Purpose: To optimize image quality and radiation dose of chest CT with respect to various iterative reconstruction levels, detector collimations and body sizes.
Method: A Kyoto Kagaku Lungman with and without extensions was scanned using fixed ultra-low doses of 0.25, 0.49 and 0.74 mGy CTDI, and collimations of 40 and 80 mm. Images were reconstructed with the lung kernel, filtered back projection (FBP) and different ASIR-V levels (10-100%). Contrast-to-noise ratios (CNR) were calculated for 12 mm simulated lesions of different densities in the lung. Image noise, signal-to-noise ratios (SNR), variations in Hounsfield units (HU), noise power spectrum (NPS) and noise texture deviations (NTD) were evaluated for all reconstructions. NTD was calculated as percentage of pixels outside 3 standard deviations to evaluate IR-specific artefacts.
Results: Compared to the FBP, image noise reduced (5-55%) with ASIR-V levels irrespective of dose or collimation. SNR correlated positively (r ≥ 0.925, p ≤ 0.001) with ASIR-V levels at all doses, collimations, and phantom sizes. ASIR-V enhanced the CNR of the lesion with the lowest contrast from 12.7-42.1 (0-100% ASIR-V) at 0.74 mGy with 40 mm collimation. As expected, higher SNR and CNR were measured in the smaller phantom than the bigger phantom. Uniform HU were observed between FBP and ASIR-V levels at all doses, collimations, and phantom sizes. NPS curves left-shifted towards lower frequencies at increasing levels of ASIR-V irrespective of collimation. A positive correlation (r ≥ 0.946, p ≥ 0.001) was observed between NTD and ASIR-V levels. NTD of the FBP was not significantly (p ≤ 0.087) different from NTD of ASIR-V ≤ 20%. The data from the NPS and NTD indicates a blotchier and coarser noise texture at higher levels of ASIR-V, especially at 100% ASIR-V.
Conclusion: In comparison with the FBP technique, ASIR-V enhanced quantitative image quality parameters at all ultra-low doses tested. Moreover, the use of ASIR-V showed consistency with body size and collimation. Hence, ASIR-V may be useful for improving image quality of chest CT at ultra-low doses.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1016/j.ejrad.2019.02.021 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
A practical approach to the spatial-domain calculation of nonprewhitening model observers in computed tomography.
Med Phys
January 2025
Department of Nuclear Medicine and Medical Physics, Karolinska University Hospital, Stockholm, Sweden.
Background: Modern reconstruction algorithms for computed tomography (CT) can exhibit nonlinear properties, including non-stationarity of noise and contrast dependence of both noise and spatial resolution. Model observers have been recommended as a tool for the task-based assessment of image quality (Samei E et al., Med Phys.
View Article and Find Full Text PDFImpact of Deep Learning-Based Image Reconstruction on Tumor Visibility and Diagnostic Confidence in Computed Tomography.
Bioengineering (Basel)
December 2024
Department of Radiology, Jena University Hospital, Friedrich Schiller University, 07747 Jena, Germany.
Deep learning image reconstruction (DLIR) has shown potential to enhance computed tomography (CT) image quality, but its impact on tumor visibility and adoption among radiologists with varying experience levels remains unclear. This study compared the performance of two deep learning-based image reconstruction methods, DLIR and Pixelshine, an adaptive statistical iterative reconstruction-volume (ASIR-V) method, and filtered back projection (FBP) across 33 contrast-enhanced CT staging examinations, evaluated by 20-24 radiologists. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were measured for tumor and surrounding organ tissues across DLIR (Low, Medium, High), Pixelshine (Soft, Ultrasoft), ASIR-V (30-100%), and FBP.
View Article and Find Full Text PDFFeasibility of Sub-milliSievert Low-dose Computed Tomography with Deep Learning Image Reconstruction in Evaluating Pulmonary Subsolid Nodules: A Prospective Intra-individual Comparison Study.
Acad Radiol
December 2024
Department of Radiology, Shanghai Chest Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China (H.Z., Q.C., W.M., J.Y., S.W., G.T., J.X., H.J., H.Y., L.Z.). Electronic address:
Rationale And Objectives: To comprehensively assess the feasibility of low-dose computed tomography (LDCT) using deep learning image reconstruction (DLIR) for evaluating pulmonary subsolid nodules, which are challenging due to their susceptibility to noise.
Materials And Methods: Patients undergoing both standard-dose CT (SDCT) and LDCT between March and June 2023 were prospectively enrolled. LDCT images were reconstructed with high-strength DLIR (DLIR-H), medium-strength DLIR (DLIR-M), adaptive statistical iterative reconstruction-V level 50% (ASIR-V-50%), and filtered back projection (FBP); SDCT with FBP as the reference standard.
Effect of Deep Learning Image Reconstruction on Image Quality and Pericoronary Fat Attenuation Index.
J Imaging Inform Med
September 2024
Department of Radiology, BenQ Hospital, Nanjing Medical University, Nanjing, Jiangsu Province, China.
Article Synopsis
- The study compares image quality and fat attenuation index (FAI) in coronary artery CT angiography (CCTA) using different tube voltages and two reconstruction methods: deep learning image reconstruction (DLIR) and adaptive statistical iterative reconstruction V (ASIR-V).
- 301 patients were divided into two groups based on tube voltage settings, with image quality assessed at 120 kV and lower voltages chosen according to the patients' BMI.
- Results showed that DLIR-H significantly reduced image noise and improved CNR, subjective scores, and FAI across all tube voltages compared to ASIR-V, while low tube voltage combined with DLIR-H maintained good image quality compared to 120 kV.
The impact of deep learning image reconstruction of spectral CTU virtual non contrast images for patients with renal stones.
Eur J Radiol Open
December 2024
Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, 200025, China.
Purpose: To compare image quality and detection accuracy of renal stones between deep learning image reconstruction (DLIR) and Adaptive Statistical Iterative Reconstruction-Veo (ASIR-V) reconstructed virtual non-contrast (VNC) images and true non-contrast (TNC) images in spectral CT Urography (CTU).
Methods: A retrospective analysis was conducted on images of 70 patients who underwent abdominal-pelvic CTU in TNC phase using non-contrast scan and contrast-enhanced corticomedullary phase (CP) and excretory phase (EP) using spectral scan. The TNC scan was reconstructed using ASIR-V70 % (TNC-AR70), contrast-enhanced scans were reconstructed using AR70, DLIR medium-level (DM), and high-level (DH) to obtain CP-VNC-AR70/DM/DH and EP-VNC-AR70/DM/DH image groups, respectively.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!