Deep-Learning Reconstruction of High-Resolution CT Improves Interobserver Agreement for the Evaluation of Pulmonary Fibrosis.

Objective: This study aimed to investigate whether deep-learning reconstruction (DLR) improves interobserver agreement in the evaluation of honeycombing for patients with interstitial lung disease (ILD) who underwent high-resolution computed tomography (CT) compared with hybrid iterative reconstruction (HIR). Methods: In this retrospective study, 35 consecutive patients suspected of ILD who underwent CT including the chest region were included. High-resolution CT images of the unilateral lung with DLR and HIR were reconstructed for the right and left lungs. A radiologist placed regions of interest on the lung and measured standard deviation of CT attenuation (i.e., quantitative image noise). In the qualitative image analyses, 5 blinded readers assessed the presence of honeycombing and reticulation, qualitative image noise, artifacts, and overall image quality using a 5-point scale (except for artifacts which was evaluated using a 3-point scale). Results: The quantitative and qualitative image noise in DLR was remarkably reduced compared to that in HIR (P < .001). Artifacts and overall DLR quality were significantly improved compared to those of HIR (P < .001 for 4 out of 5 readers). Interobserver agreement in the evaluations of honeycombing and reticulation for DLR (0.557 [0.450-0.693] and 0.525 [0.470-0.541], respectively) were higher than those for HIR (0.321 [0.211-0.520] and 0.470 [0.354-0.533], respectively). A statistically significant difference was found for honeycombing (P = .014). Conclusions: DLR improved interobserver agreement in the evaluation of honeycombing in patients with ILD on CT compared to HIR.

Can optimized model-based iterative reconstruction improve the contrast of liver lesions in CT?

BACKGROUND: Computed tomography is a standard imaging procedure for the detection of liver lesions, such as metastases, which can often be small and poorly contrasted, and therefore hard to detect. Advances in image reconstruction have shown promise in reducing image noise and improving low-contrast detectability. PURPOSE: To examine a novel, specialized, model-based iterative reconstruction (MBIR) technique for improved low-contrast liver lesion detection. MATERIAL AND METHODS: Patient images with reported poorly contrasted focal liver lesions were retrospectively reconstructed with the low-contrast attenuating algorithm (FIRST-LCD) from primary raw data. Liver-to-lesion contrast, signal-to-noise, and contrast-to-noise ratios for background and liver noise for each lesion were compared for all three FIRST-LCD presets with the established hybrid iterative reconstruction method (AIDR-3D). An additional visual conspicuity score was given by two experienced radiologists for each lesion. RESULTS: A total of 82 lesions in 57 examinations were included in the analysis. All three FIRST-LCD algorithms provided statistically significant increases in liver-to-lesion contrast, with FIRSTMILD showing the largest increase (40.47 HU in AIDR-3D; 45.84 HU in FIRSTMILD; P < 0.001). Substantial improvement was shown in contrast-to-noise metrics. Visual analysis of the lesions shows decreased lesion visibility with all FIRST methods in comparison to AIDR-3D, with FIRSTSTR showing the closest results (P < 0.001). CONCLUSION: Objective image metrics show promise for MBIR methods in improving the detectability of low-contrast liver lesions; however, subjective image quality may be perceived as inferior. Further improvements are necessary to enhance image quality and lesion detection.

Deep learning image reconstruction for quality assessment of iodine concentration in computed tomography: A phantom study.

BACKGROUND: Recently, deep learning reconstruction (DLR) technology aiming to improve image quality with minimal radiation dose has been applied not only to pediatric scans, but also to computed tomography angiography (CTA). OBJECTIVE: To evaluate image quality characteristics of filtered back projection (FBP), hybrid iterative reconstruction [Adaptive Iterative Dose Reduction 3D (AIDR 3D)], and DLR (AiCE) using different iodine concentrations and scan parameters. METHODS: Phantoms with eight iodine concentrations (ranging from 1.2 to 25.9 mg/mL) located at the edge of a cylindrical water phantom with a diameter of 19 cm were scanned. Data were reconstructed with FBP, AIDR 3D, and AiCE using various scan parameters of tube current and voltage using a 320 row-detector CT scanner. Data obtained using different reconstruction techniques were quantitatively compared by analyzing Hounsfield units (HU), noise, and contrast-to-noise ratios (CNRs). RESULTS: HU values of FBP and AIDR 3D were constant even when the iodine concentration was changed, whereas AiCE showed the highest HU value when the iodine concentration was low, but the HU value reversed when the iodine concentration exceeded a certain value. In the AIDR 3D and AiCE, the noise decreased as the tube current increased, and the change in noise when the iodine concentration was inconsistent. AIDR 3D and AiCE yielded better noise reduction rates than with FBP at a low tube current. The noise reduction rate of AIDR 3D and AiCE compared to that of FBP showed characteristics ranging from 7% to 35%, and the noise reduction rate of AiCE compared to that of AIDR 3D ranged from 2.0% to 13.3%. CONCLUSIONS: The evaluated reconstruction techniques showed different image quality characteristics (HU value, noise, and CNR) according to dose and scan parameters, and users must consider these results and characteristics before performing patient scans.

