scholarly journals Comparison of Complex k-Space Data and Magnitude-Only for Training of Deep Learning–Based Artifact Suppression for Real-Time Cine MRI

2021 ◽  
Vol 9 ◽  
Author(s):  
Hassan Haji-Valizadeh ◽  
Rui Guo ◽  
Selcuk Kucukseymen ◽  
Yankama Tuyen ◽  
Jennifer Rodriguez ◽  
...  
Keyword(s):  
Deep Learning ◽  
Image Quality ◽  
Real Time ◽  
Free Breathing ◽  
Cine Mri ◽  
Lv Function ◽  
Subjective Image Quality ◽  
Artifact Suppression ◽  
Space Data ◽  
Complex Valued

Propose: The purpose of this study was to compare the performance of deep learning networks trained with complex-valued and magnitude images in suppressing the aliasing artifact for highly accelerated real-time cine MRI.Methods: Two 3D U-net models (Complex-Valued-Net and Magnitude-Net) were implemented to suppress aliasing artifacts in real-time cine images. ECG-segmented cine images (n = 503) generated from both complex k-space data and magnitude-only DICOM were used to synthetize radial real-time cine MRI. Complex-Valued-Net and Magnitude-Net were trained with fully sampled and synthetized radial real-time cine pairs generated from highly undersampled (12-fold) complex k-space and DICOM images, respectively. Real-time cine was prospectively acquired in 29 patients with 12-fold accelerated free-breathing tiny golden-angle radial sequence and reconstructed with both Complex-Valued-Net and Magnitude-Net. Cardiac function, left-ventricular (LV) structure, and subjective image quality [1(non-diagnostic)-5(excellent)] were calculated from Complex-Valued-Net– and Magnitude-Net–reconstructed real-time cine datasets and compared to those of ECG-segmented cine (reference).Results: Free-breathing real-time cine reconstructed by both networks had high correlation (all R2 > 0.7) and good agreement (all p > 0.05) with standard clinical ECG-segmented cine with respect to LV function and structural parameters. Real-time cine reconstructed by Complex-Valued-Net had superior image quality compared to images from Magnitude-Net in terms of myocardial edge sharpness (Complex-Valued-Net = 3.5 ± 0.5; Magnitude-Net = 2.6 ± 0.5), temporal fidelity (Complex-Valued-Net = 3.1 ± 0.4; Magnitude-Net = 2.1 ± 0.4), and artifact suppression (Complex-Valued-Net = 3.1 ± 0.5; Magnitude-Net = 2.0 ± 0.0), which were all inferior to those of ECG-segmented cine (4.1 ± 1.4, 3.9 ± 1.0, and 4.0 ± 1.1).Conclusion: Compared to Magnitude-Net, Complex-Valued-Net produced improved subjective image quality for reconstructed real-time cine images and did not show any difference in quantitative measures of LV function and structure.

scholarly journals Influence of a novel deep-learning based reconstruction software on the objective and subjective image quality in low-dose abdominal computed tomography

2021 ◽  
Vol 94 (1117) ◽  
pp. 20200677
Author(s):  
Andrea Steuwe ◽  
Marie Weber ◽  
Oliver Thomas Bethge ◽  
Christin Rademacher ◽  
Matthias Boschheidgen ◽  
...  
Keyword(s):  
Deep Learning ◽  
Image Quality ◽  
Dose Reduction ◽  
Low Dose ◽  
Ct Images ◽  
Image Noise ◽  
The Novel ◽  
Subjective Image Quality ◽  
Image Information ◽  
Noise Ratio

Objectives: Modern reconstruction and post-processing software aims at reducing image noise in CT images, potentially allowing for a reduction of the employed radiation exposure. This study aimed at assessing the influence of a novel deep-learning based software on the subjective and objective image quality compared to two traditional methods [filtered back-projection (FBP), iterative reconstruction (IR)]. Methods: In this institutional review board-approved retrospective study, abdominal low-dose CT images of 27 patients (mean age 38 ± 12 years, volumetric CT dose index 2.9 ± 1.8 mGy) were reconstructed with IR, FBP and, furthermore, post-processed using a novel software. For the three reconstructions, qualitative and quantitative image quality was evaluated by means of CT numbers, noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) in six different ROIs. Additionally, the reconstructions were compared using SNR, peak SNR, root mean square error and mean absolute error to assess structural differences. Results: On average, CT numbers varied within 1 Hounsfield unit (HU) for the three assessed methods in the assessed ROIs. In soft tissue, image noise was up to 42% lower compared to FBP and up to 27% lower to IR when applying the novel software. Consequently, SNR and CNR were highest with the novel software. For both IR and the novel software, subjective image quality was equal but higher than the image quality of FBP-images. Conclusion: The assessed software reduces image noise while maintaining image information, even in comparison to IR, allowing for a potential dose reduction of approximately 20% in abdominal CT imaging. Advances in knowledge: The assessed software reduces image noise by up to 27% compared to IR and 48% compared to FBP while maintaining the image information. The reduced image noise allows for a potential dose reduction of approximately 20% in abdominal imaging.

