Is it possible to use low-dose deep learning reconstruction for the detection of liver metastases on CT routinely?

OBJECTIVES: To compare the image quality and hepatic metastasis detection of low-dose deep learning image reconstruction (DLIR) with full-dose filtered back projection (FBP)/iterative reconstruction (IR). METHODS: A contrast-detail phantom consisting of low-contrast objects was scanned at five CT dose index levels (10, 6, 3, 2, and 1 mGy). A total of 154 participants with 305 hepatic lesions who underwent abdominal CT were enrolled in a prospective non-inferiority trial with a three-arm design based on phantom results. Data sets with full dosage (13.6 mGy) and low dosages (9.5, 6.8, or 4.1 mGy) were acquired from two consecutive portal venous acquisitions, respectively. All images were reconstructed with FBP (reference), IR (control), and DLIR (test). Eleven readers evaluated phantom data sets for object detectability using a two-alternative forced-choice approach. Non-inferiority analyses were performed to interpret the differences in image quality and metastasis detection of low-dose DLIR relative to full-dose FBP/IR. RESULTS: The phantom experiment showed the dose reduction potential from DLIR was up to 57% based on the reference FBP dose index. Radiation decreases of 30% and 50% resulted in non-inferior image quality and hepatic metastasis detection with DLIR compared to full-dose FBP/IR. Radiation reduction of 70% by DLIR performed inferiorly in detecting small metastases (< 1 cm) compared to full-dose FBP (difference: -0.112; 95% confidence interval [CI]: -0.178 to 0.047) and full-dose IR (difference: -0.123; 95% CI: -0.182 to 0.053) (p < 0.001). CONCLUSION: DLIR enables a 50% dose reduction for detecting low-contrast hepatic metastases while maintaining comparable image quality to full-dose FBP and IR. KEY POINTS: • Non-inferiority study showed that deep learning image reconstruction (DLIR) can reduce the dose to oncological patients with low-contrast lesions without compromising the diagnostic information. • Radiation dose levels for DLIR can be reduced to 50% of full-dose FBP and IR for detecting low-contrast hepatic metastases, while maintaining comparable image quality. • The reduction of radiation by 70% by DLIR is clinically acceptable but insufficient for detecting small low-contrast hepatic metastases (< 1 cm).

Effect of deep learning image reconstruction in the prediction of resectability of pancreatic cancer: Diagnostic performance and reader confidence.

OBJECTIVE: To assess the diagnostic performance and reader confidence in determining the resectability of pancreatic cancer at computed tomography (CT) using a new deep learning image reconstruction (DLIR) algorithm. METHODS: A retrospective review was conduct of on forty-seven patients with pathologically confirmed pancreatic cancers who underwent baseline multiphasic contrast-enhanced CT scan. Image data sets were reconstructed using filtered back projection (FBP), hybrid model-based adaptive statistical iterative reconstruction (ASiR-V) 60 %, and DLIR "TrueFidelity" at low(L), medium(M), and high strength levels(H). Four board-certified abdominal radiologists reviewed the CT images and classified cancers as resectable, borderline resectable, or unresectable. Diagnostic performance and reader confidence for categorizing the resectability of pancreatic cancer were evaluated based on the reference standards, and the interreader agreement was assessed using Fleiss k statistics. RESULTS: For prediction of margin-negative resections(ie, R0), the average area under the receiver operating characteristic curve was significantly higher with DLIR-H (0.91; 95 % confidence interval [CI]: 0.79, 0.98) than FBP (0.75; 95 % CI:0.60, 0.86) and ASiR-V (0.81; 95 % CI:0.67, 0.91) (p = 0.030 and 0.023 respectively). Reader confidence scores were significantly better using DLIR compared to FBP and ASiR-V 60 % and increased linearly with the increase of DLIR strength level (all p < 0.001). Among the image reconstructions, DLIR-H showed the highest interreader agreement in the resectability classification and lowest subject variability in the reader confidence. CONCLUSIONS: The DLIR-H algorithm may improve the diagnostic performance and reader confidence in the CT assignment of the local resectability of pancreatic cancer while reducing the interreader variability.