Optimal slice thickness of brain computed tomography using a hybrid iterative reconstruction algorithm for identifying hyperdense middle cerebral artery sign of acute ischemic stroke.

PURPOSE: To determine the optimal slice thickness of brain non-contrast computed tomography using a hybrid iterative reconstruction algorithm to identify hyperdense middle cerebral artery sign in patients with acute ischemic stroke. METHODS: We retrospectively enrolled 30 patients who had presented hyperdense middle cerebral artery sign and 30 patients who showed no acute ischemic change in acute magnetic resonance imaging. Reformatted axial images at an angle of the orbitomeatal line in slice thicknesses of 0.5, 1, 3, 5, and 7 mm were generated. Optimal slice thickness for identifying hyperdense middle cerebral artery sign was evaluated by a receiver operating characteristics curve analysis and area under the curve (AUC). RESULTS: The mean AUC value of 0.5-mm slice (0.921; 95% confidence interval (95% CI), 0.868 to 0.975) was significantly higher than those of 3-mm (0.791; 95% CI, 0.686 to 0.895; p = 0.041), 5-mm (0.691; 95% CI, 0.583 to 0.799, p < 0.001), and 7-mm (0.695; 95% CI, 0.593 to 0.797, p < 0.001) slices, whereas it was equivalent to that of 1-mm slice (0.901; 95% CI, 0.837 to 0.965, p = 0.751). CONCLUSION: Thin slice thickness of ≤ 1 mm has a better diagnostic performance for identifying hyperdense artery sign on brain non-contrast computed tomography with a hybrid iterative reconstruction algorithm in patients with acute ischemic stroke.

Liquid Calibration Phantoms in Ultra-Low-Dose QCT for the Assessment of Bone Mineral Density.

INTRODUCTION: Cortical bone is affected by metabolic diseases. Some studies have shown that lower cortical bone mineral density (BMD) is related to increases in fracture risk which could be diagnosed by quantitative computed tomography (QCT). Nowadays, hybrid iterative reconstruction-based (HIR) computed tomography (CT) could be helpful to quantify the peripheral bone tissue. A key focus of this paper is to evaluate liquid calibration phantoms for BMD quantification in the tibia and under hybrid iterative reconstruction-based-CT with the different hydrogen dipotassium phosphate (K2HPO4) concentrations phantoms. METHODOLOGY: Four ranges of concentrations of K2HPO4 were made and tested with 2 exposure settings. Accuracy of the phantoms with ash gravimetry and intermediate K2HPO4 concentration as hypothetical patients were evaluated. The correlations and mean differences between measured equivalent QCT BMD and ash density as a gold standard were calculated. Relative percentage error (RPE) in CT numbers of each concentration over a 6-mo period was reported. RESULTS: The correlation values (R2 was close to 1.0), suggested that the precision of QCT-BMD measurements using standard and ultra-low dose settings were similar for all phantoms. The mean differences between QCT-BMD and the ash density for low concentrations (about 93 mg/cm3) were lower than high concentration phantoms with 135 and 234 mg/cm3 biases. In regard to accuracy test for hypothetical patient, RPE was up to 16.1% for the low concentration (LC) phantom for the case of high mineral content. However, the lowest RPE (0.4 to 1.8%) was obtained for the high concentration (HC) phantom, particularly for the high mineral content case. In addition, over 6 months, the K2HPO4 concentrations increased 25% for 50 mg/cm3 solution and 0.7 % for 1300 mg/cm3 solution in phantoms. CONCLUSION: The excellent linear correlations between the QCT equivalent density and the ash density gold standard indicate that QCT can be used with submilisivert radiation dose. We conclude that using liquid calibration phantoms with a range of mineral content similar to that being measured will minimize bias. Finally, we suggest performing BMD measurements with ultra-low dose scan concurrent with iterative-based reconstruction to reduce radiation exposure.