scholarly journals CINENet: deep learning-based 3D cardiac CINE MRI reconstruction with multi-coil complex-valued 4D spatio-temporal convolutions

2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Thomas Küstner ◽  
Niccolo Fuin ◽  
Kerstin Hammernik ◽  
Aurelien Bustin ◽  
Haikun Qi ◽  
...  
Keyword(s):  
Deep Learning ◽  
Cine Mri ◽  
Cardiac Cine ◽  
Mri Reconstruction ◽  
Spatio Temporal ◽  
Complex Valued

Highly accelerated free‐breathing real‐time phase contrast cardiovascular MRI via complex‐difference deep learning

2021 ◽  
Author(s):  
Hassan Haji‐Valizadeh ◽  
Rui Guo ◽  
Selcuk Kucukseymen ◽  
Amanda Paskavitz ◽  
Xiaoying Cai ◽  
...  
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Phase Contrast ◽  
Free Breathing ◽  
Cardiovascular Mri ◽  
Complex Difference

scholarly journals P736 Feasibility and accuracy of real-time automatic quantification of left ventricular ejection fraction by hand-held ultrasound device

2020 ◽  
Vol 21 (Supplement_1) ◽  
Author(s):  
A Hjorth-Hansen ◽  
M S Magelssen ◽  
G N Anderssen ◽  
T Graven ◽  
J O Kleinau ◽  
...  
Keyword(s):  
Image Quality ◽  
Ejection Fraction ◽  
Real Time ◽  
Left Ventricular ◽  
Left Ventricular Ejection ◽  
Lv Function ◽  
Cardiac Ultrasound ◽  
Automatic Assessment ◽  
Automatic Quantification ◽  
Focused Cardiac Ultrasound

Abstract Funding Acknowledgements This study is perfomed with loan of hand-held devices and technical support by GE Ultrasound. The study was financed by NTNU and Levanger Hospital Background Automatic quantification of left ventricular (LV) ejection fraction (EF) by hand-held ultrasound devices (HUDs) may increase the benefit of focused cardiac ultrasound by inexperienced users. Automatic measurements of EF (autoEF) from 4-chamber (4Ch) recordings are available for real-time use. Image quality is of major importance for EF assessment, but the influence of automatic quantification of LV function by HUD is not evaluated. Purpose To evaluate the feasibility and accuracy of autoEF for real-time quantification of LV function by HUDs, and to assess the importance of image quality and the operators’ experience for the accuracy of the method. Methods Patients referred to a cardiac outpatient clinic with suspected heart failure (HF) were included. In total, 5 GPs, 3 nurses and 5 cardiologists performed HUD examinations. Each patient was examined by a nurse and a general practitioner (GPs) using a HUD. Immediately thereafter, a cardiologist performed echocardiographic reference examination and a HUD examination for autoEF. The GPs underwent six pre-study practical training days in addition to lectures. The nurses were familiar with focused cardiac ultrasound from their work at the outpatient HF clinic. AutoEF was measured in 4Ch view only. Reference EF was measured by the cardiologists using the biplane Simpson’s method. Another cardiologist blinded to the reference measurements and operators evaluated each HUD recording on technical and qualitative parameters using a scale from 1 (poor) to 6 (very good), and recommended to accept or reject the autoEF result. In total, 510 recordings of autoEF were available for analysis. Results 87 patients (46% women) with mean age of 67.5 years were examined. Mean BMI was 29 kg/m2, 95% CI (27.5, 30.2). 24 (30%) had atrial fibrillation. Of 510 recordings with autoEF measurements, 255 (50%) were rejected during evaluation and considered not eligible for clinical use. AutoEF by HUD and refEF was mean (SD) 52.6% (16.7) and 53.2% (7.3), respectively. Overall, the quality score for autoEF recordings was mean (SD) 4.4 (0.9). The highest mean score was 5.0 (SD 0.7) by the cardiologist and lowest for the GPs 4.0 (SD 0.9) (p < 0.001). The corresponding proportions of accepted autoEF measurements were 75% and 33%, respectively. The difference compared to reference were lowest in the accepted recordings (p < 0.001). The most important parameters for correct autoEF measurements were a properly assessed 4Ch view, a well visualized mitral annulus and the number of segments with visible endocardium (all p < 0.01), the latter being the overall most important parameter. Conclusion The feasibility of evaluation of real-time automatic assessment of LV EF by HUD was only moderate. In the hands of the least experienced the use of automatic LV EF was not of adequate quality compared to reference. Thus, sufficient training and good image quality is essential for automatic assessment of LV function by HUDs.

