Spatial resolution, noise properties, and detectability index of a deep learning reconstruction algorithm for dual-energy CT of the abdomen.

BACKGROUND: Iterative reconstruction (IR) has increasingly replaced traditional reconstruction methods in computed tomography (CT). The next paradigm shift in image reconstruction is likely to come from artificial intelligence, with deep learning reconstruction (DLR) solutions already entering the clinic. An enduring disadvantage to IR has been a change in noise texture, which can affect diagnostic confidence. DLR has demonstrated the potential to overcome this issue and has recently become available for dual-energy CT. PURPOSE: To evaluate the spatial resolution, noise properties, and detectability index of a commercially available DLR algorithm for dual-energy CT of the abdomen and compare it to single-energy (SE) CT. METHODS: An oval 25 cm x 35 cm custom-made phantom was scanned on a GE Revolution CT scanner (GE Healthcare, Waukesha, WI) at two dose levels (13 and 5 mGy) and two iodine concentrations (8 and 2 mg/mL), using three typical abdominal scan protocols: dual-energy (DE), SE 80 kV (SE-80 kV) and SE 120 kV (SE-120 kV). Reconstructions were performed with three strengths of IR (ASiR-V: AR0%, AR50%, AR100%) and three strengths of DLR (TrueFidelity: low, medium, high). The DE acquisitions were reconstructed as mono-energetic images between 40 and 80 keV. The noise power spectrum (NPS), task transfer function (TTF), and detectability index (d') were determined for the reconstructions following the recommendations of AAPM Task Group 233. RESULTS: Noise magnitude reductions (relative to AR0%) for the SE protocols were on average (-29%, -21%) for (AR50%, TF-M), while for DE-70 keV were (-28%, -43%). There was less reduction in mean frequency (fav ) for DLR than for IR, with similar results for SE and DE imaging. There was, however, a substantial change in the NPS shape when using DE with DLR, quantifiable by a marked reduction in the peak frequency (fpeak ) that was absent in SE mode. All protocols and reconstructions (including AR0%) exhibited slight to moderate shifts towards lower spatial frequencies at the lower dose (<12% in fav ). Spatial resolution was consistently superior for DLR compared to IR for SE but not for DE. All protocols and reconstructions (including AR0%) showed decreased resolution with reduced dose and iodine concentration, with less decrease for DLR compared to IR. DLR displayed a higher d' than IR. The effect of energy was large: d' increased with lower keV, and SE-80 kV had higher d' than SE-120 kV. Using DE with DLR could provide higher d' than SE-80 kV at the higher dose but not at lower dose. CONCLUSIONS: DE imaging with DLR maintained spatial resolution and reduced noise magnitude while displaying less change in noise texture than IR. The d' was also higher with DLR than IR, suggesting superiority in detectability of iodinated contrast. Despite these trends being consistent with those previously established for SE imaging, there were some noteworthy differences. For DE imaging there was no improvement in resolution compared to IR and a change in noise texture. DE imaging with low keV and DLR had superior detectability to SE DLR at the high dose but was not better than SE-80 kV at low dose.

MR-Forest: A Deep Decision Framework for False Positive Reduction in Pulmonary Nodule Detection.

With the development of deep learning methods such as convolutional neural network (CNN), the accuracy of automated pulmonary nodule detection has been greatly improved. However, the high computational and storage costs of the large-scale network have been a potential concern for the future widespread clinical application. In this paper, an alternative Multi-ringed (MR)-Forest framework, against the resource-consuming neural networks (NN)-based architectures, has been proposed for false positive reduction in pulmonary nodule detection, which consists of three steps. First, a novel multi-ringed scanning method is used to extract the order ring facets (ORFs) from the surface voxels of the volumetric nodule models; Second, Mesh-LBP and mapping deformation are employed to estimate the texture and shape features. By sliding and resampling the multi-ringed ORFs, feature volumes with different lengths are generated. Finally, the outputs of multi-level are cascaded to predict the candidate class. On 1034 scans merging the dataset from the Affiliated Hospital of Liaoning University of Traditional Chinese Medicine (AH-LUTCM) and the LUNA16 Challenge dataset, our framework performs enough competitiveness than state-of-the-art in false positive reduction task (CPM score of 0.865). Experimental results demonstrate that MR-Forest is a successful solution to satisfy both resource-consuming and effectiveness for automated pulmonary nodule detection. The proposed MR-forest is a general architecture for 3D target detection, it can be easily extended in many other medical imaging analysis tasks, where the growth trend of the targeting object is approximated as a spheroidal expansion.