[Effects of Iterative Reconstruction on Image Quality of Pediatric Body Computed Tomography Images].

This study evaluated the effects of three types of hybrid iterative reconstruction (IR) on image quality of pediatric body computed tomography images. The image quality components evaluated were noise power spectrum (NPS), task-based modulation transfer function (TTF), and system performance function (SPF). As the IR strength was increased while reducing the radiation dose, the NPS increased in a low-frequency range and the TTF decreased in low-contrast regions. In the low-contrast regions, the calculated SPF decreased over the entire frequency range. Alternatively, in the high-contrast regions, the SPF decreased in the low-frequency regions and increased in the high-frequency regions. The radiation dose reduction using the hybrid IR resulted in the deterioration of the image quality in the low-contrast regions and changes in the spatial frequency characteristics in the high-contrast regions.

[Effect of Hybrid Iterative Reconstruction on CT Image Quality Using Metal Artifact Reduction].

PURPOSE: This study aimed to evaluate the effect of adaptive iterative dose reduction 3D (AIDR 3D) on the computed tomography (CT) image quality by using single energy metal artifact reduction (SEMAR). MATERIALS & METHODS: A water phantom (22 cmφ) with the stem for total hip arthroplasty made of titanium was scanned. The volume CT dose index (CTDIvol) was set to 8.9 and 5.0 mGy. The reconstruction was performed using filtered back projection and AIDR 3D by soft kernel (FC13) and SEMAR. The averaged profile method was used for the quantitative evaluation of artifacts. We placed a rectangular region-of-interest on the artifact part, and obtained the x-direction averaged profile (Profile A). Profile B was obtained using a water phantom without metal. Profiles A and B were normalized as Profiles A' and B' using the mean value calculated from Profile B. Based on the standard deviation (SD) calculated from Profile B', the background variation level was defined as ±2SD, and subtracted from Profile A' (Profile A″). Finally, the area of Profile A″ was calculated and defined as Artifacttotal. Artifactover, and Artifactunder, respectively, the positive- and negative-side components of Artifacttotal. RESULTS: Both Artifacttotal and Artifactunder increased according to the strength of AIDR 3D. The variations of Artifactover and Artifactunder, due to the AIDR 3D strength, were small and large, respectively. Further, in comparison with a high dose, the effect of artifact emphasis increased at low dose. Therefore, it should be noted that stronger AIDR 3D can emphasize the residual metal artifact.

[Improvement of Image Quality in the Axial Section Using High-resolution Scan Mode and Hybrid Iterative Reconstruction in Ultra-high-resolution Computed Tomography].

The purpose of this study is to compare the physical characteristics and visibility of high-resolution and conventional images acquired with the same X-ray dose, and to investigate the superiority of super high-resolution imaging. A Catphan phantom was scanned in the normal resolution (NR), high-resolution (HR), and super high-resolution (SHR) modes of ultra-high-resolution computed tomography at 120 kV and 75 mAs. All images were reconstructed into a 5-mm thick image slices with filtered back-projection (FBP) and hybrid image reconstruction (HIR), which included normal and enhanced adaptive iterative dose reduction 3D (AIDR and eAIDR, respectively). The modulation transfer function (MTF) and noise power spectrum (NPS) were measured using the circular edge method and radial frequency method, respectively. The signal-to-noise ratio (SNR) was then calculated. High-contrast resolution and low-contrast detectability were evaluated visually by five radiological technologists. The MTFs of HReAIDR and HRFBP images were higher than those of NRFBP images. However, the NPSs of HReAIDR and HRFBP images were larger than those of NRFBP images. The SNR of HReAIDR images was higher than that of NRFBP and HRFBP images. The scores of high-contrast resolution of HReAIDR, NRFBP, and HRFBP images were 13, 8, and 13 cycles/cm, respectively, and the scores of low-contrast detectability were 5, 5, and 6 mm, respectively. Hence, an improvement in high-contrast resolution of signal more than 400 HU in the axial section can be achieved without increasing the radiation dose and decreasing low-contrast detectability with 10 HU using the HR mode and eAIDR.

