Hybrid Iterative Reconstruction Research Articles

Relative Enhancement Ratio of Portal Venous Phase to Unenhanced CT in the Diagnosis of Lipid-poor Adrenal Adenomas.

Background The development of an accurate, practical, noninvasive, and widely available diagnostic approach to characterize lipid-poor adrenal lesions (greater than 10 HU at unenhanced CT) remains an ongoing demand. Purpose To investigate whether combined assessment of unenhanced and portal venous phase CT allows for the differentiation of lipid-poor adrenal adenomas from nonadenomas. Materials and Methods Patients with lipid-poor adrenal lesions who underwent unenhanced and portal venous phase CT with a single-energy scanner between January 2016 and March 2020 were identified retrospectively. For each lesion, the unenhanced and contrast-enhanced attenuation were measured; the absolute enhancement (contrast-enhanced minus unenhanced attenuation [HU]) and relative enhancement ratio ([absolute enhancement divided by unenhanced attenuation] × 100%) were calculated. The sensitivity achieved at 95% specificity to distinguish adenomas from nonadenomas was determined with receiver operating characteristic curve analysis and compared among parameters with use of the McNemar test. Results A total of 220 patients (mean age ± standard deviation, 66 years ± 12; 134 men) with 131 lipid-poor adenomas and 89 nonadenomas were analyzed. The sensitivity (achieved at 95% specificity) of the relative enhancement ratio (86% [113 of 131 adenomas; 95% CI: 79, 92] at a threshold of >210%) was higher than that of unenhanced attenuation (50% [66 of 131 adenomas; 95% CI: 42, 59] at a threshold of ≤21 HU), contrast-enhanced attenuation (3% [four of 131 adenomas; 95% CI: 1, 8] at a threshold of >120 HU), and absolute enhancement (24% [32 of 131 adenomas; 95% CI: 17, 33] at a threshold of >74 HU; all P < .001). The sensitivities of the relative enhancement ratio were 100% (58 of 58 adenomas; 95% CI: 94, 100), 83% (52 of 63 adenomas; 95% CI: 71, 91), and 30% (three of 10 adenomas; 95% CI: 7, 65) for adenomas measuring unenhanced attenuation of more than 10 HU up to 20 HU, 21-30 HU, and more than 30 HU, respectively. Conclusion A relative enhancement ratio threshold of greater than 210%, measured at unenhanced and portal venous phase CT, accurately differentiated lipid-poor adenomas from nonadenomas, particularly for lesions with unenhanced attenuation of 10-30 HU. © RSNA, 2021 Online supplemental material is available for this article.

Diagnostic performance and image quality of deep learning image reconstruction (DLIR) on unenhanced low-dose abdominal CT for urolithiasis.

Patients with urolithiasis undergo radiation overexposure from computed tomography (CT) scans. Improvement of image reconstruction is necessary for radiation dose reduction. To evaluate a deep learning-based reconstruction algorithm for CT (DLIR) in the detection of urolithiasis at low-dose non-enhanced abdominopelvic CT. A total of 75 patients who underwent low-dose abdominopelvic CT for urolithiasis were retrospectively included. Each examination included three reconstructions: DLIR; filtered back projection (FBP); and hybrid iterative reconstruction (IR; ASiR-V 70%). Image quality was subjectively and objectively assessed using attenuation and noise measurements in order to calculate the signal-to-noise ratio (SNR), absolute contrast, and contrast-to-noise ratio (CNR). Attenuation of the largest stones were also compared. Detectability of urinary stones was assessed by two observers. Image noise was significantly reduced with DLIR: 7.2 versus 17 and 22 for ASiR-V 70% and FBP, respectively. Similarly, SNR and CNR were also higher compared to the standard reconstructions. When the structures had close attenuation values, contrast was lower with DLIR compared to ASiR-V. Attenuation of stones was also lowered in the DLIR series. Subjective image quality was significantly higher with DLIR. The detectability of all stones and stones >3 mm was excellent with DLIR for the two observers (intraclass correlation [ICC] = 0.93 vs. 0.96 and 0.95 vs. 0.99). For smaller stones (<3 mm), results were different (ICC = 0.77 vs. 0.86). For low-dose abdominopelvic CT, DLIR reconstruction exhibited image quality superior to ASiR-V and FBP as well as an excellent detection of urinary stones.

Study on Model Iterative Reconstruction Algorithm vs. Filter Back Projection Algorithm for Diagnosis of Acute Cerebral Infarction Using CT Images.