Multisite multivendor validation of a quantitative MRI and CT compatible fat phantom.

PURPOSE: Chemical shift-encoded magnetic resonance imaging enables accurate quantification of liver fat content though estimation of proton density fat-fraction (PDFF). Computed tomography (CT) is capable of quantifying fat, based on decreased attenuation with increased fat concentration. Current quantitative fat phantoms do not accurately mimic the CT number of human liver. The purpose of this work was to develop and validate an optimized phantom that simultaneously mimics the MRI and CT signals of fatty liver. METHODS: An agar-based phantom containing 12 vials doped with iodinated contrast, and with a granular range of fat fractions was designed and constructed within a novel CT and MR compatible spherical housing design. A four-site, three-vendor validation study was performed. MRI (1.5T and 3T) and CT images were obtained using each vendor's PDFF and CT reconstruction, respectively. An ROI centered in each vial was placed to measure MRI-PDFF (%) and CT number (HU). Mixed-effects model, linear regression, and Bland-Altman analysis were used for statistical analysis. RESULTS: MRI-PDFF agreed closely with nominal PDFF values across both field strengths and all MRI vendors. A linear relationship (slope = -0.54 ± 0.01%/HU, intercept = 37.15 ± 0.03%) with an R2 of 0.999 was observed between MRI-PDFF and CT number, replicating established in vivo signal behavior. Excellent test-retest repeatability across vendors (MRI: mean = -0.04%, 95% limits of agreement = [-0.24%, 0.16%]; CT: mean = 0.16 HU, 95% limits of agreement = [-0.15HU, 0.47HU]) and good reproducibility using GE scanners (MRI: mean = -0.21%, 95% limits of agreement = [-1.47%, 1.06%]; CT: mean = -0.18HU, 95% limits of agreement = [-1.96HU, 1.6HU]) were demonstrated. CONCLUSIONS: The proposed fat phantom successfully mimicked quantitative liver signal for both MRI and CT. The proposed fat phantom in this study may facilitate broader application and harmonization of liver fat quantification techniques using MRI and CT across institutions, vendors and imaging platforms.

Weed germinable seedbanks of rice-wheat systems in the Eastern Indo-Gangetic Plains: Do tillage and edaphic factors explain community variation?

Zero tillage (ZT) is widely promoted throughout India's Eastern Indo-Gangetic Plains (IGP) because of its potential to increase wheat productivity and resilience to abiotic stresses. Weeds remain a major barrier to ZT adoption, yet it remains unclear how ZT will influence weed communities in the Eastern-IGP. The primary objective of this study was to characterise the composition of the germinable weed seedbank sampled just prior to the wheat phase of rice-wheat farms in Bihar and Eastern Uttar Pradesh, and examine whether adoption of ZT wheat has shifted weed community composition compared to conventional tillage (CT). Additionally, we examined whether edaphic properties and topography (upland vs. lowland) explain variation in germinable weed seedbank communities. In December 2014, we evaluated the germinable seedbank from 72 fields differing in their historic (>=3 year) tillage practices (ZT vs. CT) in three regions: Samastipur-Vaishali-Muzaffarpur (SVM), Ara-Buxar and Maharajgunj-Kushinagar. Weed community composition and species richness varied by region and topography. ZT adoption was associated with lower relative density of Chenopodium album in the germinable seedbank and lower emergence of Phalaris minor seedlings within farmers' fields. In upland topographies of the SVM region, ZT adoption was also associated with greater relative abundance of Solanum nigrum in the weed seedbank. However, differences between tillage systems in individual species were not large enough to result in detection of differences at the whole-community level. Variation in edaphic properties, most notably soil texture and pH, explained 51% of the variation in the weed seedbank community. Our work suggests several frequent but poorly understood species (e.g. Mazus pumilus and Grangea maderaspatana) in Eastern IGP for which future research should quantify their effects on crop yields. Finally, future work surveying weed species abundance at harvest could further determine the dominant problematic species in these regions.

Has agricultural intensification impacted maize root traits and rhizosphere interactions related to organic N acquisition?