Optimization of hybrid iterative reconstruction level and evaluation of image quality and radiation dose for pediatric cardiac computed tomography angiography.

BACKGROUND: Hybrid iterative reconstruction can reduce image noise and produce better image quality compared with filtered back-projection (FBP), but few reports describe optimization of the iteration level. OBJECTIVE: We optimized the iteration level of iDose4 and evaluated image quality for pediatric cardiac CT angiography. MATERIALS AND METHODS: Children (n = 160) with congenital heart disease were enrolled and divided into full-dose (n = 84) and half-dose (n = 76) groups. Four series were reconstructed using FBP, and iDose4 levels 2, 4 and 6; we evaluated subjective quality of the series using a 5-grade scale and compared the series using a Kruskal-Wallis H test. For FBP and iDose4-optimal images, we compared contrast-to-noise ratios (CNR) and size-specific dose estimates (SSDE) using a Student's t-test. We also compared diagnostic-accuracy of each group using a Kruskal-Wallis H test. RESULTS: Mean scores for iDose4 level 4 were the best in both dose groups (all P < 0.05). CNR was improved in both groups with iDose4 level 4 as compared with FBP. Mean decrease in SSDE was 53% in the half-dose group. Diagnostic accuracy for the four datasets were in the range 92.6-96.2% (no statistical difference). CONCLUSION: iDose4 level 4 was optimal for both the full- and half-dose groups. Protocols with iDose4 level 4 allowed 53% reduction in SSDE without significantly affecting image quality and diagnostic accuracy.

Reduced-dose abdominopelvic CT using hybrid iterative reconstruction in suspected left-sided colonic diverticulitis.

OBJECTIVES: To assess the effect of hybrid iterative reconstruction (HIR) and filtered back projection (FBP) on abdominopelvic CT with reduced-dose (RD-APCT) in the evaluation of acute left-sided colonic diverticulitis (ALCD). METHODS: Twenty-five consecutive patients with suspected ALCD who underwent RD-APCT (mean CTDIvol 11.2 ± 4.2 mGy) were enrolled in this study. Raw data were reconstructed using FBP and two increasing HIR levels, L4 & L6. Two radiologists assessed image quality, image noise and reviewer confidence in interpreting findings of ALCD, including wall thickening, pericolic fat inflammation, pericolic abscess, and contained or free extraluminal air. Objective image noise (OIN) was measured. RESULTS: OIN was reduced up to 54 % with HIR compared to FBP. Subjective image quality of HIR images was superior to FBP; subjective image noise was reduced. The detection rate of extraluminal air was higher with HIR L6. Reviewer confidence in interpreting CT findings of ALCD significantly improved with application of HIR. CONCLUSIONS: RD-APCT with HIR offers superior image quality and lower image noise compared to FBP, allowing a high level of reviewer confidence in interpreting CT findings in ALCD. HIR facilitates detection of ALCD findings that may be missed with the FBP algorithm. KEY POINTS: HIR significantly reduces objective image noise in comparison to conventional FBP. HIR offers superior subjective image quality in comparison to conventional FBP. HIR allows reduced-dose abdominopelvic CT with acceptable image quality. Reviewer confidence in interpreting CT findings in ALCD significantly improves with HIR.

Improved stent sharpness evaluation with super-resolution deep learning reconstruction in coronary CT angiography.