scholarly journals Reliability and reproducibility of cardiac MRI quantification of peak exercise function with long-axis views

PLoS ONE ◽  
2021 ◽  
Vol 16 (2) ◽  
pp. e0245912
Author(s):  
Amy A. Kirkham ◽  
Michelle V. Goonasekera ◽  
Brenna C. Mattiello ◽  
Justin G. Grenier ◽  
Mark J. Haykowsky ◽  
...  
Keyword(s):  
Cardiovascular Risk ◽  
Real Time ◽  
Global Longitudinal Strain ◽  
Intraclass Correlation ◽  
Free Breathing ◽  
Left Ventricular ◽  
Peak Exercise ◽  
Breath Hold ◽  
Lv Function ◽  
Image Stack

The conventional approach to cardiac magnetic resonance (CMR) involving breath holds, electrocardiography-gating, and acquisition of a short-axis (SAX) image stack, introduces technical and logistical challenges for assessing exercise left ventricular (LV) function. Real-time, free-breathing CMR acquisition of long-axis (LAX) images overcomes these issues and also enables assessment of global longitudinal strain (GLS). We evaluated the reliability of a free-breathing LAX approach compared to the standard SAX approach and the reproducibility of free-breathing LAX. LV SAX (contiguous stack) and LAX (two-chamber and four-chamber) 3T CMR cine images were acquired four times within one scan in 32 women with cardiovascular risk factors (56±10 years, 28±4 kg/m2) as follows: 1) resting, gated-segmented, end-expiration breath-hold; 2) resting, real-time, free-breathing; 3) test-retest set of resting, real-time, free-breathing; 4) peak exercise (incremental-to-maximum, in-magnet, stepper test), real-time, free-breathing. A second scan was performed within one week in a subset (n = 5) to determine reproducibility of peak exercise measures. Reliability and agreement of the free-breathing LAX approach with the conventional SAX approach were assessed by intraclass correlation coefficient (ICC) and Bland-Altman plots, respectively. Normal control GLS reserve was also acquired in a separate set of 12 young, healthy control women (25±4 years, 22±2 kg/m2) for comparison. Comparisons of LV volumes and function among all techniques at rest had good-to-excellent reliability (ICC = 0.80–0.96), and excellent reliability between peak exercise free-breathing LAX and SAX evaluations (ICC = 0.92–0.96). Higher resting heart rates with free-breathing acquisitions compared to breath-hold (mean difference, limits of agreement: 5, 1–12 beats per minute) reduced reliability for cardiac output (ICC = 0.67–0.79). Reproducibility of the free-breathing LAX approach was good-to-excellent at rest and peak exercise (ICC = 0.74–0.99). GLS exercise reserve was impaired in older women at cardiovascular risk compared to young healthy women (-4.7±2.3% vs -7.4±2.1%, p = 0.001). Real-time, free-breathing CMR with LAX evaluation provides a reliable and reproducible method to assess rest and peak exercise cardiac function, including GLS.

Abstract 1872: New Fusion Technique Significantly Improves Image Quality and Completeness of Datasets in Real Time 3 Dimensional Echocardiography

Circulation ◽  
2007 ◽  
Vol 116 (suppl_16) ◽  
Author(s):  
Cezary Szmigielski ◽  
Vicente Grau ◽  
Martin Hoeschen ◽  
J. A Noble ◽  
Richard Kerber ◽  
...  
Keyword(s):  
Image Quality ◽  
Real Time ◽  
Lv Function ◽  
Wall Thickening ◽  
Endocardial Border ◽  
Fusion Technique ◽  
3 Dimensional ◽  
Dimensional Echocardiography ◽  
Number Of Segments ◽  
Full Volume