Plant-microbe interactions in the rhizosphere influence rates of organic matter mineralization and nutrient cycling that are critical to sustainable agricultural productivity. Agricultural intensification, particularly the introduction of synthetic fertilizer in the USA, altered the abundance and dominant forms of nitrogen (N), a critical plant nutrient, potentially imposing selection pressure on plant traits and plant-microbe interactions regulating N cycling and acquisition. We hypothesized that maize adaptation to synthetic N fertilization altered root functional traits and rhizosphere microbial nutrient cycling, reducing maize ability to acquire N from organic sources. Six maize genotypes released pre-fertilizer (1936, 1939, 1942) or post-fertilizer (1984, 1994, 2015) were grown in rhizoboxes containing patches of 15N-labelled clover/vetch residue. Multivariate approaches did not identify architectural traits that strongly and consistently predicted rhizosphere processes, though metrics of root morphological plasticity were linked to carbon- and N-cycling enzyme activities. Root traits, potential activities of extracellular enzymes (BG, LAP, NAG, urease), abundances of N-cycling genes (amoA, narG, nirK, nirS, nosZ) and uptake of organic N did not differ between eras of release despite substantial variation among genotypes and replicates. Thus, agricultural intensification does not appear to have impaired N cycling and acquisition from organic sources by modern maize and its rhizobiome. Improved mechanistic understanding of rhizosphere processes and their response to selective pressures will contribute greatly to rhizosphere engineering for sustainable agriculture.

Contrast Administration in CT: A Patient-Centric Approach.

Patient-centric care has garnered the attention of the radiology community. The authors describe a patient-centric approach to iodinated contrast administration designed to optimize the diagnostic yield of contrast-enhanced CT while minimizing patient iodine load and exposure to ionizing radiation, thereby enhancing patient safety while providing reasonable diagnostic efficacy. Patient-centric CT hardware settings and contrast media administration are important considerations for clinical CT quality and safety.

EFFECTIVE DOSE ESTIMATION FROM ORGAN DOSE MEASUREMENTS IN FAST-kV SWITCH DUAL ENERGY COMPUTED TOMOGRAPHY.

The purpose of this study was to validate a novel approach to estimating effective dose (E) in 'fast-kV switch dual energy computed tomography' using MOSFET detectors. The effective energy of the combined dual energy environment was characterized with the dual energy CT scanner and then MOSFETs were calibrated matching to the effective energy of the dual energy CT beam with a conventional CT beam. The calibration method was then experimentally validated by comparing the dose between MOSFET and an ion chamber (IC) using a standard CTDI body phantom. The measured doses of the MOSFET and IC were 17.1 mGy ± 3.8% and 17.1 mGy ± 0.4%, respectively. To measure organ doses, an adult anthropomorphic phantom loaded with 18 MOSFET detectors was scanned using a standard fast-kV switch dual energy abdomen/pelvis CT protocol. E was calculated by applying ICRP 103 tissue weighting factors as well as partial volume correction factors for organs that were not completely covered by the protocol field-of-view. E from the dual energy abdomen/pelvis CT was calculated to be 17.8 mSv ± 11.6%. This calculation was then compared to E from dose length product method, which yielded 14.62 mSv.

An Optimized Rhizobox Protocol to Visualize Root Growth and Responsiveness to Localized Nutrients.

Roots are notoriously difficult to study. Soil is both a visual and mechanical barrier, making it difficult to track roots in situ without destructive harvest or expensive equipment. We present a customizable and affordable rhizobox method that allows the non-destructive visualization of root growth over time and is particularly well-suited to studying root plasticity in response to localized resource patches. The method was validated by assessing maize genotypic variation in plasticity responses to patches containing 15N-labeled legume residue. Methods are described to obtain representative developmental measurements over time, measure root length density in resource-containing and control patches, calculate root growth rates, and determine 15N recovery by plant roots and shoots. Advantages, caveats, and potential future applications of the method are also discussed. Although care must be taken to ensure that experimental conditions do not bias root growth data, the rhizobox protocol presented here yields reliable results if carried out with sufficient attention to detail.