OBJECTIVES: This study aimed to assess the impact of super-resolution deep learning reconstruction (SR-DLR) on coronary CT angiography (CCTA) image quality and blooming artifacts from coronary artery stents in comparison to conventional methods, including hybrid iterative reconstruction (HIR) and deep learning-based reconstruction (DLR). METHODS: A retrospective analysis included 66 CCTA patients from July to November 2022. Major coronary arteries were evaluated for image noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR). Stent sharpness was quantified using 10%-90% edge rise slope (ERS) and 10%-90% edge rise distance (ERD). Qualitative analysis employed a 5-point scoring system to assess overall image quality, image noise, vessel wall, and stent structure. RESULTS: SR-DLR demonstrated significantly lower image noise compared to HIR and DLR. SNR and CNR were notably higher in SR-DLR. Stent ERS was significantly improved in SR-DLR, with mean ERD values of 0.70 ± 0.20 mm for SR-DLR, 1.13 ± 0.28 mm for HIR, and 0.85 ± 0.26 mm for DLR. Qualitatively, SR-DLR scored higher in all categories. CONCLUSIONS: SR-DLR produces images with lower image noise, leading to improved overall image quality, compared with HIR and DLR. SR-DLR is a valuable image reconstruction algorithm for enhancing the spatial resolution and sharpness of coronary artery stents without being constrained by hardware limitations. ADVANCES IN KNOWLEDGE: The overall image quality was significantly higher in SR-DLR, resulting in sharper coronary artery stents compared to HIR and DLR.

[Physical Properties of Small Focal Spot Imaging with Deep Learning Reconstruction in Chest-abdominal Plain CT].

PURPOSE: The aim of this study was to compare the physical properties of small focal spot imaging with deep learning reconstruction (DLR) and small or large focal spot imaging with hybrid iterative reconstruction (IR) in chest-abdominal plain computed tomography. METHOD: In small focal spot imaging using DLR and hybrid IR, tube currents were set at 350 mA. For the large focal spot imaging using hybrid IR, the tube current was set at 360, 400, 450, and 500 mA. The spatial frequencies with 50% task transfer function (TTF) for delrin and acrylic were calculated to compare spatial resolution properties for lung and soft tissue in the chest. Additionally, the low-contrast object-specific contrast-to-noise ratio (CNRLO) was measured as noise property was measured for a 7-mm module with a CT value contrast of 10 HU in the abdomen. RESULT: Spatial frequencies with 50% TTF for delrin and acrylic were found to be greater in small focal spot imaging using DLR compared to those in small and large focal spot imaging using hybrid IR. Moreover, the CNRLO obtained from small focal spot imaging with DLR was also nearly equivalent to that of large focal spot imaging with hybrid IR at tube currents of 450 and 500 mA. CONCLUSION: In chest-abdominal plain computed tomography, small focal spot imaging with DLR has been demonstrated to exhibit greater spatial resolution properties compared to small and large focal spot imaging with hybrid IR, with equivalent or better noise performance.

The use of artificial intelligence and deep learning reconstruction in urological computed tomography: Dose reduction at ghost level.

Objective: The objective of the study is to demonstrate that with the use of artificial intelligence (AI) in computed tomography (CT), radiation doses of CT kidney-ureter-bladder (KUB) and CT urogram (CTU) can be reduced to less than that of X-ray KUB and CT KUB, respectively, while maintaining the good image quality. Materials and Methods: We reviewed all CT KUBs (n = 121) performed in September 2019 and all CTUs (n = 74) performed in December 2019 at our institution. The dose length product (DLP) of all CT KUBs and each individual phase of CTU were recorded. DLP of each scan done with new scanner (Canon Aquilion One Genesis with AiCE [CAOG]) which uses AI and deep learning reconstruction (DLR) were compared against traditional non-AI scanner (GE OPTIMA 660 [GEO-660]). We also compared DLPs of both scanners against the United Kingdom, National Diagnostic Reference Levels (NDRL) for CT. Results: One hundred and twenty-one patient's CT KUBs and 74 patient's CTUs were reviewed. For CT KUB group, the mean DLP of 81/121 scans done using AI/DLR scanner (CAOG) was 77.8 mGy cm (1.16 mSv), while the mean DLP of 40/121 CT KUB done with GEO-660 was 317.1 mGy cm (4.75 mSv). For CTU group, the mean DLP for 46/74 scans done using AI/DLR scanner (CAOG) was 401.9 mGy cm (6 mSv), compared to mean DLP of 1352.6 mGy cm (20.2 mSv) from GEO-660. Conclusion: We propose that CT scanners using AI/DLR method have the potential of reducing radiation doses of CT KUB and CTU to such an extent that it heralds the extinction of plain film XR KUB for follow-up of urinary tract stones. To the best of our knowledge, this is the first study comparing CT KUB and CTU doses from new scanners utilizing AI/DLR technology with traditional scanners using hybrid iterative reconstruction technology. Moreover, we have shown that this technology can markedly reduce the cumulative radiation burden in all urological patients undergoing CT examinations, whether this is CT KUB or CTU.