The aim was to explore the application value of computed tomography (CT) perfusion (CTP) imaging based on the iterative model reconstruction (IMR) in the diagnosis of acute cerebral infarction (ACI). 80 patients with ACI, admitted to hospital, were selected as the research objects and divided randomly into a routine treatment group (group A) and a low-dose group (group B) (each group with 40 patients). Patients in group A were scanned at 80 kV–150 mAs, and the traditional filtered back projection (FBP) algorithm was employed to reconstruct the images; besides, 80 kV–30 mAs was adopted to scan the patients in group B, and the images were reconstructed by IMR1, IMR2, IMR3, iDose4 (a kind of hybrid iterative reconstruction technology), and FBP, respectively. The application values of different algorithms were evaluated by CTP based on the collected CTP images of patients and detecting indicators. The results showed that the gray and white matter CT value, SD value, SNR, CNR, and subjective image scores of patients in group B were basically consistent with those of group A (p > 0.05) after the IMR1 reconstruction, and the CT and SD of gray and white matter in patients from group B reduced steeply (p < 0.05), while SNR and CNR increased dramatically after IMR2 and IMR3 reconstruction in contrast to group A (p < 0.05). Furthermore, the cerebral blood volume (CBV), cerebral blood flow (CBF), mean transit time (MTT) of contrast agent, and time to peak (TTP) of contrast agent in patients from group B after iDose4 and IMR reconstruction were basically the same as those of group A (p > 0.05). Therefore, IMR combined with low-dose CTP could obtain high-quality CTP images of the brain with stable perfusion indicators and low radiation dose, which could be clinically applied in the diagnosis of ACI.

Effect of Different Iterative Reconstruction Algorithms on Ultra-Low Dose CT of Inflammatory Bowel Disease in a Rabbit Model

Purpose: To evaluate the feasibility of ultra-low dose 80 kVp CT for the evaluation of inflammatory bowel disease (IBD) in a rabbit model and to investigate the effect of two different iterative reconstruction (IR) algorithms on 80 kVp CT in terms of image quality and diagnostic performance in comparison with same session conventional 120 kVp images.Materials and Methods: This study was approved by the Animal Care and Use Committee of our hospital. Twenty-eight New Zealand white rabbits were randomly divided into two groups: IBD group (n=18) and a control group (n=10). To create an acute IBD model, 3mL of a 5% w/v tri-nitrobenzene sulfonic acid solution was administered to the sigmoid colon of the rabbits. CT was performed at 80 kVp and 120 kVp and was reconstructed using filtered back projection (FBP), hybrid statistic-based IR, and full IR algorithms for 80 kVp and using FBP only for 120 kVp. Effective radiation dose, image noise, image quality, and diagnostic performance by two reviewers were recorded and compared using repeated measure analysis of variance, McNemar test, and receiver operating curve (ROC) analysis.Results: Mean effective radiation dose of 80 kVp CT (0.05 mSv) was significantly lower than that (0.285 mSv) of 120 kVp CT. Mean image noise was highest in the 80 kVp FBP setting (60.36) but significantly decreased with IR algorithms (47.02 with hybrid IR and 12.92 with full IR) (P&lt;0.0001). Mean overall image quality score was lowest in the 80 kVp FBP setting (1.57 and 1.46 for reviewers 1 and 2, respectively) but significantly improved with IR algorithms (2.43 and 2.25 for hybrid IR and 4.79 and 4.93 for full IR) (P&lt;0.0001). Sensitivity and area under the curve (AUC) for differentiating a normal bowel from IBD was lowest with 80 kVp FBP (61.1% and 83.3%; 0.883 and 0.967) but improved with IR algorithms (83.3%~100%; 0.992~1), to similar levels as the 120 kVp FBP setting (100% and 1). Differences in sensitivity and AUC between 80 kVp FBP and 80 kVp IR algorithms were statistically significant in reviewer 1.Conclusions: Ultra-low dose 80 kVp CT in a rabbit IBD model is not feasible using the standard FBP algorithm. However, with the application of IR algorithms, diagnostic performance of 80 kVp CT was acceptable and on par with that at conventional 120 kVp FBP.

Comparison of image quality between spectral photon-counting CT and dual-layer CT for the evaluation of lung nodules: a phantom study.

To evaluate the image quality (IQ) of a spectral photon-counting CT (SPCCT) using filtered back projection (FBP) and hybrid iterative reconstruction (IR) algorithms (iDose4), in comparison with a dual-layer CT (DLCT) system, and to choose the best image quality according to the IR level for SPCCT. Two phantoms were scanned using a standard lung protocol (120 kVp, 40 mAs) with SPCCT and DLCT systems. Raw data were reconstructed using FBP and 9 iDose4 levels (i1/i2/i3/i4/i5/i6/i7/i9/i11) for SPCCT and 7 for DLCT (i1/i2/i3/i4/i5/i6/i7). Noise power spectrum and task-based transfer function (TTF) were computed. Detectability index (d') was computed for detection of 4 mm ground-glass nodule (GGN) and solid nodule. Two chest radiologists performed an IQ evaluation (noise/nodule sharpness/nodule conspicuity/overall IQ) in consensus, and chose the best image for SPCCT. Noise magnitude was -47% ± 2% lower on average with SPCCT than with DLCT for iDose4 range from i1 to i6. Average NPS spatial frequencies increased for SPCCT in comparison with DLCT. TTF also increased, except for the air insert with FBP, and i1/i2/i3. Higher detectability was found for SPCCT for both GGN and solid nodules. IQ for both types of nodule was rated consistently higher with SPCCT than with DLCT for the same iDose4 level. For SPCCT and both nodules, the scores for noise and conspicuity improved with increasing iDose4 level. iDose4 level 6 provided the best subjective IQ for both types of nodule. Higher IQ for GGN and solid nodules was demonstrated with SPCCT compared with DLCT with better detectability using iDose4. Using spectral photon-counting CT compared with dual-layer CT, noise magnitude was reduced with improvements in spatial resolution and detectability of ground-glass nodules and solid lung nodules. As the iDose4 level increased, noise magnitude was reduced and detectability of ground-glass and solid lung nodules was better for both CT systems. For spectral photon-counting CT imaging, two chest radiologists determined iDose4 level 6 as the best image quality for detecting ground-glass nodules and solid lung nodules.