Emulsification Increases the Acute Ketogenic Effect and Bioavailability of Medium-Chain Triglycerides in Humans: Protein, Carbohydrate, and Fat Metabolism.

Background: Lower-brain glucose uptake is commonly present before the onset of cognitive deterioration associated with aging and may increase the risk of Alzheimer disease. Ketones are the brain's main alternative energy substrate to glucose. Medium-chain triglycerides (MCTs) are rapidly ß-oxidized and are ketogenic but also have gastrointestinal side effects. We assessed whether MCT emulsification into a lactose-free skim-milk matrix [emulsified MCTs (MCT-Es)] would improve ketogenesis, reduce side effects, or both compared with the same oral dose of MCTs consumed without emulsification [nonemulsified MCTs (MCT-NEs)]. Objectives: Our aims were to show that, in healthy adults, MCT-Es will induce higher ketonemia and have fewer side effects than MCT-NEs and the effects of MCT-NEs and MCT-Es on ketogenesis and plasma medium-chain fatty acids (MCFAs) will be dose-dependent. Methods: Using a metabolic study day protocol, 10 healthy adults were each given 3 separate doses (10, 20, or 30 g) of MCT-NEs or MCT-Es with a standard breakfast or no treatment [control (CTL)]. Blood samples were taken every 30 min for 4 h to measure plasma ketones (ß-hydroxybutyrate and acetoacetate), octanoate, decanoate, and other metabolites. Participants completed a side-effects questionnaire at the end of each study day. Results: Compared with CTL, MCT-NEs increased ketogenesis by 2-fold with no significant differences between doses. MCT-Es increased total plasma ketones by 2- to 4-fold in a dose-dependent manner. Compared with MCT-NEs, MCT-Es increased plasma MCFA bioavailability (F) by 2- to 3-fold and decreased the number of side effects by ∼50%. Conclusions: Emulsification increased the ketogenic effect and decreased side effects in a dose-dependent manner for single doses of MCTs ≤30 g under matching conditions. Further investigation is needed to establish whether emulsification could sustain ketogenesis and minimize side effects and therefore be used as a treatment to change brain ketone availability over a prolonged period of time. This trial was registered at clinicaltrials.gov as NCT02409927.

A NOVEL APPROACH FOR EFFECTIVE DOSE MEASUREMENTS IN DUAL-ENERGY CT.

A novel method was presented for the effective dose (ED) measurement with metal-oxide-semiconductor field-effect transistor (MOSFET) detectors in dual-energy (DE) dual-source (DS) computed tomography (CT) scanner. This study demonstrated that the mean energy of the combined spectrum in dual-source computed tomography can be used to measure the ED. For validation, the MOSFET dose at the centre cavity of a CT dose index (CTDI) body phantom was compared with the dose measured with an energy-independent ion chamber (IC). A clinical abdomen/pelvis scan was performed with an adult anthropomorphic phantom, and ED was compared between the MOSFET method and the dose length product (DLP) method. The tissue doses in the CTDI phantom were 2.08±(2.70 %) with IC and 2.20±(4.82 %) cGy with MOSFET; the per cent difference was 5.91 %, and the t-test showed that there was no statistically significant difference. EDs for the abdomen/pelvis scans were 5.01±(2.34 %) mSv with MOSFET and 5.56 mSv with the DLP method.

Feasibility of using the computed tomography dose indices to estimate radiation dose to partially and fully irradiated brains in pediatric neuroradiology examinations.