Breast pectoral muscle segmentation in mammograms using a modified holistically-nested edge detection network.

This paper presents a method for automatic breast pectoral muscle segmentation in mediolateral oblique mammograms using a Convolutional Neural Network (CNN) inspired by the Holistically-nested Edge Detection (HED) network. Most of the existing methods in the literature are based on hand-crafted models such as straight-line, curve-based techniques or a combination of both. Unfortunately, such models are insufficient when dealing with complex shape variations of the pectoral muscle boundary and when the boundary is unclear due to overlapping breast tissue. To compensate for these issues, we propose a neural network framework that incorporates multi-scale and multi-level learning, capable of learning complex hierarchical features to resolve spatial ambiguity in estimating the pectoral muscle boundary. For this purpose, we modified the HED network architecture to specifically find 'contour-like' objects in mammograms. The proposed framework produced a probability map that can be used to estimate the initial pectoral muscle boundary. Subsequently, we process these maps by extracting morphological properties to find the actual pectoral muscle boundary. Finally, we developed two different post-processing steps to find the actual pectoral muscle boundary. Quantitative evaluation results show that the proposed method is comparable with alternative state-of-the-art methods producing on average values of 94.8 ±â€¯8.5% and 97.5 ±â€¯6.3% for the Jaccard and Dice similarity metrics, respectively, across four different databases.

Iodine Maps from Subtraction CT or Dual-Energy CT to Detect Pulmonary Emboli with CT Angiography: A Multiple-Observer Study.

Background Dual-energy CT iodine maps are used to detect pulmonary embolism (PE) with CT angiography but require dedicated hardware. Subtraction CT, a software-only solution, results in iodine maps with high contrast-to-noise ratios. Purpose To compare the use of subtraction CT versus dual-energy CT iodine maps to CT angiography for PE detection. Materials and Methods In this prospective study ( https://clinicaltrials.gov , NCT02890706), 274 participants suspected of having PE underwent precontrast CT followed by contrast material-enhanced dual-energy CT angiography between July 2016 and April 2017. Iodine maps from dual-energy CT were derived. Subtraction maps (contrast-enhanced CT minus precontrast CT) were calculated after motion correction. Truth was established by expert consensus. A total of 75 randomly selected participants with and without PE (1:1 ratio) were evaluated by three radiologists and six radiology residents (blinded to final diagnosis) for the presence of PE using three types of CT: CT angiography alone, dual-energy CT, and subtraction CT. The partial area under the receiver operating characteristic curve (AUC) for the clinically relevant specificity region (maximum partial AUC, 0.11) was compared by using multireader multicase variance. A P value less than or equal to .025 was considered indicative of a significant difference due to multiple comparisons. Results There were 35 men and 40 women in the reader study (mean age, 63 years ± 12 [standard deviation]). The pooled sensitivities were not different (P ≥ .31 among techniques) (95% confidence intervals [CIs]: 67%, 89% for CT angiography; 72%, 91% for dual-energy CT; 70%, 91% for subtraction CT). However, pooled specificity was higher for subtraction CT (95% CI: 100%, 100%) than for CT angiography (95% CI: 89%, 97%) or dual-energy CT (95% CI: 89%, 98%) (P < .001). Partial AUCs for the average observer improved equally when adding iodine maps (subtraction CT [0.093] vs CT angiography [0.088], P = .03; dual-energy CT [0.094] vs CT angiography, P = .01; dual-energy CT vs subtraction CT, P = .68). Average reading times were equivalent (range, 97-101 seconds; P ≥ .41) among techniques. Conclusion Subtraction CT iodine maps had greater specificity than CT angiography alone in pulmonary embolism detection. Subtraction CT had comparable diagnostic performance to that of dual-energy CT, without the need for dedicated hardware. © RSNA, 2019 Online supplemental material is available for this article.

Improved visualization of the coronary arteries using motion correction during vasodilator stress CT myocardial perfusion imaging.