Cycle-consistent learning-based hybrid iterative reconstruction for whole-body PET imaging.

Objective. To develop a cycle-consistent learning-based hybrid iterative reconstruction (IR) method that takes only slightly longer than analytic reconstruction, while pursuing the image resolution and tumor quantification achievable by IR for whole-body PET imaging.Approach. We backproject the raw positron emission tomography (PET) data to generate a blurred activity distribution. From the backprojection to the IR label, a reconstruction mapping that approximates the deblurring filters for the point spread function and the physical effects of the PET system is unrolled to a neural network with stacked convolutional layers. By minimizing the cycle-consistent loss, we train the reconstruction and inverse mappings simultaneously.Main results. In phantom study, the proposed method results in an absolute relative error (RE) of the mean activity of 4.0% ± 0.7% in the largest hot sphere, similar to the RE of the full-count IR and significantly smaller than that obtained by CycleGAN postprocessing. Achieving a noise reduction of 48.1% ± 0.5% relative to the low-count IR, the proposed method demonstrates advantages over the low-count IR and CycleGAN in terms of resolution maintenance, contrast recovery, and noise reduction. In patient study, the proposed method obtains a noise reduction of 44.6% ± 8.0% for the lung and the liver, while maintaining the regional mean activity in both simulated lesions and real tumors. The run time of the proposed method is only half that of the conventional IR.Significance. The proposed cycle-consistent learning from the backprojection rather than the raw PET data or an IR result enables improved reconstruction accuracy, reduced memory requirements, and fast implementation speeds for clinical whole-body PET imaging.

Effects of contrast enhancement boost postprocessing technique in combination with different reconstruction algorithms on the image quality of abdominal CT angiography.

OBJECTIVES: To investigate whether contrast-enhancement-boost (CE-boost) in combination with hybrid iterative reconstruction (Hybrid IR, also named HIR [AIDR 3D, adaptive iterative dose reduction three dimensional]) and model-based iterative reconstruction (MBIR [FIRST, forward projected model-based IR solution]) algorithms can improve the image quality of abdominal CT angiography (CTA). METHODS: This retrospective study included 50 patients who underwent abdominal CTA. Both arterial and portal phases were reconstructed using three different algorithms [filtered-back projection (FBP), AIDR 3D, and FIRST] separately. CE-boost was performed additionally to generate AIDR 3D-boost and FIRST-boost images. Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the arteries and portal system were compared among the five datasets (FBP, AIDR 3D, FIRST, AIDR 3D-boost, FIRST-boost). In subjective analyses, two radiologists independently ordered images (5, best; 1, worst) based on the visual image quality of distal arterial or portal venous branches. The Friedman and the Dunn-Bonferroni post-hoc tests were used for statistical analysis. RESULTS: FIRST-boost arterial and portal images had the lowest noise compared with FBP, AIDR 3D, FIRST, and AIDR 3D-boost images (all P < 0.05), and significantly higher SNR and CNR than FBP, AIDR 3D, and FIRST images (all P < 0.05). AIDR 3D-boost images showed lower noise, and higher SNR and CNR than FBP and AIDR 3D images (all P < 0.05). FIRST-boost images had higher subjective grading scores than FBP, AIDR 3D, and AIDR 3D-boost images (all P < 0.05). CONCLUSION: The postprocessing technique CE-boost can improve the image quality of abdominal CTA images. MBIR in combination with CE-boost (FIRST-boost) images had the best image quality compared with the other four image datasets.

Impact of noise reduction on radiation dose reduction potential of virtual monochromatic spectral images: Comparison of phantom images with conventional 120 kVp images using deep learning image reconstruction and hybrid iterative reconstruction.