Introduction: Real time 3 dimensional echocardiography (RT3DE) still presents a challenge for full volume acquisition in many patients because of incomplete datasets due to inadequate sector width and/or limited acoustic windows. Hypothesis: We assessed the hypothesis that fusion of standard datasets from different transducer positions improves the completeness and image quality of RT3DE. Methods: We studied 14 patients (mean age 47.0±22 yrs, 9 men, 5 women) referred for echocardiographic assessment of LV function. Standard full volume RT3DE was performed with an iE33 scanner with X3 probe (Philips Medical Systems, Andover, USA). We fused at least 2 standard RT3DE datasets from different transducer positions using a new, developed on site technique. All standard and fused datasets were processed with research software (Research Arena, TomTec, Germany). Readings were done by an experienced echocardiographer, who graded endocardial border definition and wall thickening: good quality: grade 2; intermediate: 1; poor or out of sector: 0. We studied separately segments in the short axis view (PSAX) and in apical views, yielding 32 segments per patient and total 896 cardiac segments for standard and fused datasets. Results: Table 1 summarizes the segmental quality assessments using the 2 techniques, and the number of segments per patient (mean±SD). Quality gains from grade 0 to 1 and 1 to 2 in the fused images were similarly distributed among PSAX and apical segments. Conclusions: The new fusion technique improved analysis of segmentation of the heart with RT3DE when compared with standard datasets. It increased the number of segments available for assessment and upgraded their image quality. Table 1

scholarly journals Free-Breathing 3D Imaging of Right Ventricular Structure and Function Using Respiratory and Cardiac Self-Gated Cine MRI

2015 ◽  
Vol 2015 ◽  
pp. 1-9 ◽  
Author(s):  
Yanchun Zhu ◽  
Jing Liu ◽  
Jonathan Weinsaft ◽  
Pascal Spincemaille ◽  
Thanh D. Nguyen ◽  
...  
Keyword(s):  
Image Quality ◽  
Right Ventricular ◽  
3D Imaging ◽  
Free Breathing ◽  
Cine Mri ◽  
Cine Imaging ◽  
Breath Hold ◽  
Imaging Method ◽  
Long Term Outcome ◽  
And Function

Providing a movie of the beating heart in a single prescribed plane, cine MRI has been widely used in clinical cardiac diagnosis, especially in the left ventricle (LV). Right ventricular (RV) morphology and function are also important for the diagnosis of cardiopulmonary diseases and serve as predictors for the long term outcome. The purpose of this study is to develop a self-gated free-breathing 3D imaging method for RV quantification and to evaluate its performance by comparing it with breath-hold 2D cine imaging in 7 healthy volunteers. Compared with 2D, the 3D RV functional measurements show a reduction of RV end-diastole volume (RVEDV) by 10%, increase of RV end-systole volume (RVESV) by 1.8%, reduction of RV systole volume (RVSV) by 21%, and reduction of RV ejection fraction (RVEF) by 12%. High correlations between the two techniques were found (RVEDV: 0.94; RVESV: 0.85; RVSV: 0.95; and RVEF: 0.89). Compared with 2D, the 3D image quality measurements show a small reduction in blood SNR, myocardium-blood CNR, myocardium contrast, and image sharpness. In conclusion, the proposed self-gated free-breathing 3D cardiac cine imaging technique provides comparable image quality and correlated functional measurements to those acquired with the multiple breath-hold 2D technique in RV.

scholarly journals Deep Learning-Based Image Reconstruction for CT Angiography of the Aorta

Diagnostics ◽  
2021 ◽  
Vol 11 (11) ◽  
pp. 2037
Author(s):  
Andra Heinrich ◽  
Felix Streckenbach ◽  
Ebba Beller ◽  
Justus Groß ◽  
Marc-André Weber ◽  
...  
Keyword(s):  
Deep Learning ◽  
Image Quality ◽  
Ct Angiography ◽  
Thoracic Aorta ◽  
Abdominal Aorta ◽  
Ascending Aorta ◽  
Image Noise ◽  
Descending Thoracic Aorta ◽  
Iliac Arteries ◽  
Subjective Image Quality

To evaluate the impact of a novel, deep-learning-based image reconstruction (DLIR) algorithm on image quality in CT angiography of the aorta, we retrospectively analyzed 51 consecutive patients who underwent ECG-gated chest CT angiography and non-gated acquisition for the abdomen on a 256-dectector-row CT. Images were reconstructed with adaptive statistical iterative reconstruction (ASIR-V) and DLIR. Intravascular image noise, the signal-to-noise ratio (SNR) and the contrast-to-noise ratio (CNR) were quantified for the ascending aorta, the descending thoracic aorta, the abdominal aorta and the iliac arteries. Two readers scored subjective image quality on a five-point scale. Compared to ASIR-V, DLIR reduced the median image noise by 51-54% for the ascending aorta and the descending thoracic aorta. Correspondingly, median CNR roughly doubled for the ascending aorta and descending thoracic aorta. There was a 38% reduction in image noise for the abdominal aorta and the iliac arteries, with a corresponding improvement in CNR. Median subjective image quality improved from good to excellent at all anatomical levels. In CT angiography of the aorta, DLIR substantially improved objective and subjective image quality beyond what can be achieved by state-of-the-art iterative reconstruction. This can pave the way for further radiation or contrast dose reductions.

Sign in / Sign up

Export Citation Format

Share Document