BACKGROUND: Vasodilator stress computed tomography perfusion (sCTP) imaging is complementary to coronary CT angiography (CCTA), used to determine the hemodynamic significance of coronary artery disease. However, it requires a separate image acquisition due to motion artifacts caused by higher heart rates during stress, resulting in increased iodine contrast dose and radiation. We sought to determine whether a novel motion correction algorithm applied to stress images would improve the visualization of the coronary arteries to potentially allow CCTA + sCTP evaluation in a single scan. METHODS: 28 patients referred for clinically indicated CCTA (iCT, Philips) underwent sCTP imaging (retrospective-gating with dose modulation; 100 kVp and 250 mA; 5.2 ± 4.3 mSv) after regadenoson (0.4 mg, Astellas). Stress images were reconstructed using standard filtered back-projection (FBP) and also processed to generate interaction-free coronary motion-compensated back-projection reconstructions (MCR). Each coronary artery from standard FBP and MCR images was viewed side-by-side by a reader blinded to the reconstruction technique, who graded severity of motion artifact by segment (scale 0-5, with 3 as the threshold for diagnostic quality) and to measure signal-to-noise and contrast-to-noise ratios (SNR, CNR). RESULTS: Visualization scores were higher with MCR for all coronary segments, including 14/86 (16%) segments deemed as non-diagnostic on FBP images. SNR (7 ± 2) and CNR (15 ± 8) were unchanged by motion-correction (7 ± 3, p = 0.88 and 15 ± 5, p = 0.94, respectively). CONCLUSIONS: MCR improves the visualization of coronary anatomy on sCTP images without degrading image characteristics. This algorithm is an important step towards the combined assessment of coronary anatomy and myocardial perfusion in a single scan, which will reduce study time, radiation exposure and contrast dose.

Abdominal multi-organ segmentation with organ-attention networks and statistical fusion.

Accurate and robust segmentation of abdominal organs on CT is essential for many clinical applications such as computer-aided diagnosis and computer-aided surgery. But this task is challenging due to the weak boundaries of organs, the complexity of the background, and the variable sizes of different organs. To address these challenges, we introduce a novel framework for multi-organ segmentation of abdominal regions by using organ-attention networks with reverse connections (OAN-RCs) which are applied to 2D views, of the 3D CT volume, and output estimates which are combined by statistical fusion exploiting structural similarity. More specifically, OAN is a two-stage deep convolutional network, where deep network features from the first stage are combined with the original image, in a second stage, to reduce the complex background and enhance the discriminative information for the target organs. Intuitively, OAN reduces the effect of the complex background by focusing attention so that each organ only needs to be discriminated from its local background. RCs are added to the first stage to give the lower layers more semantic information thereby enabling them to adapt to the sizes of different organs. Our networks are trained on 2D views (slices) enabling us to use holistic information and allowing efficient computation (compared to using 3D patches). To compensate for the limited cross-sectional information of the original 3D volumetric CT, e.g., the connectivity between neighbor slices, multi-sectional images are reconstructed from the three different 2D view directions. Then we combine the segmentation results from the different views using statistical fusion, with a novel term relating the structural similarity of the 2D views to the original 3D structure. To train the network and evaluate results, 13 structures were manually annotated by four human raters and confirmed by a senior expert on 236 normal cases. We tested our algorithm by 4-fold cross-validation and computed Dice-Sørensen similarity coefficients (DSC) and surface distances for evaluating our estimates of the 13 structures. Our experiments show that the proposed approach gives strong results and outperforms 2D- and 3D-patch based state-of-the-art methods in terms of DSC and mean surface distances.

Calibration-free beam hardening correction for myocardial perfusion imaging using CT.