PURPOSE: To assess the effects of deep learning image reconstruction (DLIR) and hybrid iterative reconstruction (HIR) on the image quality of virtual monochromatic spectral (VMS) images and to investigate the dose reduction potential of the VMS and conventional 120 kVp images. METHODS: A cylindrical phantom simulating an adult abdomen was used. The contrast was set to 60 (medium) and 300 (high) Hounsfield units. CT acquisitions were performed at three dose levels: 12, 9, and 6 mGy. Images were reconstructed via filtered back projection (FBP), DLIR, and HIR. The noise power spectrum (NPS) and task transfer function (TTF) were measured, and the system performance (SP) function was calculated (TTF2/NPS). RESULTS: The noise magnitudes at low spatial frequencies with DLIR and HIR were lower than that with FBP by 45.6% and 24.4%, respectively. Compared to the FBP results, the TTF values at 50% with DLIR at medium and high contrast changed by -13.2% and +25.3% with the VMS images and -2.0% and +9.3% with the 120 kVp images, respectively. In the VMS and 120 kVp images, compared to the SP values of 12 mGy FBP images, SP values of 6 mGy DLIR images decreased at medium contrast and increased at high contrast. CONCLUSIONS: DLIR achieved better noise reduction than HIR. The spatial resolution of VMS-DLIR varied significantly depending on the contrast. The image quality of VMS-DLIR and 120 kVp-DLIR potentially decrease in medium contrast tasks and increase in high contrast tasks with 50% dose reduction.

Improvement in Image Quality and Visibility of Coronary Arteries, Stents, and Valve Structures on CT Angiography by Deep Learning Reconstruction.

OBJECTIVE: This study aimed to investigate whether a deep learning reconstruction (DLR) method improves the image quality, stent evaluation, and visibility of the valve apparatus in coronary computed tomography angiography (CCTA) when compared with filtered back projection (FBP) and hybrid iterative reconstruction (IR) methods. MATERIALS AND METHODS: CCTA images of 51 patients (mean age ± standard deviation [SD], 63.9 ± 9.8 years, 36 male) who underwent examination at a single institution were reconstructed using DLR, FBP, and hybrid IR methods and reviewed. CT attenuation, image noise, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and stent evaluation, including 10%-90% edge rise slope (ERS) and 10%-90% edge rise distance (ERD), were measured. Quantitative data are summarized as the mean ± SD. The subjective visual scores (1 for worst -5 for best) of the images were obtained for the following: overall image quality, image noise, and appearance of stent, vessel, and aortic and tricuspid valve apparatus (annulus, leaflets, papillary muscles, and chordae tendineae). These parameters were compared between the DLR, FBP, and hybrid IR methods. RESULTS: DLR provided higher Hounsfield unit (HU) values in the aorta and similar attenuation in the fat and muscle compared with FBP and hybrid IR. The image noise in HU was significantly lower in DLR (12.6 ± 2.2) than in hybrid IR (24.2 ± 3.0) and FBP (54.2 ± 9.5) (p < 0.001). The SNR and CNR were significantly higher in the DLR group than in the FBP and hybrid IR groups (p < 0.001). In the coronary stent, the mean value of ERS was significantly higher in DLR (1260.4 ± 242.5 HU/mm) than that of FBP (801.9 ± 170.7 HU/mm) and hybrid IR (641.9 ± 112.0 HU/mm). The mean value of ERD was measured as 0.8 ± 0.1 mm for DLR while it was 1.1 ± 0.2 mm for FBP and 1.1 ± 0.2 mm for hybrid IR. The subjective visual scores were higher in the DLR than in the images reconstructed with FBP and hybrid IR. CONCLUSION: DLR reconstruction provided better images than FBP and hybrid IR reconstruction.

Quantitative measurements of emphysema in ultra-high resolution computed tomography using model-based iterative reconstruction in comparison to that using hybrid iterative reconstruction.

The percentage of low attenuation volume ratio (LAVR), which is measured using computed tomography (CT), is an index of the severity of emphysema. For LAVR evaluation, ultra-high-resolution (U-HR) CT images are useful. To improve the image quality of U-HRCT, iterative reconstruction is used. There are two types of iterative reconstruction: hybrid iterative reconstruction (HIR) and model-based iterative reconstruction (MBIR). In this study, we physically and clinically evaluated U-HR images reconstructed with HIR and MBIR, and demonstrated the usefulness of U-HR images with MBIR for quantitative measurements of emphysema. Both images were reconstructed with a slice thickness of 0.25 mm and an image matrix size of 1024 × 1024 pixels. For physical evaluation, the modulation transfer function (MTF) and noise power spectrum (NPS) of HIR and MBIR were compared. For clinical evaluation, LAVR calculated from HIR and MBIR were compared using the Wilcoxon matched-pairs signed-rank test. In addition, the correlation between LAVR and forced expiratory volume in one second (FEV1%) was evaluated using the Spearman rank correlation test. The MTFs of HIR and MBIR were comparable. The NPS of MBIR was lower than that of HIR. The mean LAVR values calculated from HIR and MBIR were 19.5 ± 12.6% and 20.4 ± 11.7%, respectively (p = 0.84). The correlation coefficients between LAVR and FEV1% that were taken from HIR and MBIR were 0.64 and 0.74, respectively (p < 0.01). MBIR is more useful than HIR for the quantitative measurements of emphysema with U-HR images.

