a higher CNR. The ability of SAFIRE to improve low-contrast object detection and conspicuity depends on the radiation dose level. At low radiation doses, low-contrast objects are invisible, regardless of reconstruction technique.
    American Journal of Roentgenology 07/2012; 199(1):8-18. DOI:10.2214/AJR.11.7421 · 2.74 Impact Factor

Publication Stats

2k Citations
182.75 Total Impact Points

Institutions

  • 2010
    • Cleveland Clinic
      Cleveland, Ohio, United States
  • 2005–2010
    • Mayo Clinic - Rochester
      • Department of Radiology
      Рочестер, Minnesota, United States
  • 2007–2008
    • University of California, Los Angeles
      • Department of Radiology
      Los Angeles, CA, United States
  • 2006
    • University of Michigan
      Ann Arbor, Michigan, United States