PURPOSE: Computed tomography myocardial perfusion imaging (CT-MPI) and coronary CTA have the potential to make CT an ideal noninvasive imaging gatekeeper exam for invasive coronary angiography. However, beam hardening (BH) artifacts prevent accurate blood flow calculation in CT-MPI. BH correction methods require either energy-sensitive CT, not widely available, or typically, a calibration-based method in conventional CT. We propose a calibration-free, automatic BH correction (ABHC) method suitable for CT-MPI and evaluate its ability to reduce BH artifacts in single "static-perfusion" images and to create accurate myocardial blood flow (MBF) in dynamic CT-MPI. METHODS: In the algorithm, we used input CT DICOM images and iteratively optimized parameters in a polynomial BH correction until a BH-sensitive cost function was minimized on output images. An input image was segmented into a soft tissue image and a highly attenuating material (HAM) image containing bones and regions of high iodine concentrations, using mean HU and temporal enhancement properties. We forward projected HAM, corrected projection values according to a polynomial correction, and reconstructed a correction image to obtain the current iteration's BH corrected image. The cost function was sensitive to BH streak artifacts and cupping. We evaluated the algorithm on simulated CT and physical phantom images, and on preclinical porcine with optional coronary obstruction and clinical CT-MPI data. Assessments included measures of BH artifact in single images as well as MBF estimates. We obtained CT images on a prototype spectral detector CT (SDCT, Philips Healthcare) scanner that provided both conventional and virtual keV images, allowing us to quantitatively compare corrected CT images to virtual keV images. To stress test the method, we evaluated results on images from a different scanner (iCT, Philips Healthcare) and different kVp values. RESULTS: In a CT-simulated digital phantom consisting of water with iodine cylinder insets, BH streak artifacts between simulated iodine inserts were reduced from 13 ± 2 to 0 ± 1 HU. In a similar physical phantom having higher iodine concentrations, BH streak artifacts were reduced from 48 ± 6 to 1 ± 5 HU and cupping was reduced by 86%, from 248 to 23 HU. In preclinical CT-MPI images without coronary obstruction, BH artifact was reduced from 24 ± 6 HU to less than 5 ± 4 HU at peak enhancement. Standard deviation across different regions of interest (ROI) along the myocardium was reduced from 13.26 to 6.86 HU for ABHC, comparing favorably to measurements in the corresponding virtual keV image. Corrections greatly reduced variations in preclinical MBF maps as obtained in normal animals without obstruction (FFR = 1). Coefficients of variations were 22% (conventional CT), 9% (ABHC), and 5% (virtual keV). Moreover, variations in flow tended to be localized after ABHC, giving result which would not be confused with a flow deficit in a coronary vessel territory. CONCLUSION: The automated algorithm can be used to reduce BH artifact in conventional CT and improve CT-MPI accuracy particularly by removing regions of reduced estimated flow which might be misinterpreted as flow deficits.

Fully automatic detection and segmentation of abdominal aortic thrombus in post-operative CTA images using Deep Convolutional Neural Networks.

Computerized Tomography Angiography (CTA) based follow-up of Abdominal Aortic Aneurysms (AAA) treated with Endovascular Aneurysm Repair (EVAR) is essential to evaluate the progress of the patient and detect complications. In this context, accurate quantification of post-operative thrombus volume is required. However, a proper evaluation is hindered by the lack of automatic, robust and reproducible thrombus segmentation algorithms. We propose a new fully automatic approach based on Deep Convolutional Neural Networks (DCNN) for robust and reproducible thrombus region of interest detection and subsequent fine thrombus segmentation. The DetecNet detection network is adapted to perform region of interest extraction from a complete CTA and a new segmentation network architecture, based on Fully Convolutional Networks and a Holistically-Nested Edge Detection Network, is presented. These networks are trained, validated and tested in 13 post-operative CTA volumes of different patients using a 4-fold cross-validation approach to provide more robustness to the results. Our pipeline achieves a Dice score of more than 82% for post-operative thrombus segmentation and provides a mean relative volume difference between ground truth and automatic segmentation that lays within the experienced human observer variance without the need of human intervention in most common cases.

Renal Lesion Characterization with Spectral CT: Determining the Optimal Energy for Virtual Monoenergetic Reconstruction.

Purpose To investigate the relationship between energy level of virtual monoenergetic (VM) imaging and sensitivity in the detection of minimally enhancing renal lesions. Materials and Methods Phantoms simulating unenhanced and contrast material-enhanced renal parenchyma were equipped with inserts containing different concentrations of iodine (range, 0-1.15 mg iodine per milliliter). A total of 180 patients (117 men; mean age, 65.2 years ± 13.0 [standard deviation]) with 194 (62 solid, 132 cystic) renal lesions larger than 10 mm in diameter underwent unenhanced single-energy CT and contrast-enhanced dual-energy CT. VM imaging data sets were created for 70, 80, 90, and 100 keV. Renal lesions were measured, and enhancement was calculated. Area under the receiver operating characteristic curve (AUC) for renal lesion characterization was determined by using the DeLong method. Results The AUC was highest at 70 keV and decreased as energy increased toward 100 keV. AUC in the phantom decreased from 98% (95% confidence interval [CI]: 95, 100) at 70 keV to 88% (95% CI: 79, 96) at 100 keV (P = .004). AUC in patients decreased from 96% (95% CI: 94, 98) at 70 keV to 79% (95% CI: 71, 86) at 100 keV (P = .001). In patients with an enhancement threshold of 15 HU, sensitivity in the detection of solid renal lesions decreased between from 91% (49 of 62 [95% CI: 78, 97]) at 70 keV to 48% (33 of 62 [95% CI: 25, 71]) at 100 keV (P < .05), with no change in specificity (93% [120 of 132 {95% CI: 87, 97}] at 70 keV, 97% [125 of 132 {95% CI: 92, 99}] at 100 keV). Conclusion There is a reduction in diagnostic accuracy for renal lesion characterization with increasing VM imaging energy. The 70-keV setting may provide an optimal trade-off between sensitivity and specificity. © RSNA, 2018 Online supplemental material is available for this article.