Development and implementation of an ultralow-dose CT protocol for the assessment of cerebrospinal shunts in adult hydrocephalus

BackgroundCerebrospinal fluid shunts in the treatment of hydrocephalus, although associated with clinical benefit, have a high failure rate with repeat computed tomography (CT) imaging resulting in a substantial cumulative radiation dose. Therefore, we sought to develop a whole-body ultralow-dose (ULD) CT protocol for the investigation of shunt malfunction and compare it with the reference standard, plain radiographic shunt series (PRSS).MethodsFollowing ethical approval, using an anthropomorphic phantom and a human cadaveric ventriculoperitoneal shunt model, a whole-body ULD-CT protocol incorporating two iterative reconstruction (IR) algorithms, pure IR and hybrid IR, including 60% filtered back projection and 40% IR was evaluated in 18 adult patients post new shunt implantation or where shunt malfunction was suspected. Effective dose (ED) and image quality were analysed.ResultsULD-CT permitted a 36% radiation dose reduction (median ED 0.16 mSv, range 0.07–0.17, versus 0.25 mSv (0.06–1.69 mSv) for PRSS (p = 0.002). Shunt visualisation in the thoracoabdominal cavities was improved with ULD-CT with pure IR (p = 0.004 and p = 0.031, respectively) and, in contrast to PRSS, permitted visualisation of the entire shunt course (p < 0.001), the distal shunt entry point and location of the shunt tip in all cases. For shunt complications, ULD-CT had a perfect specificity. False positives (3/22, 13.6%) were observed with PRSS.ConclusionsAt a significantly reduced radiation dose, whole body ULD-CT with pure IR demonstrated diagnostic superiority over PRSS in the evaluation of cerebrospinal fluid shunt malfunction.

Image quality improvement with deep learning-based reconstruction on abdominal ultrahigh-resolution CT: A phantom study.

PurposeIn an ultrahigh‐resolution CT (U‐HRCT), deep learning‐based reconstruction (DLR) is expected to drastically reduce image noise without degrading spatial resolution. We assessed a new algorithm's effect on image quality at different radiation doses assuming an abdominal CT protocol.MethodsFor the normal‐sized abdominal models, a Catphan 600 was scanned by U‐HRCT with 100%, 50%, and 25% radiation doses. In all acquisitions, DLR was compared to model‐based iterative reconstruction (MBIR), filtered back projection (FBP), and hybrid iterative reconstruction (HIR). For the quantitative assessment, we compared image noise, which was defined as the standard deviation of the CT number, and spatial resolution among all reconstruction algorithms.ResultsDeep learning‐based reconstruction yielded lower image noise than FBP and HIR at each radiation dose. DLR yielded higher image noise than MBIR at the 100% and 50% radiation doses (100%, 50%, DLR: 15.4, 16.9 vs MBIR: 10.2, 15.6 Hounsfield units: HU). However, at the 25% radiation dose, the image noise in DLR was lower than that in MBIR (16.7 vs. 26.6 HU). The spatial frequency at 10% of the modulation transfer function (MTF) in DLR was 1.0 cycles/mm, slightly lower than that in MBIR (1.05 cycles/mm) at the 100% radiation dose. Even when the radiation dose decreased, the spatial frequency at 10% of the MTF of DLR did not change significantly (50% and 25% doses, 0.98 and 0.99 cycles/mm, respectively).ConclusionDeep learning‐based reconstruction performs more consistently at decreasing dose in abdominal ultrahigh‐resolution CT compared to all other commercially available reconstruction algorithms evaluated.

Physics-based iterative reconstruction for dual-source and flying focal spot computed tomography.