The purpose of this study was two-fold: (a) to measure the dose to the brain using clinical protocols at our institution, and (b) to develop a scanner-independent dosimetry method to estimate brain dose. Radiation dose was measured with a pediatric anthropomorphic phantom and MOSFET detectors. Six current neuroradiology protocols were used: brain, sinuses, facial bones, orbits, temporal bones, and craniofacial areas. Two different CT vendor scanners (scanner A and B) were used. Partial volume correction factors (PVCFs) were determined for the brain to account for differences between point doses measured by the MOSFETs and average organ dose. The CTDIvol and DLP for each protocol were recorded. The dose to the brain (mGy) for scanners A and B was 10.7 and 10.0 for the brain protocol, 7.8 and 3.2 for the sinus, 10.2 and 8.6 for the facial bones, 7.4 and 4.7 for the orbits and 1.6 and 1.9 for the temporal bones, respectively. On scanner A, the craniofacial protocol included a standard and high dose option; the dose measured for these exams was 3.9 and 16.9 mGy, respectively. There was only one craniofacial protocol on scanner B; the brain dose measured on this exam was 4.8 mGy. A linear correlation was found between DLP and brain dose with the conversion factors: 0.049 (R(2) = 0.87), 0.046 (R(2) = 0.89) for scanner A and B, and 0.048 (R(2) = 0.89) for both scanners. The range of dose observed was between 1.8 and 16.9 mGy per scan. This suggests that brain dose estimates may be made from DLP.

Lifetime Attributable Risk of Cancer From Radiation Exposure During Parathyroid Imaging: Comparison of 4D CT and Parathyroid Scintigraphy.

OBJECTIVE: The purpose of this study is to measure the organ doses and effective dose (ED) for parathyroid 4D CT and scintigraphy and to estimate the lifetime attributable risk of cancer incidence associated with imaging. MATERIALS AND METHODS: Organ radiation doses for 4D CT and scintigraphy were measured on the basis of imaging with our institution's protocols. An anthropomorphic phantom with metal oxide semiconductor field effect transistor detectors was scanned to measure CT organ dose. Organ doses from the radionuclide were based on International Commission for Radiological Protection report 80. ED was calculated for 4D CT and scintigraphy and was used to estimate the lifetime attributable risk of cancer incidence for patients differing in age and sex with the approach established by the Biologic Effects of Ionizing Radiation VII report. A 55-year-old woman was selected as the standard patient according to the demographics of patients with primary hyperparathyroidism. RESULTS: Organs receiving the highest radiation dose from 4D CT were the thyroid (150.6 mGy) and salivary glands (137.8 mGy). For scintigraphy, the highest organ doses were to the colon (41.5 mGy), gallbladder (39.8 mGy), and kidneys (32.3 mGy). The ED was 28 mSv for 4D CT, compared with 12 mSv for scintigraphy. In the exposed standard patient, the lifetime attributable risk for cancer incidence was 193 cancers/100,000 patients for 4D CT and 68 cancers/100,000 patients for scintigraphy. Given a baseline lifetime incidence of cancer of 46,300 cancers/100,000 patients, imaging results in an increase in lifetime incidence of cancer over baseline of 0.52% for 4D CT and 0.19% for scintigraphy. CONCLUSION: The ED of 4D CT is more than double that of scintigraphy, but both studies cause negligible increases in lifetime risk of cancer. Clinicians should not allow concern for radiation-induced cancer to influence decisions regarding workup in older patients.

A novel method of estimating effective dose from the point dose method: a case study--parathyroid CT scans.

The purpose of this study was to validate a novel approach of applying a partial volume correction factor (PVCF) using a limited number of MOSFET detectors in the effective dose (E) calculation. The results of the proposed PVCF method were compared to the results from both the point dose (PD) method and a commercial CT dose estimation software (CT-Expo). To measure organ doses, an adult female anthropomorphic phantom was loaded with 20 MOSFET detectors and was scanned using the non-contrast and 2 phase contrast-enhanced parathyroid imaging protocols on a 64-slice multi-detector computed tomography scanner. E was computed by three methods: the PD method, the PVCF method, and the CT-Expo method. The E (in mSv) for the PD method, the PVCF method, and CT-Expo method was 2.6  ±  0.2, 1.3  ±  0.1, and 1.1 for the non-contrast scan, 21.9  ±  0.4, 13.9  ±  0.2, and 14.6 for the 1st phase of the contrast-enhanced scan, and 15.5  ±  0.3, 9.8  ±  0.1, and 10.4 for the 2nd phase of the contrast-enhanced scan, respectively. The E with the PD method differed from the PVCF method by 66.7% for the non-contrast scan, by 44.9% and by 45.5% respectively for the 1st and 2nd phases of the contrast-enhanced scan. The E with PVCF was comparable to the results from the CT-Expo method with percent differences of 15.8%, 5.0%, and 6.3% for the non-contrast scan and the 1st and 2nd phases of the contrast-enhanced scan, respectively. To conclude, the PVCF method estimated E within 16% difference as compared to 50-70% in the PD method. In addition, the results demonstrate that E can be estimated accurately from a limited number of detectors.

