Ultra-high-resolution CT of the temporal bone: Comparison between deep learning reconstruction and hybrid and model-based iterative reconstruction.

PURPOSE: The purpose of this study was to evaluate the ability of ultra-high-resolution computed tomography (UHR-CT) to assess stapes and chorda tympani nerve anatomy using a deep learning (DLR), a model-based, and a hybrid iterative reconstruction algorithm compared to simulated conventional CT. MATERIALS AND METHODS: CT acquisitions were performed with a Mercury 4.0 phantom. Images were acquired with a 1024 × 1024 matrix and a 0.25 mm slice thickness and reconstructed using DLR, model-based, and hybrid iterative reconstruction algorithms. To simulate conventional CT, images were also reconstructed with a 512 × 512 matrix and a 0.5 mm slice thickness. Spatial resolution, noise power spectrum, and objective high-contrast detectability were compared. Three radiologists evaluated the clinical acceptability of these algorithms by assessing the thickness and image quality of the stapes footplate and superstructure elements, as well as the image quality of the chorda tympani nerve bony and tympanic segments using a 5-point confidence scale on 13 temporal bone CT examinations reconstructed with the four algorithms. RESULTS: UHR-CT provided higher spatial resolution than simulated conventional CT at the penalty of higher noise. DLR and model-based iterative reconstruction provided better noise reduction than hybrid iterative reconstruction, and DLR had the highest detectability index, regardless of the dose level. All stapedial structure thicknesses were thinner using UHR-CT by comparison with conventional simulated CT (P < 0.009). DLR showed the best visualization scores compared to the other reconstruction algorithms (P < 0.032). CONCLUSION: UHR-CT with DLR results in less noise than UHR-CT with hybrid iterative reconstruction and significantly improves stapes and tympanic chorda tympani nerve depiction compared to simulated conventional CT and UHR-CT with iterative reconstruction.

Spectral CT imaging: Technical principles of dual-energy CT and multi-energy photon-counting CT.

Spectral computed tomography (CT) imaging encompasses a unique generation of CT systems based on a simple principle that makes use of the energy-dependent information present in CT images. Over the past two decades this principle has been expanded with the introduction of dual-energy CT systems. The first generation of spectral CT systems, represented either by dual-source or dual-layer technology, opened up a new imaging approach in the radiology community with their ability to overcome the limitations of tissue characterization encountered with conventional CT. Its expansion worldwide can also be considered as an important leverage for the recent groundbreaking technology based on a new chain of detection available on photon counting CT systems, which holds great promise for extending CT towards multi-energy CT imaging. The purpose of this article was to detail the basic principles and techniques of spectral CT with a particular emphasis on the newest technical developments of dual-energy and multi-energy CT systems.

Impact of Simulation-Based Training on Radiation Exposure of Young Interventional Cardiologists.

Reducing radiation exposure during cardiovascular catheterization is of paramount importance to ensure patient and staff safety. Our study aimed to assess the transferability of acquired skills from virtual reality to the real world, including radioprotection measures during mentored simulation training (ST) in coronary angiography. A total of 10 cardiology residents were evaluated during real-life cases in the catheterization laboratory before (group A) and after mentored ST. The educational effect of mentored simulator training on real-life case performance was evaluated at 2 different time points: within the first week (group B) and after 12 weeks (group C). Compared with group A, the total dose area product (DAP) (µGy•m2) and total air kerma (mGy) were lower after ST: group A: 2,633 (1,723 to 3,617) versus group B: 1,618 (1,032 to 2,562), p <0.05 and 214 (136 to 297) versus 135 (84 to 222), p <0.05, respectively. Concerning operator radiation exposure (µSv), left finger dose: 1,090 (820 to 1,460) versus 635 (300 to 900), p = 0.028; left leg dose 80 (0 to 110) versus 0 (0 to 0), p = 0.027; left eye lens dose: 39 (24 to 69) versus 11 (8 to 20), p <0.0001; and chest dose outside the lead apron: 50 (34 to 88) versus 29 (21 to 50), p <0.003 were significantly lower in the group B than group A. A total of 12 weeks after ST, the total DAP and total air kerma remained stable along with operator exposure except left eye lens dose (µSv): group B: 11 (8 to 20) versus group C: 16 (12 to 27), p = 0.02. In addition, left eye lens dose, left wrist dose, and chest dose outside the lead apron were significantly correlated with total DAP (rs = 0.635, rs = 0.729, and rs = 0, 629, respectively) and total air kerma (rs = 0.488, rs = 0.514, and rs = 0.548, respectively) at 12 weeks. In conclusion, ST for coronary angiography may improve radioprotection learning and should be incorporated into training curricula.