For single-source helical Computed Tomography (CT), both Filtered-Back Projection (FBP) and statistical iterative reconstruction have been investigated. However, for dual-source CT with flying focal spot (DS-FFS CT), a statistical iterative reconstruction that accurately models the scanner geometry and acquisition physics remains unknown to researchers. Therefore, our purpose is to present a novel physics-based iterative reconstruction method for DS-FFS CT and assess its image quality. Our algorithm uses precise physics models to reconstruct from the native cone-beam geometry and interleaved dual-source helical trajectory of a DS-FFS CT. To do so, we construct a noise physics model to represent data acquisition noise and a prior image model to represent image noise and texture. In addition, we design forward system models to compute the locations of deflected focal spots, the dimension, and sensitivity of voxels and detector units, as well as the length of intersection between x-rays and voxels. The forward system models further represent the coordinated movement between the dual sources by computing their x-ray coverage gaps and overlaps at an arbitrary helical pitch. With the above models, we reconstruct images by an advanced Consensus Equilibrium (CE) numerical method to compute the maximum a posteriori estimate to a joint optimization problem that simultaneously fits all models. We compared our reconstruction with Siemens ADMIRE, which is the clinical standard hybrid iterative reconstruction (IR) method for DS-FFS CT, in terms of spatial resolution, noise profile, and image artifacts through both phantoms and clinical scan datasets. Experiments show that our reconstruction has a higher spatial resolution, with a Task-Based Modulation Transfer Function (MTFtask ) consistently higher than the clinical standard hybrid IR. In addition, our reconstruction shows a reduced magnitude of image undersampling artifacts than the clinical standard. By modeling a precise geometry and avoiding data rebinning or interpolation, our physics-based reconstruction achieves a higher spatial resolution and fewer image artifacts with smaller magnitude than the clinical standard hybrid IR.

Deep learning image reconstruction algorithm for pancreatic protocol dual-energy computed tomography: image quality and quantification of iodine concentration.

To evaluate the image quality and iodine concentration (IC) measurements in pancreatic protocol dual-energy computed tomography (DECT) reconstructed using deep learning image reconstruction (DLIR) and compare them with those of images reconstructed using hybrid iterative reconstruction (IR). The local institutional review board approved this prospective study. Written informed consent was obtained from all participants. Thirty consecutive participants with pancreatic cancer (PC) underwent pancreatic protocol DECT for initial evaluation. DECT data were reconstructed at 70 keV using 40% adaptive statistical iterative reconstruction-Veo (hybrid-IR) and DLIR at medium and high levels (DLIR-M and DLIR-H, respectively). The diagnostic acceptability and conspicuity of PC were qualitatively assessed using a 5-point scale. IC values of the abdominal aorta, pancreas, PC, liver, and portal vein; standard deviation (SD); and coefficient of variation (CV) were calculated. Qualitative and quantitative parameters were compared between the hybrid-IR, DLIR-M, and DLIR-H groups. The diagnostic acceptability and conspicuity of PC were significantly better in the DLIR-M group compared with those in the other groups (p < .001-.001). The IC values of the anatomical structures were almost comparable between the three groups (p = .001-.9). The SD of IC values was significantly lower in the DLIR-H group (p < .001) and resulted in the lowest CV (p < .001-.002) compared with those in the hybrid-IR and DLIR-M groups. DLIR could significantly improve image quality and reduce the variability of IC values than could hybrid-IR. Image quality and conspicuity of pancreatic cancer were the best in DLIR-M. DLIR significantly reduced background noise and improved SNR and CNR. The variability of iodine concentration was reduced in DLIR.

Deep learning image reconstruction for pancreatic low-dose computed tomography: comparison with hybrid iterative reconstruction.

To evaluate image quality, image noise, and conspicuity of pancreatic ductal adenocarcinoma (PDAC) in pancreatic low-dose computed tomography (LDCT) reconstructed using deep learning image reconstruction (DLIR) and compare with those of images reconstructed using hybrid iterative reconstruction (IR). Our institutional review board approved this prospective study. Written informed consent was obtained from all patients. Twenty-eight consecutive patients with PDAC undergoing chemotherapy (14 men and 14 women; mean age, 68.4 years) underwent pancreatic LDCT for therapy evaluation. The LDCT images were reconstructed using 40% adaptive statistical iterative reconstruction-Veo (hybrid-IR) and DLIR at medium and high levels (DLIR-M and DLIR-H). The image noise, diagnostic acceptability, and conspicuity of PDAC were qualitatively assessed using a 5-point scale. CT numbers of the abdominal aorta, portal vein, pancreas, PDAC, background noise, signal-to-noise ratio (SNR) of the anatomical structures, and tumor-to-pancreas contrast-to-noise ratio (CNR) were calculated. Qualitative and quantitative parameters were compared between the hybrid-IR, DLIR-M, and DLIR-H images. CT dose-index volumes and dose-length product in pancreatic LDCT were 2.3 ± 1.0 mGy and 74.9 ± 37.0 mGy•cm, respectively. The image noise, diagnostic acceptability, and conspicuity of PDAC were significantly better in DLIR-H than those in hybrid-IR and DLIR-M (all P < 0.001). The background noise was significantly lower in the DLIR-H images (P < 0.001) and resulted in improved SNRs (P < 0.001) and CNR (P < 0.001) compared with those in the hybrid-IR and DLIR-M images. DLIR significantly reduced image noise and improved image quality in pancreatic LDCT images compared with hybrid-IR.

Feasibility of low-dose contrast media in run-off CT angiography on dual-layer spectral detector CT.