Automated Analysis of Unregistered Multi-View Mammograms With Deep Learning.

We describe an automated methodology for the analysis of unregistered cranio-caudal (CC) and medio-lateral oblique (MLO) mammography views in order to estimate the patient's risk of developing breast cancer. The main innovation behind this methodology lies in the use of deep learning models for the problem of jointly classifying unregistered mammogram views and respective segmentation maps of breast lesions (i.e., masses and micro-calcifications). This is a holistic methodology that can classify a whole mammographic exam, containing the CC and MLO views and the segmentation maps, as opposed to the classification of individual lesions, which is the dominant approach in the field. We also demonstrate that the proposed system is capable of using the segmentation maps generated by automated mass and micro-calcification detection systems, and still producing accurate results. The semi-automated approach (using manually defined mass and micro-calcification segmentation maps) is tested on two publicly available data sets (INbreast and DDSM), and results show that the volume under ROC surface (VUS) for a 3-class problem (normal tissue, benign, and malignant) is over 0.9, the area under ROC curve (AUC) for the 2-class "benign versus malignant" problem is over 0.9, and for the 2-class breast screening problem (malignancy versus normal/benign) is also over 0.9. For the fully automated approach, the VUS results on INbreast is over 0.7, and the AUC for the 2-class "benign versus malignant" problem is over 0.78, and the AUC for the 2-class breast screening is 0.86.

Hybrid spectral CT reconstruction.

Current photon counting x-ray detector (PCD) technology faces limitations associated with spectral fidelity and photon starvation. One strategy for addressing these limitations is to supplement PCD data with high-resolution, low-noise data acquired with an energy-integrating detector (EID). In this work, we propose an iterative, hybrid reconstruction technique which combines the spectral properties of PCD data with the resolution and signal-to-noise characteristics of EID data. Our hybrid reconstruction technique is based on an algebraic model of data fidelity which substitutes the EID data into the data fidelity term associated with the PCD reconstruction, resulting in a joint reconstruction problem. Within the split Bregman framework, these data fidelity constraints are minimized subject to additional constraints on spectral rank and on joint intensity-gradient sparsity measured between the reconstructions of the EID and PCD data. Following a derivation of the proposed technique, we apply it to the reconstruction of a digital phantom which contains realistic concentrations of iodine, barium, and calcium encountered in small-animal micro-CT. The results of this experiment suggest reliable separation and detection of iodine at concentrations ≥ 5 mg/ml and barium at concentrations ≥ 10 mg/ml in 2-mm features for EID and PCD data reconstructed with inherent spatial resolutions of 176 µm and 254 µm, respectively (point spread function, FWHM). Furthermore, hybrid reconstruction is demonstrated to enhance spatial resolution within material decomposition results and to improve low-contrast detectability by as much as 2.6 times relative to reconstruction with PCD data only. The parameters of the simulation experiment are based on an in vivo micro-CT experiment conducted in a mouse model of soft-tissue sarcoma. Material decomposition results produced from this in vivo data demonstrate the feasibility of distinguishing two K-edge contrast agents with a spectral separation on the order of the energy resolution of the PCD hardware.

Vertebral Body Compression Fractures and Bone Density: Automated Detection and Classification on CT Images.

