Validation of deep-learning image reconstruction for coronary computed tomography angiography: Impact on noise, image quality and diagnostic accuracy

Benz, Dominik C.; Benetos, Georgios; Rampidis, Georgios; Felten, Elia von; Bakula, Adam; Šustar, Aleksandra; Kudura, Ken; Messerli, Michael; Fuchs, Tobias A.; Gebhard, Cathérine; Pazhenkottil, Aju P.; Kaufmann, Philipp A.; Buechel, Ronny R.

doi:10.1016/j.jcct.2020.01.002

Cited by 120 publications

(90 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…DLR performs better for all image quality characteristics compared to MBIR. These results are in line with studies evaluating DLR in body applications [9,[12][13][14]. These studies showed a lower noise and a higher SDNR for DLR compared to Hybrid-IR and MBIR, although these studies also showed a higher Fig.…”

Section: Discussionsupporting

confidence: 90%

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

et al. 2021

View full text Add to dashboard Cite

Objectives To 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). Methods Cerebral 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. Results For 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. Conclusions With 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.

show abstract

Section: Discussionsupporting

confidence: 90%

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

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The evaluated criteria usually consisted of quantitative and qualitative measurements. Utilizing DLIR in coronary computed tomography angiography, Benz et al showed a significant reduction in noise and higher image quality, while DL image reconstruction was equal to IR with regard to diagnostic accuracy, sensitivity, and specificity in the detection of significant coronary artery stenosis with invasive coronary angiography as criterion standard [30]. The phantom study by Greffier et al published in 2020 showed a potential for dose reduction of up to 56 % for True-Fidelity, by achieving comparable detectability with iterative reconstruction and DL while lowering the dose [31].…”

Section: Deep Learning Reconstruction Algorithms In the Clinical Routinementioning

confidence: 99%

Deep Learning CT Image Reconstruction in Clinical Practice

Arndt

Güttler

Heinrich

et al. 2020

Rofo

View full text Add to dashboard Cite

Background Computed tomography (CT) is a central modality in modern radiology contributing to diagnostic medicine in almost every medical subspecialty, but particularly in emergency services. To solve the inverse problem of reconstructing anatomical slice images from the raw output the scanner measures, several methods have been developed, with filtered back projection (FBP) and iterative reconstruction (IR) subsequently providing criterion standards. Currently there are new approaches to reconstruction in the field of artificial intelligence utilizing the upcoming possibilities of machine learning (ML), or more specifically, deep learning (DL). Method This review covers the principles of present CT image reconstruction as well as the basic concepts of DL and its implementation in reconstruction. Subsequently commercially available algorithms and current limitations are being discussed. Results and Conclusion DL is an ML method that utilizes a trained artificial neural network to solve specific problems. Currently two vendors are providing DL image reconstruction algorithms for the clinical routine. For these algorithms, a decrease in image noise and an increase in overall image quality that could potentially facilitate the diagnostic confidence in lesion conspicuity or may translate to dose reduction for given clinical tasks have been shown. One study showed equal diagnostic accuracy in the detection of coronary artery stenosis for DL reconstructed images compared to IR at higher image quality levels. Consequently, a lot more research is necessary and should aim at diagnostic superiority in the clinical context covering a broadness of pathologies to demonstrate the reliability of such DL approaches. Key Points: Citation Format

show abstract

“…Other studies have demonstrated that using DLIR can further reduce image noise while avoiding blurring artifacts with IR algorithms. These studies focused on imaging coronary arteries, lower extremities, and the brain (21)(22)(23). As far as we know, no previous studies have applied DLIR to low radiation dose and contrast medium dose protocols in pediatric chest CTA.…”

Section: Introductionmentioning

confidence: 99%

Improving the image quality of pediatric chest CT angiography with low radiation dose and contrast volume using deep learning image reconstruction

Sun

et al. 2021

Quant Imaging Med Surg

View full text Add to dashboard Cite

Background: Chest CT angiography (CTA) is a common clinical examination technique for children.Iterative reconstruction algorithms are often used to reduce image noise but encounter limitations under low dose conditions. Deep learning-based image reconstruction algorithms have been developed to overcome these limitations. We assessed the quantitative and qualitative image quality of thin-slice chest CTA in children acquired with low radiation dose and contrast volume by using a deep learning image reconstruction (DLIR) algorithm.Methods: A total of 33 children underwent chest CTA with 70 kVp and automatic tube current modulation for noise indices of 11-15 based on their age and contrast volume of 0.8-1.2 mL/kg. Images were reconstructed with 50% and 100% adaptive statistical iterative reconstruction-V (ASIR-V) and highsetting DLIR (DLIR-H) at 0.625 mm slice thickness. Two radiologists evaluated images in consensus for overall image noise, artery margin, and artery contrast separately on a 5-point scale (5, excellent; 4, good; 3, acceptable; 2, sub-acceptable, and 1, not acceptable). The CT value and image noise of the descending aorta and back muscle were measured. Radiation dose and contrast volume was recorded. Results:The volume CT dose index, dose length product, and contrast volume were 1.37±0.29 mGy, 35.43±10.59 mGy•cm, and 25.43±13.32 mL, respectively. The image noises (in HU) of the aorta with DLIR-H (19.24±5.77) and 100% ASIR-V (20.45±6.93) were not significantly different (P>0.05) and were substantially lower than 50% ASIR-V (29.45±7.59) (P<0.001). The 100% ASIR-V images had oversmoothed artery margins, but only the DLIR-H images provided acceptable scores on all 3 aspects of the qualitative image quality evaluation.Conclusions: It is feasible to improve the image quality of a low radiation dose and contrast volume chest CTA in children using the high-setting DLIR algorithm.

show abstract

Validation of deep-learning image reconstruction for coronary computed tomography angiography: Impact on noise, image quality and diagnostic accuracy

Cited by 120 publications

References 25 publications

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

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

Deep Learning CT Image Reconstruction in Clinical Practice

Improving the image quality of pediatric chest CT angiography with low radiation dose and contrast volume using deep learning image reconstruction

Contact Info

Product

Resources

About