The aim of the present study was to assess the feasibility of applying low-dose contrast media (CM), and to explore the optimal virtual monoenergetic images (VMIs) in run-off computed tomography (CT) angiography (CTA) on dual-layer spectral detector CT (SDCT). Forty patients were randomly assigned into a control group using routine volume CM (group A) and an experimental group using half-volume CM (group B). In groups A and B, 120 kVp polychromatic conventional images were generated via hybrid iterative reconstruction algorithm defined as A1 and B1, respectively. Additionally, in group B, VMIs (range, 40-120 keV) were reconstructed via a spectral reconstruction algorithm defined as B2-B10. Vascular attenuation, noise, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and radiation dose were evaluated. Subjective evaluation was performed using a 5-point scale. The patient demographics and radiation dose demonstrated no significant difference between groups A and B [dose length product (DLP): 1,823.45±512.68 vs. 2,014.40±453.25 mGy·cm, P=0.229; volume CT dose index: 14.92±3.40 vs. 16.26±2.85 mGy, P=0.208; the effective dose (ED): 10.82±3.02 vs. 11.88±2.67 mSv, P=0.229]. The mean vascular attenuation was higher in group B2 (40 keV) and was lower in group B3 (50 keV) in comparison with that in group A1 (487.07±154.21 vs. 414.35±71.66 HU, 329.90±100.25 vs. 414.35±71.66 HU, P>0.05). Compared with group A1, the mean noise was similar in group B2 (40 keV) and was lower in group B1 and groups B3-B10 (50-120 keV) (14.81±5.67 vs. 17.29±4.70 HU, P>0.05; 6.75±1.23-11.26±3.24 vs. 17.29±4.70 HU, P<0.05). The mean SNR and CNR in group B2 (40 keV), as well as the mean SNR in group B3 (50 keV), were significantly higher than those of group A1 (38.21±7.52 vs. 28.25±7.20, 32.70±7.79 vs. 24.54±6.60, 32.85±7.10 vs. 28.25±7.20, P<0.05), and the mean CNR in group B3 (50 keV) was similar to that in group A1 (26.66±7.32 vs. 24.54±6.60, P>0.05). Scores of subjective image quality (IQ) in group B2 (40 keV) and B3 (50 keV) were similar to those in group A1 {5 [4.25, 5] vs. 5 [4, 5], 5 [5, 5] vs. 5 [4, 5], P>0.05}, and showed a declining trend in group B4 (60 keV) {4 [4, 5] vs. 5 [4, 5], P>0.05}. It is feasible to perform run-off CTA using low-dose CM with VMI on SDCT. The VMIs at 40-50 keV were the optimal choice and did not compromise IQ.

Metal artefact reduction in the oral cavity using deep learning reconstruction algorithm in ultra-high-resolution computed tomography: a phantom study.

This study aimed to improve the impact of the metal artefact reduction (MAR) algorithm for the oral cavity by assessing the effect of acquisition and reconstruction parameters on an ultra-high-resolution CT (UHRCT) scanner. The mandible tooth phantom with and without the lesion was scanned using super-high-resolution, high-resolution (HR), and normal-resolution (NR) modes. Images were reconstructed with deep learning-based reconstruction (DLR) and hybrid iterative reconstruction (HIR) using the MAR algorithm. Two dental radiologists independently graded the degree of metal artefact (1, very severe; 5, minimum) and lesion shape reproducibility (1, slight; 5, almost perfect). The signal-to-artefact ratio (SAR), accuracy of the CT number of the lesion, and image noise were calculated quantitatively. The Tukey-Kramer method with a p-value of less than 0.05 was used to determine statistical significance. The HRDLR visual score was better than the NRHIR score in terms of degree of metal artefact (4.6 ± 0.5 and 2.6 ± 0.5, p < 0.0001) and lesion shape reproducibility (4.5 ± 0.5 and 2.9 ± 1.1, p = 0.0005). The SAR of HRDLR was significantly better than that of NRHIR (4.9 ± 0.4 and 2.1 ± 0.2, p < 0.0001), and the absolute percentage error of the CT number in HRDLR was lower than that in NRHIR (0.8% in HRDLR and 23.8% in NRIR). The image noise of HRDLR was lower than that of NRHIR (15.7 ± 1.4 and 51.6 ± 15.3, p < 0.0001). Our study demonstrated that the combination of HR mode and DLR in UHRCT scanner improved the impact of the MAR algorithm in the oral cavity.

Sinogram-based deep learning image reconstruction technique in abdominal CT: image quality considerations.