Effects of model-based iterative reconstruction on image quality for low-dose computed tomographic angiography of the thoracic aorta in a Swine model.

PURPOSE: The aim of the study was to assess the image quality of multi-detector-row computed tomography (CT) angiographic images of the thoracic aorta reconstructed with filtered back projection (FBP), adaptive statistical iterative reconstruction, and model-based iterative reconstruction (MBIR) at different kVp and mA settings. METHODS: A healthy 56.1-kg Yorkshire pig underwent sequential arterial CT angiograms on a 64-slice multi-detector-row CT scanner (Discovery CT 750HD; GE Healthcare Inc, Milwaukee, Wis) at progressively lower kVp and mA settings. At 120-, 100-, and 80-kVp levels, the pig was scanned at 700, 400, 200, 100, and 50 mA at, for a total of 15 scans. Each scan was reconstructed with FBP, adaptive statistical iterative reconstruction (50% blend), and MBIR. Relative noise and contrast-to-noise ratio (CNR) were calculated from regions of interest over the aorta and paraspinous muscle. In addition, selected axial and oblique sagittal images were scored subjectively for both aortic wall visibility and for overall image quality. RESULTS: Averaged across all kVp and mA variations, MBIR reduced relative noise by 73.9% and improved CNR by 227% compared with FBP; MBIR reduced relative noise by 63.4% and improved CNR by 107% compared with ASIR. The effects were more pronounced in lower tube output settings. At 100 kVp/700 mA, MBIR reduced noise by 57% compared with FBP and 40% compared with ASIR. At 100 kVp/50 mA, MBIR reduced noise by 82% compared with FBP and 75% compared with ASIR. Subjective improvements in image quality were noted only in higher noise settings. CONCLUSIONS: Model-based iterative reconstruction reduces relative noise and improves CNR compared with ASIR and FBP at all kVp and mA settings, which were significantly greater at lower mA settings.

Accuracy of effective dose estimation in personal dosimetry: a comparison between single-badge and double-badge methods and the MOSFET method.

The purpose of this study was three-fold: (1) to measure the transmission properties of various lead shielding materials, (2) to benchmark the accuracy of commercial film badge readings, and (3) to compare the accuracy of effective dose (ED) conversion factors (CF) of the U.S. Nuclear Regulatory Commission methods to the MOSFET method. The transmission properties of lead aprons and the accuracy of film badges were studied using an ion chamber and monitor. ED was determined using an adult male anthropomorphic phantom that was loaded with 20 diagnostic MOSFET detectors and scanned with a whole body CT protocol at 80, 100, and 120 kVp. One commercial film badge was placed at the collar and one at the waist. Individual organ doses and waist badge readings were corrected for lead apron attenuation. ED was computed using ICRP 103 tissue weighting factors, and ED CFs were calculated by taking the ratio of ED and badge reading. The measured single badge CFs were 0.01 (±14.9%), 0.02 (±9.49%), and 0.04 (±15.7%) for 80, 100, and 120 kVp, respectively. Current regulatory ED CF for the single badge method is 0.3; for the double-badge system, they are 0.04 (collar) and 1.5 (under lead apron at the waist). The double-badge system provides a better coefficient for the collar at 0.04; however, exposure readings under the apron are usually negligible to zero. Based on these findings, the authors recommend the use of ED CF of 0.01 for the single badge system from 80 kVp (effective energy 50.4 keV) data.

Estimation of radiation exposure for brain perfusion CT: standard protocol compared with deviations in protocol.