Purpose To create and validate a computer system with which to detect, localize, and classify compression fractures and measure bone density of thoracic and lumbar vertebral bodies on computed tomographic (CT) images. Materials and Methods Institutional review board approval was obtained, and informed consent was waived in this HIPAA-compliant retrospective study. A CT study set of 150 patients (mean age, 73 years; age range, 55-96 years; 92 women, 58 men) with (n = 75) and without (n = 75) compression fractures was assembled. All case patients were age and sex matched with control subjects. A total of 210 thoracic and lumbar vertebrae showed compression fractures and were electronically marked and classified by a radiologist. Prototype fully automated spinal segmentation and fracture detection software were then used to analyze the study set. System performance was evaluated with free-response receiver operating characteristic analysis. Results Sensitivity for detection or localization of compression fractures was 95.7% (201 of 210; 95% confidence interval [CI]: 87.0%, 98.9%), with a false-positive rate of 0.29 per patient. Additionally, sensitivity was 98.7% and specificity was 77.3% at case-based receiver operating characteristic curve analysis. Accuracy for classification by Genant type (anterior, middle, or posterior height loss) was 0.95 (107 of 113; 95% CI: 0.89, 0.98), with weighted κ of 0.90 (95% CI: 0.81, 0.99). Accuracy for categorization by Genant height loss grade was 0.68 (77 of 113; 95% CI: 0.59, 0.76), with a weighted κ of 0.59 (95% CI: 0.47, 0.71). The average bone attenuation for T12-L4 vertebrae was 146 HU ± 29 (standard deviation) in case patients and 173 HU ± 42 in control patients; this difference was statistically significant (P < .001). Conclusion An automated machine learning computer system was created to detect, anatomically localize, and categorize vertebral compression fractures at high sensitivity and with a low false-positive rate, as well as to calculate vertebral bone density, on CT images. © RSNA, 2017 Online supplemental material is available for this article.

Characterization of Small Focal Renal Lesions: Diagnostic Accuracy with Single-Phase Contrast-enhanced Dual-Energy CT with Material Attenuation Analysis Compared with Conventional Attenuation Measurements.

Purpose To determine whether single-phase contrast material-enhanced dual-energy material attenuation analysis improves the characterization of small (1-4 cm) renal lesions compared with conventional attenuation measurements by using histopathologic analysis and follow-up imaging as the clinical reference standards. Materials and Methods In this retrospective, HIPAA-compliant, institutional review board-approved study, 136 consecutive patients (95 men and 41 women; mean age, 54 years) with 144 renal lesions (111 benign, 33 malignant) measuring 1-4 cm underwent single-energy unenhanced and contrast-enhanced dual-energy computed tomography (CT) of the abdomen. For each renal lesion, attenuation measurements were obtained; attenuation change of greater than or equal to 15 HU was considered evidence of enhancement. Dual-energy attenuation measurements were also obtained by using iodine-water, water-iodine, calcium-water, and water-calcium material basis pairs. Mean lesion attenuation values and material densities were compared between benign and malignant renal lesions by using the two-sample t test. Diagnostic accuracy of attenuation measurements and dual-energy material densities was assessed and validated by using 10-fold cross-validation to limit the effect of optimistic bias. Results By using cross-validated optimal thresholds at 100% sensitivity, iodine-water material attenuation images significantly improved specificity for differentiating between benign and malignant renal lesions compared with conventional enhancement measurements (93% [103 of 111]; 95% confidence interval: 86%, 97%; vs 81% [90 of 111]; 95% confidence interval: 73%, 88%) (P = .02). Sensitivity with iodine-water and calcium-water material attenuation images was also higher than that with conventional enhancement measurements, although the difference was not statistically significant. Conclusion Contrast-enhanced dual-energy CT with material attenuation analysis improves specificity for characterization of small (1-4 cm) renal lesions compared with conventional attenuation measurements. © RSNA, 2017 Online supplemental material is available for this article.

Polynomial dual energy inverse functions for bone Calcium/Phosphorus ratio determination and experimental evaluation.

An X-ray dual energy (XRDE) method was examined, using polynomial nonlinear approximation of inverse functions for the determination of the bone Calcium-to-Phosphorus (Ca/P) mass ratio. Inverse fitting functions with the least-squares estimation were used, to determine calcium and phosphate thicknesses. The method was verified by measuring test bone phantoms with a dedicated dual energy system and compared with previously published dual energy data. The accuracy in the determination of the calcium and phosphate thicknesses improved with the polynomial nonlinear inverse function method, introduced in this work, (ranged from 1.4% to 6.2%), compared to the corresponding linear inverse function method (ranged from 1.4% to 19.5%).

Benefits of sinogram-affirmed iterative reconstruction in 0.4 mSv ultra-low-dose CT of the upper abdomen following transarterial chemoembolisation: comparison to low-dose and standard-dose CT and filtered back projection technique.