To investigate the image quality and perception of a sinogram-based deep learningimage reconstruction (DLIR) algorithm for single-energy abdominal CT compared to standard-of-care strength of ASIR-V. In this retrospective study, 50 patients (62% F; 56.74 ± 17.05 years) underwent portal venous phase. Four reconstructions (ASIR-V at 40%, and DLIR at three strengths: low (DLIR-L), medium (DLIR-M), and high (DLIR-H)) were generated. Qualitative and quantitative image quality analysis was performed on the 200 image datasets. Qualitative scores were obtained for image noise, contrast, small structure visibility, sharpness, and artifact by three blinded radiologists on a 5-point scale (1, excellent; 5, very poor). Radiologists also indicated image preference on a 3-point scale (1, most preferred; 3, least preferred). Quantitative assessment was performed by measuring image noise and contrast-to-noise ratio (CNR). DLIR had better image quality scores compared to ASIR-V. Scores on DLIR-H for noise (1.40 ± 0.53), contrast (1.41 ± 0.55), small structure visibility (1.51 ± 0.61), and sharpness (1.60 ± 0.54) were the best (p < 0.05) followed by DLIR-M (1.85 ± 0.52, 1.66 ± 0.57, 1.69 ± 0.59, 1.68 ± 0.46), DLIR-L (2.29 ± 0.58, 1.96 ± 0.61, 1.90 ± 0.65, 1.86 ± 0.46), and ASIR-V (2.86 ± 0.67, 2.55 ± 0.58, 2.34 ± 0.66, 2.01 ± 0.36). Ratings for artifacts were similar for all reconstructions (p > 0.05). DLIRs did not influence subjective textural perceptions and were preferred over ASIR-V from the beginning. All DLIRs had a higher CNR (26.38-102.30%) and lower noise (20.64-48.77%) than ASIR-V. DLIR-H had the best objective scores. Sinogram-based deep learning image reconstructions were preferred over iterative reconstruction subjectively and objectively due to improved image quality and lower noise, even in large patients. Use in clinical routine may allow for radiation dose reduction. • Deep learningimage reconstructions (DLIRs) have a higher contrast-to-noise ratio compared to medium-strength hybrid iterative reconstruction techniques. • DLIR may be advantageous in patients with large body habitus due to a lower image noise. • DLIR can enable further optimization of radiation doses used in abdominal CT.

Model-based reconstruction algorithm in the detection of acute trauma-related lesions in brain CT examinations.

To evaluate the added value of a model-based reconstruction algorithm in the assessment of acute traumatic brain lesions in emergency non-enhanced computed tomography, in comparison with a standard hybrid iterative reconstruction approach. We retrospectively evaluated a total of 350 patients who underwent a 256-row non-enhanced computed tomography scan at the emergency department for brain trauma. Images were reconstructed both with hybrid and model-based iterative algorithm. Two radiologists, blinded to clinical data, recorded the presence, nature, number, and location of acute findings. Subjective image quality was performed using a 4-point scale. Objective image quality was determined by computing the signal-to-noise ratio and contrast-to-noise ratio. The agreement between the two readers was evaluated using k-statistics. A subjective image quality analysis using model-based iterative reconstruction gave a higher detection rate of acute trauma-related lesions in comparison to hybrid iterative reconstruction (extradural haematomas 116 vs. 68, subdural haemorrhages 162 vs. 98, subarachnoid haemorrhages 118 vs. 78, parenchymal haemorrhages 94 vs. 64, contusive lesions 36 vs. 28, diffuse axonal injuries 75 vs. 31; all P<0.001). Inter-observer agreement was moderate to excellent in evaluating all injuries (extradural haematomas k=0.79, subdural haemorrhages k=0.82, subarachnoid haemorrhages k=0.91, parenchymal haemorrhages k=0.98, contusive lesions k=0.88, diffuse axonal injuries k=0.70). Quantitatively, the mean standard deviation of the thalamus on model-based iterative reconstruction images was lower in comparison to hybrid iterative one (2.12 ± 0.92 vsa 3.52 ± 1.10; P=0.030) while the contrast-to-noise ratio and signal-to-noise ratio were significantly higher (contrast-to-noise ratio 3.06 ± 0.55 vs. 1.55 ± 0.68, signal-to-noise ratio 14.51 ± 1.78 vs. 8.62 ± 1.88; P<0.0001). Median subjective image quality values for model-based iterative reconstruction were significantly higher (P=0.003). Model-based iterative reconstruction, offering a higher image quality at a thinner slice, allowed the identification of a higher number of acute traumatic lesions than hybrid iterative reconstruction, with a significant reduction of noise.

Deep learning\u2013based reconstruction may improve non-contrast cerebral CT imaging compared to other current reconstruction algorithms

ObjectivesTo evaluate image quality and reconstruction times of a commercial deep learning reconstruction algorithm (DLR) compared to hybrid-iterative reconstruction (Hybrid-IR) and model-based iterative reconstruction (MBIR) algorithms for cerebral non-contrast CT (NCCT).MethodsCerebral NCCT acquisitions of 50 consecutive patients were reconstructed using DLR, Hybrid-IR and MBIR with a clinical CT system. Image quality, in terms of six subjective characteristics (noise, sharpness, grey-white matter differentiation, artefacts, natural appearance and overall image quality), was scored by five observers. As objective metrics of image quality, the noise magnitude and signal-difference-to-noise ratio (SDNR) of the grey and white matter were calculated. Mean values for the image quality characteristics scored by the observers were estimated using a general linear model to account for multiple readers. The estimated means for the reconstruction methods were pairwise compared. Calculated measures were compared using paired t tests.ResultsFor all image quality characteristics, DLR images were scored significantly higher than MBIR images. Compared to Hybrid-IR, perceived noise and grey-white matter differentiation were better with DLR, while no difference was detected for other image quality characteristics. Noise magnitude was lower for DLR compared to Hybrid-IR and MBIR (5.6, 6.4 and 6.2, respectively) and SDNR higher (2.4, 1.9 and 2.0, respectively). Reconstruction times were 27 s, 44 s and 176 s for Hybrid-IR, DLR and MBIR respectively.ConclusionsWith a slight increase in reconstruction time, DLR results in lower noise and improved tissue differentiation compared to Hybrid-IR. Image quality of MBIR is significantly lower compared to DLR with much longer reconstruction times.Key Points• Deep learning reconstruction of cerebral non-contrast CT results in lower noise and improved tissue differentiation compared to hybrid-iterative reconstruction.• Deep learning reconstruction of cerebral non-contrast CT results in better image quality in all aspects evaluated compared to model-based iterative reconstruction.• Deep learning reconstruction only needs a slight increase in reconstruction time compared to hybrid-iterative reconstruction, while model-based iterative reconstruction requires considerably longer processing time.