OBJECTIVE: The purpose of this study was to measure the organ doses and estimate the effective dose for the standard brain perfusion CT protocol and erroneous protocols. MATERIALS AND METHODS: An anthropomorphic phantom with metal oxide semiconductor field effect transistor (MOSFET) detectors was scanned on a 64-MDCT scanner. Protocol 1 used a standard brain perfusion protocol with 80 kVp and fixed tube current of 200 mA. Protocol 2 used 120 kVp and fixed tube current of 200 mA. Protocol 3 used 120 kVp with automatic tube current modulation (noise index, 2.4; minimum, 100 mA; maximum, 520 mA). RESULTS: Compared with protocol 1, the effective dose was 2.8 times higher with protocol 2 and 7.8 times higher with protocol 3. For all protocols, the peak dose was highest in the skin, followed by the brain and calvarial marrow. Compared with protocol 1, the peak skin dose was 2.6 times higher with protocol 2 and 6.7 times higher with protocol 3. The peak skin dose for protocol 3 exceeded 3 Gy. The ocular lens received significant scatter radiation: 177 mGy for protocol 2 and 435 mGy for protocol 3, which were 4.6 and 11.3 times the dose for protocol 1, respectively. CONCLUSION: Compared with the standard protocol, erroneous protocols of increasing the tube potential from 80 kVp to 120 kVp will lead to a three- to fivefold increase in organ doses, and concurrent use of high peak kilovoltage with incorrectly programmed tube current modulation can increase dose to organs by 7- to 11-fold. Tube current modulation with a low noise index can lead to doses to the skin and ocular lens that are close to thresholds for tissue reactions.

Radiation exposure in urology: a genitourinary catalogue for diagnostic imaging.

PURPOSE: Computerized tomography use increased exponentially in the last 3 decades, and it is commonly used to evaluate many urological conditions. Ionizing radiation exposure from medical imaging is linked to the risk of malignancy. We measured the organ and calculated effective doses of different studies to determine whether the dose-length product method is an accurate estimation of radiation exposure. MATERIALS AND METHODS: An anthropomorphic male phantom validated for human organ dosimetry measurements was used to determine radiation doses. High sensitivity metal oxide semiconductor field effect transistor dosimeters were placed at 20 organ locations to measure specific organ doses. For each study the phantom was scanned 3 times using our institutional protocols. Organ doses were measured and effective doses were calculated on dosimetry. Effective doses measured by a metal oxide semiconductor field effect transistor dosimeter were compared to calculated effective doses derived from the dose-length product. RESULTS: The mean±SD effective dose on dosimetry for stone protocol, chest and abdominopelvic computerized tomography, computerized tomography urogram and renal cell carcinoma protocol computerized tomography was 3.04±0.34, 4.34±0.27, 5.19±0.64, 9.73±0.71 and 11.42±0.24 mSv, respectively. The calculated effective dose for these studies Was 3.33, 2.92, 5.84, 9.64 and 10.06 mSv, respectively (p=0.8478). CONCLUSIONS: The effective dose varies considerable for different urological computerized tomography studies. Renal stone protocol computerized tomography shows the lowest dose, and computerized tomography urogram and the renal cell carcinoma protocol accumulate the highest effective doses. The calculated effective dose derived from the dose-length product is a reasonable estimate of patient radiation exposure.

Radiation exposure in the follow-up of patients with urolithiasis comparing digital tomosynthesis, non-contrast CT, standard KUB, and IVU.