AIM: To evaluate the advantage of sinogram-affirmed iterative reconstruction (SIR) compared to filtered back projection (FBP) in upper abdomen computed tomography (CT) after transarterial chemoembolisation (TACE) at different tube currents. MATERIALS AND METHODS: The study was approved by the institutional review board. Written informed consent was obtained from all patients. Post-TACE CT was performed with different tube currents successively varied in four steps (180, 90, 45 and 23 mAs) with 40 patients per group (mean age: 60±12 years, range: 23-85 years, sex: 70 female, 90 male). The data were reconstructed with standard FBP and five different SIR strengths. Image quality was independently rated by two readers on a five-point scale. High (Lipiodol-to-liver) as well as low (liver-to-fat) contrast-to-noise ratios (CNRs) were intra-individually compared within one dose to determine the optimal strength (S1-S5) and inter-individually between different doses to determine the possibility of dose reduction using the Kruskal-Wallis test. RESULTS: Subjective image quality and objective CNR analysis were concordant: intra-individually, SIR was significantly (p<0.001) superior to FBP. Inter-individually, regarding different doses (180 versus 23 ref mAs), there was no significant (p=1.00) difference when using S5 SIR at 23 mAs instead of FBP. CONCLUSION: SIR allows for an 88% dose reduction from 3.43 to 0.4 mSv in unenhanced CT of the liver following TACE without subjective or objective loss in image quality.

Radiation risk reduction in cardiac electrophysiology through use of a gridless imaging technique.

AIMS: It has been previously demonstrated that use of appropriate frame rates coupled with minimal use of high-dose digital acquisition can limit radiation risk to patients undergoing diagnostic and therapeutic electrophysiology (EP). Imaging without the anti-scatter grid has been proposed as a means of achieving further radiation reduction. We evaluate application of a gridless imaging technique to deliver further reductions in radiation risk to both patients and personnel. METHODS AND RESULTS: Radiation and clinical data for EP procedures performed for 16 months from March 2012 were monitored. The period was divided into three phases: Phase 1 (March 2012-June 2012) provided a performance baseline (radiation output modelling and procedural risk adjustment calibration), Phase 2 (July 2012-September 2012) confirmation of performance with the grid, and Phase 3 (September 2012-June 2013) gridless imaging period. Statistical process control (SPC) charts were used to monitor for changes in radiation use and clinical outcomes (procedural success). Imaging without the grid halved the levels of radiation delivered in undertaking EP procedures. Although there was a perceptible impact on image quality with the grid removed. Review of the SPC chart monitoring procedural outcomes did not identify any discernable adverse impact on success rates. Selected use of the gridless technique is recommended with re-introduction of the grid in larger patients or during aspects of the procedure where image quality is important (e.g. transeptal punctures). CONCLUSION: Use of a gridless imaging technique can contribute to a significant reduction in radiation risk to both patients and operators during cardiac EP procedures.

A minimum spanning forest based classification method for dedicated breast CT images.

PURPOSE: To develop and test an automated algorithm to classify different types of tissue in dedicated breast CT images. METHODS: Images of a single breast of five different patients were acquired with a dedicated breast CT clinical prototype. The breast CT images were processed by a multiscale bilateral filter to reduce noise while keeping edge information and were corrected to overcome cupping artifacts. As skin and glandular tissue have similar CT values on breast CT images, morphologic processing is used to identify the skin based on its position information. A support vector machine (SVM) is trained and the resulting model used to create a pixelwise classification map of fat and glandular tissue. By combining the results of the skin mask with the SVM results, the breast tissue is classified as skin, fat, and glandular tissue. This map is then used to identify markers for a minimum spanning forest that is grown to segment the image using spatial and intensity information. To evaluate the authors' classification method, they use DICE overlap ratios to compare the results of the automated classification to those obtained by manual segmentation on five patient images. RESULTS: Comparison between the automatic and the manual segmentation shows that the minimum spanning forest based classification method was able to successfully classify dedicated breast CT image with average DICE ratios of 96.9%, 89.8%, and 89.5% for fat, glandular, and skin tissue, respectively. CONCLUSIONS: A 2D minimum spanning forest based classification method was proposed and evaluated for classifying the fat, skin, and glandular tissue in dedicated breast CT images. The classification method can be used for dense breast tissue quantification, radiation dose assessment, and other applications in breast imaging.

Reduction of Radiation Exposure in Atrial Fibrillation Ablation Using a New Image Integration Module: A Prospective Randomized Trial in Patients Undergoing Pulmonary Vein Isolation.