Low-dose whole-body CT using deep learning image reconstruction: image quality and lesion detection.

To evaluate image quality and lesion detection capabilities of low-dose (LD) portal venous phase whole-body computed tomography (CT) using deep learning image reconstruction (DLIR). The study cohort of 59 consecutive patients (mean age, 67.2 years) who underwent whole-body LD CT and a prior standard-dose (SD) CT reconstructed with hybrid iterative reconstruction (SD-IR) within one year for surveillance of malignancy were assessed. The LD CT images were reconstructed with hybrid iterative reconstruction of 40% (LD-IR) and DLIR (LD-DLIR). The radiologists independently evaluated image quality (5-point scale) and lesion detection. Attenuation values in Hounsfield units (HU) of the liver, pancreas, spleen, abdominal aorta, and portal vein; the background noise and signal-to-noise ratio (SNR) of the liver, pancreas, and spleen were calculated. Qualitative and quantitative parameters were compared between the SD-IR, LD-IR, and LD-DLIR images. The CT dose-index volumes (CTDIvol) and dose-length product (DLP) were compared between SD and LD scans. The image quality and lesion detection rate of the LD-DLIR was comparable to the SD-IR. The image quality was significantly better in SD-IR than in LD-IR (p < 0.017). The attenuation values of all anatomical structures were comparable between the SD-IR and LD-DLIR (p = 0.28-0.96). However, background noise was significantly lower in the LD-DLIR (p < 0.001) and resulted in improved SNRs (p < 0.001) compared to the SD-IR and LD-IR images. The mean CTDIvol and DLP were significantly lower in the LD (2.9 mGy and 216.2 mGy•cm) than in the SD (13.5 mGy and 1011.6 mGy•cm) (p < 0.0001). LD CT images reconstructed with DLIR enable radiation dose reduction of >75% while maintaining image quality and lesion detection rate and superior SNR in comparison to SD-IR. Deep learning image reconstruction algorithm enables around 80% reduction in radiation dose while maintaining the image quality and lesion detection compared to standard-dose whole-body CT.

Ultra-low-dose thoracic CT with model-based iterative reconstruction (MBIR) in cystic fibrosis patients undergoing treatment with cystic fibrosis transmembrane conductance regulators (CFTR)

To assess the utility of a volumetric low-dose computed tomography (CT) thorax (LDCTT) protocol at a dose equivalent to a posteroanterior (PA) and lateral chest radiograph for surveillance of cystic fibrosis (CF) patients. A prospective study was undertaken of 19 adult patients with CF that proceeded to LDCTT at 12 and 24 months following initiation of ivacaftor. A previously validated seven-section, low-dose axial CT protocol was used for the 12-month study. A volumetric LDCTT protocol was developed for the 24-month study and reconstructed with hybrid iterative reconstruction (LD-ASIR) and pure iterative reconstruction (model-based IR [LD-MBIR]). Radiation dose was recorded for each scan. Image quality was assessed quantitatively and qualitatively, and disease severity was assessed using a modified Bhalla score. Statistical analysis was performed and p-values of <0.05 were considered statistically significant. Volumetric LD-MBIR studies were acquired at a lower radiation dose than the seven-section studies (0.08±0.01 versus 0.10±0.02 mSv; p=0.02). LD-MBIR and seven-section ASIR images had significantly lower levels of image noise compared with LD-ASIR images (p<0.0001). Diagnostic acceptability scores and depiction of bronchovascular structures were found to be acceptable for axial and coronal LD-MBIR images. LD-MBIR images were superior to LD-ASIR images for all qualitative parameters assessed (p<0.0001). No significant change was observed in mean Bhalla score between 1-year and 2-year studies (p=0.84). The use of a volumetric LDCTT protocol (reconstructed with pure IR) enabled acquisition of diagnostic quality CT images, which were considered extremely useful for surveillance of CF patients, at a dose equivalent to a PA and lateral chest radiograph.

Model-based iterative reconstruction for 320-detector row CT angiography reduces radiation exposure in infants with complex congenital heart disease.