Clinical acceptance of deep learning reconstruction for abdominal CT imaging: objective and subjective image quality and low-contrast detectability assessment.

OBJECTIVE: To evaluate the image quality and clinical acceptance of a deep learning reconstruction (DLR) algorithm compared to traditional iterative reconstruction (IR) algorithms. METHODS: CT acquisitions were performed with two phantoms and a total of nine dose levels. Images were reconstructed with two types of IR algorithms, DLR and filtered-back projection. Spatial resolution, image texture, mean noise value, and objective and subjective low-contrast detectability were compared. Ten senior radiologists evaluated the clinical acceptance of these algorithms by scoring ten CT exams reconstructed with the DLR and IR algorithms evaluated. RESULTS: Compared to MBIR, DLR yielded a lower noise and a higher low-contrast detectability index at low doses (CTDIvol ≤ 2.2 and ≤ 4.5 mGy, respectively). Spatial resolution and detectability at higher doses were better with MBIR. Compared to HIR, DLR yielded a higher spatial resolution, a lower noise, and a higher detectability index. Despite these differences in algorithm performance, significant differences in subjective low-contrast performance were not found (p ≥ 0.005). DLR texture was finer than that of MBIR and closer to that of HIR. Radiologists preferred DLR images for all criteria assessed (p < 0.0001), whereas MBIR was rated worse than HIR (p < 0.0001) in all criteria evaluated, except for noise (p = 0.044). DLR reconstruction time was 12 times faster than that of MBIR. CONCLUSION: DLR yielded a gain in objective detection and noise at lower dose levels with the best clinical acceptance among the evaluated reconstruction algorithms. KEY POINTS: • DLR yielded improved objective low-contrast detection and noise at lower dose levels. • Despite the differences in objective detectability among the algorithms evaluated, there were no differences in subjective detectability. • DLR presented significantly higher clinical acceptability scores compared to MBIR and HIR.

Ultra-high resolution computed tomography of joints: practical recommendations for acquisition protocol optimization.

BACKGROUND: To assess the influence on the spatial resolution of various Ultra-high-resolution computed tomography (CT) parameters and provide practical recommendations for acquisition protocol optimization in musculoskeletal imaging. METHODS: All acquisitions were performed with an Ultra-high resolution scanner, and variations of the following parameters were evaluated: field-of-view (150-300 mm), potential (80-140 KVp), current (25-250 mAs), focal spot size (0.4×0.5 to 0.8×1.3 mm2), slice thickness (0.25-0.5 mm), reconstruction matrix (512×512 to 2048×2048), and iso-centering (up to 85 mm off-center). Two different image reconstruction algorithms were evaluated: hybrid iterative reconstruction (HIR) and model-based iterative reconstruction (MBIR). CATPHAN 600 phantom images were analyzed to calculate the number of visible line pairs per centimeter (lp/cm). Task transfer function (TTF) curves were calculated to quantitatively evaluate spatial resolution. Cadaveric knee acquisitions were also performed. RESULTS: Under the conditions studied, the factor that most intensely influenced spatial resolution was the matrix size (additional visualization of up to 8 lp/cm). Increasing the matrix from 512×512 to 2048×2048 led to a 28.2% increase in TTF10% values with a high-dose protocol and a 5.6% increase with a low-dose protocol with no change in the number of visually distinguishable line pairs. The second most important factor affecting spatial resolution was the tube output (29.6% TTF10% gain and 5 additional lp/cm visualized), followed by the reconstruction algorithm choice and lateral displacement (both with a 4 lp/cm gain). Decreasing the slice thickness from 0.5 to 0.25 mm, led to an increase of 3 lp/cm (from 17 to 20 lp/cm) and a 17.3% increase in TTF10% values with no change in the "in-plane" spatial resolution. CONCLUSIONS: This study provides practical recommendations for spatial resolution optimization using Ultra-high-resolution CT.

Metal artifact reduction for small metal implants on CT: Which image reconstruction algorithm performs better?