INTRODUCTION: Recently, a new image integration module (IIM, CartoUnivu™ Module) has been introduced to combine and merge fluoroscopy images with 3-dimensional-(3D)-electroanatomical maps (Carto® 3 System) into an accurate 3D view. The aim of the study was to investigate the influence of IIM on the fluoroscopy exposure during pulmonary vein isolation (PVI) for paroxysmal atrial fibrillation (PAF) in a prospective randomized trial. METHODS AND RESULTS: Between June and November 2014, a total of 60 patients with PAF (73.3% male, 64.0 ± 9.2 years), who underwent PVI with the endpoint of unexcitability of the ablation line, were randomized to either a conventional 3D mapping system (Carto® 3 System) or to an additional IIM on the basis of an assumed reduction of fluoroscopy exposure by the use of IIM. There were no significant differences in baseline characteristics. The median ablation procedure time was identical in both groups (140.7 ± 27.8 minutes vs. 140.8 ± 39.5 minutes; P = 0.851). A significant decrease of mean fluoroscopy time from 11.9 ± 2.1 to 7.4 ± 2.6 minutes (P < 0.0006) and median fluoroscopy dose from 882.9 to 476.5 cGycm(2) (P < 0.001) was achieved. The main reduction of radiation could be realized during creation of the 3D-map. No major complications occurred during the procedures. After a median follow-up of 125.7 ± 45.6 days 80% of the patients were free from any atrial arrhythmias. CONCLUSION: CartoUnivu™ module easily integrates into the workflow of PVI with the endpoint of unexcitability of the ablation line without prolonging the procedure time. It is associated with a marked reduction in fluoroscopic dose when compared to a conventional 3D mapping system.

Predicting interatrial septum rotation: is the position of the heart or the direction of the coronary sinus reliable?: Implications for interventional electrophysiologists from CT studies.

BACKGROUND: Determining the location of the interatrial septum (IAS) is crucial for cardiac electrophysiology procedures. Empirical methods of predicting IAS orientation depend on anatomical landmarks, including determining it from the direction of the coronary sinus (CS) and the position of the heart (e.g., vertical or transverse). However, the reliability of these methods for predicting IAS rotation warrants further study. The purpose of this study was to assess the clinical utility of the relationship between IAS orientation, CS direction, and heart position. METHODS: Data from 115 patients undergoing coronary computed tomography (CT) angiography with no evidence of cardiac structural disease were collected and analyzed. Angulations describing IAS orientation, CS direction, and heart position were measured. The relationships between IAS orientation and each of the other two parameters were subsequently analyzed. RESULTS: The mean angulations for IAS orientation, CS direction, and heart position were 36.8 ± 7.3° (range 19.1-53.6), 37.7 ± 6.6° (range 21.3-50.1), and 37.1 ± 8.3° (range 19.2-61.0), respectively. We found a significant correlation between IAS orientation and CS direction (r = 0.928; P < 0.01), and the linear regression equation was drawn: IAS orientation = 2.01 + 1.03 × CS direction (r(2) = 0.86). No correlation was observed between IAS orientation and heart position (P = 0.86). CONCLUSION: In patients without structural heart disease, CS direction may be a reliable predictor of IAS orientation, and may serve as a helpful reference for clinicians during invasive electrophysiological procedures. Further study is warranted to clarify the relationship between IAS orientation and heart position.

A breast-specific, negligible-dose scatter correction technique for dedicated cone-beam breast CT: a physics-based approach to improve Hounsfield Unit accuracy.

The purpose of this research was to develop a method to correct the cupping artifact caused from x-ray scattering and to achieve consistent Hounsfield Unit (HU) values of breast tissues for a dedicated breast CT (bCT) system. The use of a beam passing array (BPA) composed of parallel-holes has been previously proposed for scatter correction in various imaging applications. In this study, we first verified the efficacy and accuracy using BPA to measure the scatter signal on a cone-beam bCT system. A systematic scatter correction approach was then developed by modeling the scatter-to-primary ratio (SPR) in projection images acquired with and without BPA. To quantitatively evaluate the improved accuracy of HU values, different breast tissue-equivalent phantoms were scanned and radially averaged HU profiles through reconstructed planes were evaluated. The dependency of the correction method on object size and number of projections was studied. A simplified application of the proposed method on five clinical patient scans was performed to demonstrate efficacy. For the typical 10-18 cm breast diameters seen in the bCT application, the proposed method can effectively correct for the cupping artifact and reduce the variation of HU values of breast equivalent material from 150 to 40 HU. The measured HU values of 100% glandular tissue, 50/50 glandular/adipose tissue, and 100% adipose tissue were approximately 46, -35, and -94, respectively. It was found that only six BPA projections were necessary to accurately implement this method, and the additional dose requirement is less than 1% of the exam dose. The proposed method can effectively correct for the cupping artifact caused from x-ray scattering and retain consistent HU values of breast tissues.