We investigated the impact of model-based iterative reconstruction (MBIR) on 320-detector row computed tomography angiography (CTA) in infants with complex congenital heart disease (CHD). Seventy infants with complex CHD who underwent 320-detector row CTA (40 boys and 30 girls; age range, 0-22 months; median age, 60 days) were retrospectively evaluated. First, the images were reconstructed by filtered back projection (FBP), hybrid iterative reconstruction (HIR), or MBIR in 20 cases, and variables were compared among the three iterative reconstruction methods (IR test). Second, the variables were compared between 25 cases scanned using HIR and 25 cases scanned using MBIR, with a 20 standard deviation noise level for both. Attenuation values and contrast-to-noise ratios (CNRs) of the great vessels and heart chambers were calculated. Total dose-length products were recorded for all patients (radiation dose: RD test). In the IR test, the mean CNR values were 4.8±1.3 for FBP, 6.9±1.4 for HIR, and 8.2±1.7 for MBIR (P < 0.0001). The best subjective image qualities in the great vessels and heart chambers were obtained with MBIR. In RD testing, no significant differences between HIR and MBIR in image quality (CNR: HIR, 8.4±2.4; MBIR, 8.3±2.4) were observed. The effective dose was significantly lower for MBIR than for HIR (0.7±0.2 vs. 1.1±0.3 mSv; P < 0.001). The MBIR algorithm significantly improved image quality and decreased radiation exposure in 320-row CTA of infants with complex CHD, providing an alternative to FBP or HIR that is both safer and produces better results.

Improved Reliability of Automated ASPECTS Evaluation Using Iterative Model Reconstruction from Head CT Scans.

Iterative model reconstruction (IMR) has shown to improve computed tomography (CT) image quality compared to hybrid iterative reconstruction (HIR). Alberta Stroke Program Early CT Score (ASPECTS) assessment in early stroke is particularly dependent on high-image quality. Purpose of this study was to investigate the reliability of ASPECTS assessed by humans and software based on HIR and IMR, respectively. Forty-seven consecutive patients with acute anterior circulation large vessel occlusions (LVOs) and successful endovascular thrombectomy were included. ASPECTS was assessed by three neuroradiologists (one attending, two residents) and by automated software in noncontrast axial CT with HIR (iDose4; 5mm) and IMR (5 and 0.9mm). Two expert neuroradiologists determined consensus ASPECTS reading using all available image data including MRI. Agreement between four raters (three humans, one software) and consensus were compared using square-weighted kappa (κ). Human raters achieved moderate to almost perfect agreement (κ=.557-.845) with consensus reading. The attending showed almost perfect agreement for 5mm HIR (κHIR =.845), while residents had mostly substantial agreements without clear trends across reconstructions. Software had substantial to almost perfect agreement with consensus, increasing with IMR 5 and 0.9mm slice thickness (κHIR =.751, κIMR =.777, and κIMR0.9 =.814). Agreements inversely declined for these reconstructions for the attending (κHIR =.845, κIMR =.763, and κIMR0.9 =.681). Human and software rating showed good reliability of ASPECTS across different CT reconstructions. Human raters performed best with the reconstruction algorithms they had most experience with (HIR for the attending). Automated software benefits from higher resolution with better contrasts in IMR with 0.9mm slice thickness.

Deep learning trained algorithm maintains the quality of half-dose contrast-enhanced liver computed tomography images: Comparison with hybrid iterative reconstruction: Study for the application of deep learning noise reduction technology in low dose

PurposeThis study compares the image and diagnostic qualities of a DEep Learning Trained Algorithm (DELTA) for half-dose contrast-enhanced liver computed tomography (CT) with those of a commercial hybrid iterative reconstruction (HIR) method used for standard-dose CT (SDCT). MethodsThis study enrolled 207 adults, and they were divided into two groups: SDCT and low-dose CT (LDCT). SDCT was reconstructed using the HIR method (SDCTHIR), and LDCT was reconstructed using both the HIR method (LDCTHIR) and DELTA (LDCTDL). Noise, Hounsfield unit (HU) values, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were compared between three image series. Two radiologists assessed the noise, artefacts, overall image quality, visualisation of critical anatomical structures and lesion detection, characterisation and visualisation. ResultsThe mean effective doses were 5.64 ± 1.96 mSv for SDCT and 2.87 ± 0.87 mSv for LDCT. The noise of LDCTDL was significantly lower than that of SDCTHIR and LDCTHIR. The SNR and CNR of LDCTDL were significantly higher than those of the other two groups. The overall image quality, visualisation of anatomical structures and lesion visualisation between LDCTDL and SDCTHIR were not significantly different. For lesion detection, the sensitivities and specificities of SDCTHIR vs. LDCTDL were 81.9 % vs. 83.7 % and 89.1 % vs. 86.3 %, respectively, on a per-patient basis. SDCTHIR showed 75.4 % sensitivity and 82.6 % specificity for lesion characterisation on a per-patient basis, whereas LDCTDL showed 73.5 % sensitivity and 82.4 % specificity. ConclusionsLDCT with DELTA had approximately 49 % dose reduction compared with SDCT with HIR while maintaining image quality on contrast-enhanced liver CT.