PURPOSE: To compare the performance of different image reconstruction algorithms in the presence of small metal objects of different sizes and at different dose levels. METHOD: A fresh bone of bovine femur was drilled with seven drill bits of increasing diameter. CT images with eight different dose levels were acquired and reconstructed with three algorithms: hybrid iterative reconstruction - HIR, Full model-based iterative reconstruction - full MBIR and a single energy metal artifact reduction - SEMAR. Trabecular distortion adjacent to metal was evaluated subjectively with a four-point scale. Edge profile artifacts were evaluated quantitatively by measuring drill bit diameter overestimation and the width of the low-density halo surrounding the drill bit. RESULTS: Trabecular distortion was higher with full MBIR compared to HIR and SEMAR (P < 0.0001) and increased with drill bits larger than 1.2 mm and with doses lower than 18.1 mGy.cm. Low-density halos size and drill bit diameter overestimation decreased with full MBIR compared to the other two reconstruction algorithms and with SEMAR compared to HIR (P < 0.0001). There was a mean drill bit overestimation of 0.56 ± 0.25 mm for full MBIR versus 0.68 ± 0.09 mm for SEMAR and mean low-density halo diameters of 0.03 mm ± 0.08 for full MBIR versus 0.42 mm ± 0.09 for SEMAR. CONCLUSION: Algorithm performance is influenced by dose levels and metal object size and no individual algorithm provides the best overall performance. Full MBIR is better in reducing edge artifacts and SEMAR is the best option for larger metal implants and low dose protocols.

Full model-based iterative reconstruction (MBIR) in abdominal CT increases objective image quality, but decreases subjective acceptance.

OBJECTIVE: Evaluate and compare the image quality and acceptance of a full MBIR algorithm to that of an earlier full IR hybrid algorithm and filtered back projection (FBP). METHODS: Acquisitions were performed with a 320 detector-row CT scanner with seven different dose levels. Images were reconstructed with three algorithms: FBP, full hybrid iterative reconstruction (HIR), and a full model-based iterative reconstruction algorithm (full MBIR). The sensitometry, spatial resolution, image texture, and low-contrast detectability of these algorithms were compared. Subjective analysis of low-contrast detectability was performed. Ten radiologists answered a questionnaire on image quality and confidence in full MBIR images in clinical practice. RESULTS: The contrast-to-noise ratio of full MBIR was significantly higher than in the other algorithms (p < 0.0015). The spatial resolution was also higher with full MBIR at high frequencies (> 0.3 lp/mm). Full MBIR at low dose levels led to better low-contrast detectability and more inserts being identified with a higher confidence (p < 0.0001). Full MBIR was associated with a change in image texture compared to HIR and FBP. Eighty percent of radiologists judged general appearance and texture of full MBIR images worse than HIR. Moreover, compared with HIR, for 50% of radiologists, the diagnostic confidence on full MBIR images was worse. Questionnaire reliability was considered acceptable (Cronbach alpha 0.7). CONCLUSION: Compared to conventional iterative reconstruction algorithms, full MBMIR presented a higher image quality and low-contrast detectability and a worse acceptance among radiologists. KEY POINTS: • Full MBIR used led to an overall improvement in image quality compared with FBP and HIR. • Full MBIR leads to image texture change which reduces the confidence in these images among radiologists. • Awareness of the image texture change and improved quality of full MBIR reconstructed images could improve the acceptance of this technique in clinical practice.

The Impact of Dose Reduction in Quantitative Kinematic CT of Ankle Joints Using a Full Model-Based Iterative Reconstruction Algorithm: A Cadaveric Study.

OBJECTIVE: The objective of our study was to evaluate the quality and reproducibility of semiautomatic measurements of the ankle in low-dose kinematic CT studies using a full model-based iterative reconstruction (MBIR). MATERIALS AND METHODS: Kinematic CT was performed in five cadaveric ankles at three acquisition dose levels: standard dose (1020 mGy × cm), low dose (10% of the standard dose), and ultra-low-dose (1.5% of the standard dose). All images were reconstructed using a full MBIR algorithm. Two semiautomatic measurements (one distance and one angle) were performed by two readers. Registration error was evaluated. The bone aspect on CT and presence of metallic implants were considered in the analysis. The influence of dose on the measurements obtained, reproducibility, and image quality was assessed. RESULTS: With the standard- and low-dose protocols, registration quality was good (registration error, 0.65-4.72%), measurements were similar (p = 0.9), and reproducibility was excellent (intraclass correlation coefficient [ICC] = 0.881). With the ultra-low-dose protocol, the registration quality was poor, yielding measurements significantly different from the other protocols (p < 0.001) and poor reproducibility (ICC = 0.39). In a specimen with normal bone and no metal implant, the registration error was low (0.61-1.01%), measurements were similar (p > 0.5), and reproducibility was excellent (ICC, 0.885-0.996) for the three dose levels tested. CONCLUSION: Full MBIR allows reliable and reproducible measurements in ankle kinematic CT with a low-dose protocol, but an ultra-low-dose protocol may lead to unreliable results.