Improvement of Image Quality Using Hybrid Iterative Reconstruction with Noise Power Spectrum Model in Computed Tomography During Hepatic Arteriography.

OBJECTIVES: In CT during hepatic arteriography (CTHA), the addition of a noise power spectrum (NPS) model to conventional hybrid iterative reconstruction (HIR) may improve spatial resolution and reduce image noise. This study aims at assessing the image quality provided by HIR with a NPS model at CTHA. METHODS: This institutional review board-approved retrospective analysis included 26 patients with hepatocellular carcinomas (HCCs) who underwent CTHA. In all acquisitions, images were reconstructed with filtered back projection (FBP), adaptive iterative dose reduction 3D (AIDR), and AIDR enhanced (eAIDR) with the NPS model. Four radiologists analyzed the signal-to-noise ratio (SNR) of HCC nodules and its associated feeding arteries. The radiologists used a semiquantitative scale (-3 to +3) to rate the subjective image quality comparing both the FBP and eAIDR images with the AIDR images. RESULTS: The feeding arteries' attenuation was significantly higher in eAIDR compared to AIDR [514.3 ± 121.4 and 448.3 ± 107.3 Hounsfield units (HU), p < 0.05]. The image noise of eAIDR was significantly lower than that of FBP (15.2 ± 2.2 and 28.5 ± 4.8 HU, p < 0.05) and comparable to that of AIDR. The SNR of feeding arteries on eAIDR was significantly higher than on AIDR (34.1 ± 7.9 and 27.4 ± 6.3, p < 0.05). Subjective assessment scores showed that eAIDR provided better visibility of feeding arteries and overall image quality compared to AIDR (p < 0.05). The HCC nodule visibility was not significantly different among the three reconstructions. CONCLUSION: In CTHA, eAIDR improved the visibility of feeding arteries associated with HCC nodules without compromising nodule detection.

Image quality of ct angiography using model-based iterative reconstruction in infants with congenital heart disease: Comparison with filtered back projection and hybrid iterative reconstruction.

PURPOSE: To compare the image quality, rate of coronary artery visualization and diagnostic accuracy of 256-slice multi-detector computed tomography angiography (CTA) with prospective electrocardiographic (ECG) triggering at a tube voltage of 80kVp between 3 reconstruction algorithms (filtered back projection (FBP), hybrid iterative reconstruction (iDose4) and iterative model reconstruction (IMR)) in infants with congenital heart disease (CHD). METHODS: Fifty-one infants with CHD who underwent cardiac CTA in our institution between December 2014 and March 2015 were included. The effective radiation doses were calculated. Imaging data were reconstructed using the FBP, iDose4 and IMR algorithms. Parameters of objective image quality (noise, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR)); subjective image quality (overall image quality, image noise and margin sharpness); coronary artery visibility; and diagnostic accuracy for the three algorithms were measured and compared. RESULTS: The mean effective radiation dose was 0.61±0.32 mSv. Compared to FBP and iDose4, IMR yielded significantly lower noise (P<0.01), higher SNR and CNR values (P<0.01), and a greater subjective image quality score (P<0.01). The total number of coronary segments visualized was significantly higher for both iDose4 and IMR than for FBP (P=0.002 and P=0.025, respectively), but there was no significant difference in this parameter between iDose4 and IMR (P=0.397). There was no significant difference in the diagnostic accuracy between the FBP, iDose4 and IMR algorithms (χ2=0.343, P=0.842). CONCLUSIONS: For infants with CHD undergoing cardiac CTA, the IMR reconstruction algorithm provided significantly increased objective and subjective image quality compared with the FBP and iDose4 algorithms. However, IMR did not improve the diagnostic accuracy or coronary artery visualization compared with iDose4.