OBJECTIVE: To compare the effective doses (EDs) associated with imaging modalities for follow-up of patients with urolithiasis, including stone protocol non-contrast computed tomography (NCCT), kidney, ureter, and bladder radiograph (KUB), intravenous urogram (IVU), and digital tomosynthesis (DT). METHODS: A validated Monte-Carlo simulation-based software PCXMC 2.0 (STUK) designed for estimation of patient dose from medical X-ray exposures was used to determine the ED for KUB, IVU (KUB scout plus three tomographic images), and DT (two scouts and one tomographic sweep). Simulations were performed using a two-dimensional stationary field onto the corresponding body area of the built-in digital phantom, with actual kVp, mAs, and geometrical parameters of the protocols. The ED for NCCT was determined using an anthropomorphic male phantom that was placed prone on a 64-slice GE Healthcare volume computed tomography (VCT) scanner. High-sensitivity metal oxide semiconductor field effect transistors dosimeters were placed at 20 organ locations and used to measure organ radiation doses. RESULTS: The ED for a stone protocol NCCT was 3.04±0.34 mSv. The ED for a KUB was 0.63 and 1.1 mSv for the additional tomographic film. The total ED for IVU was 3.93 mSv. The ED for DT performed with two scouts and one sweep (14.2°) was 0.83 mSv. CONCLUSIONS: Among the different imaging modalities for follow-up of patients with urolithiasis, DT was associated with the least radiation exposure (0.83 mSv). This ED corresponds to a fifth of NCCT or IVU studies. Further studies are needed to demonstrate the sensitivity and specificity of DT for the follow-up of nephrolithiasis patients.

Obesity triples the radiation dose of stone protocol computerized tomography.

PURPOSE: Patients with recurrent nephrolithiasis are often evaluated and followed with computerized tomography. Obesity is a risk factor for nephrolithiasis. We evaluated the radiation dose of computerized tomography in obese and nonobese adults. MATERIALS AND METHODS: We scanned a validated, anthropomorphic male phantom according to our institutional renal stone evaluation protocol. The obese model consisted of the phantom wrapped in 2 Custom Fat Layers (CIRS, Norfolk, Virginia), which have been verified to have the same radiographic tissue density as fat. High sensitivity metal oxide semiconductor field effect transistor dosimeters were placed at 20 organ locations in the phantoms to measure organ specific radiation doses. The nonobese and obese models have an approximate body mass index of 24 and 30 kg/m(2), respectively. Three runs of renal stone protocol computerized tomography were performed on each phantom under automatic tube current modulation. Organ specific absorbed doses were measured and effective doses were calculated. RESULTS: The bone marrow of each model received the highest dose and the skin received the second highest dose. The mean ± SD effective dose for the nonobese and obese models was 3.04 ± 0.34 and 10.22 ± 0.50 mSv, respectively (p <0.0001). CONCLUSIONS: The effective dose of stone protocol computerized tomography in obese patients is more than threefold higher than the dose in nonobese patients using automatic tube current modulation. The implication of this finding extends beyond the urological stone population and adds to our understanding of radiation exposure from medical imaging.

Radiation dose estimations to the thorax using organ-based dose modulation.

OBJECTIVE: The purpose of this study was to assess the radiation dose distribution and image quality for organ-based dose modulation during adult thoracic MDCT. MATERIALS AND METHODS: Organ doses were measured using an anthropomorphic adult female phantom containing 30 metal oxide semiconductor field-effect transistor detectors on a dual-source MDCT scanner with two protocols: standard tube current modulation thoracic CT and organ-based dose modulation using a 120° radial arc. Radiochromic film measured the relative axial dose. Noise was measured to evaluate image quality. Breast tissue location across the anterior aspect of the thorax was retrospectively assessed in 100 consecutive thoracic MDCT examinations. RESULTS: There was a 17-47% decrease (p = < 0.05) in anterior thoracic organ dose and a maximum 52% increase (p = < 0.05) in posterior thoracic organ dose using organ-based dose modulation compared with tube current modulation. Effective dose (SD) for tube current modulation and organ-based dose modulation were 5.25 ± 0.36 mSv and 4.42 ± 0.30 mSv, respectively. Radiochromic film analysis showed a 30% relative midline anterior-posterior gradient. There was no statistically significant difference in image noise. Adult female breast tissue was located within an average anterior angle of 155° (123-187°). CONCLUSION: Organ-based dose modulation CT using an anterior 120° arc can reduce the organ dose in the anterior aspect of the thorax with a compensatory organ dose increase posteriorly without impairment of image quality. Laterally located breast tissue will have higher organ doses than medially located breast tissue when using organ-based dose modulation. The benefit of this dose reduction must be clinically determined on the basis of the relationship of the irradiated organs to the location of the prescribed radial arc used in organ-based dose modulation.