Super‐resolution for upper abdominal MRI: Acquisition and post‐processing protocol optimization using brain MRI control data and expert reader validation

Ebner, Michael; Patel, Premal A.; Atkinson, David; Caselton, Lucy; Firmin, Louisa; Amin, Zahir; Bainbridge, Alan; Coppi, Paolo De; Taylor, Stuart A.; Ourselin, Sébastien; Chouhan, Manil; Vercauteren, Tom

doi:10.1002/mrm.27852

Cited by 11 publications

(8 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, the feasibility of SR in MRI has been reported for the robust acquisition of images with better sharpness and lesion conspicuity than the standard method in abdominal (29,30) or fetal brain images (31). These reports provide encouraging results for further clinical application.…”

Section: Discussionmentioning

confidence: 99%

Generative adversarial network-based post-processed image super-resolution technology for accelerating brain MRI: comparison with compressed sensing

et al. 2022

View full text Add to dashboard Cite

Background It is unclear whether deep-learning–based super-resolution technology (SR) or compressed sensing technology (CS) can accelerate magnetic resonance imaging (MRI) . Purpose To compare SR accelerated images with CS images regarding the image similarity to reference 2D- and 3D gradient-echo sequence (GRE) brain MRI. Material and Methods We prospectively acquired 1.3× and 2.0× faster 2D and 3D GRE images of 20 volunteers from the reference time by reducing the matrix size or increasing the CS factor. For SR, we trained the generative adversarial network (GAN), upscaling the low-resolution images to the reference images with twofold cross-validation. We compared the structural similarity (SSIM) index of accelerated images to the reference image. The rate of incorrect answers of a radiologist discriminating faster and reference image was used as a subjective image similarity (ISM) index. Results The SR demonstrated significantly higher SSIM than the CS (SSIM=0.9993–0.999 vs. 0.9947–0.9986; P < 0.001). In 2D GRE, it was challenging to discriminate the SR image from the reference image, compared to the CS (ISM index 40% vs. 17.5% in 1.3×; P = 0.039 and 17.5% vs. 2.5% in 2.0×; P = 0.034). In 3D GRE, the CS revealed a significantly higher ISM index than the SR (22.5% vs. 2.5%; P = 0.011) in 2.0 × faster images. However, the ISM index was identical for the 2.0× CS and 1.3× SR (22.5% vs. 27.5%; P = 0.62) with comparable time costs. Conclusion The GAN-based SR outperformed CS in image similarity with 2D GRE for MRI acceleration. In addition, CS was more advantageous in 3D GRE than SR.

show abstract

Section: Discussionmentioning

confidence: 99%

Generative adversarial network-based post-processed image super-resolution technology for accelerating brain MRI: comparison with compressed sensing

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Inclusion of the CA 19-9 level and laparoscopy might also help to reduce the NRR and consequently spare the patient from useless exploratory surgery. In the future, imaging [23][24][25] development in association with EUS [26,27], and assistance during laparoscopy [28] could improve the relevance of tumor staging. However, the aim of staging is to detect existing distant metastasis or vascular invasion that precludes resection.…”

Section: Perspectivesmentioning

confidence: 99%

Has the non-resection rate decreased during the last two decades among patients undergoing surgical exploration for pancreatic adenocarcinoma?

et al. 2020

View full text Add to dashboard Cite

Purpose: To determine if improvement in imaging reduces the non-resection rate (NRR) among patients with pancreatic ductal adenocarcinoma (PDAC). Methods: From 2000 to 2019, 751 consecutive patients with PDAC were considered eligible for a intention-to-treat pancreatectomy and entered the operating room. In April 2011, our institution acquired a dual energy spectral computed tomography (CT) scanner and liver diffusion weighted magnetic resonance imaging (DW-MRI) was included in the imaging workup. We consequently considered 2 periods of inclusion: period #1 (February 2000-March 2011) and period #2 (April 2011-August 2019). Results: All patients underwent a preoperative CT scan with a median delay to surgery of 18 days. Liver DW-MRI was performed among 407 patients (54%). Median delay between CT and surgery decreased (21 days to 16 days, P < .01), and liver DW-MRI was significantly most prescribed during period #2 (14% vs 75%, P < .01). According to the intraoperative findings, the overall NRR was 24.5%, and remained stable over the two periods (25% vs 24%, respectively). While vascular invasion, liver metastasis, and carcinomatosis rates remained stable, para-aortic lymph nodes invasion rate (0.4% vs 4.6%; P < 0.001) significantly increased over the 2 periods. The mean size of the bigger extra pancreatic tumor significantly decrease (7.9 mm vs 6.4 mm (P < .01), respectively) when the resection was not done. In multivariate analysis, CA 19-9 < 500 U/mL (P < .01), and liver DW-MRI prescription (P < .01) favoured the resection. Conclusions: Due to changes in our therapeutic strategies, the NRR did not decrease during two decades despite imaging improvement.

show abstract

“…Compared to VVR-based techniques, motion correction using retrospective SVR followed by image reconstruction has shown significantly improved results in a range of MRI applications including diffusion-weighted imaging (DWI) of non-cooperative patients, e.g. ( Bastiani et al, 2019 ; Hutter et al, 2018 ; Marami et al, 2016 ; 2019 ), fetal brain MRI ( Alansary et al, 2017 ; Ebner et al, 2019b ; Gholipour et al, 2010 ; Kainz et al, 2015 ; Marami et al, 2017 ), fetal cardiac MRI ( van Amerom et al, 2019 ; Lloyd et al, 2019 ), and body MRI ( Ebner et al, 2019a ; Kurugol et al, 2017 ).…”

Section: Introductionmentioning

confidence: 99%

SLIMM: Slice localization integrated MRI monitoring

et al. 2020

View full text Add to dashboard Cite

Functional MRI (fMRI) is extremely challenging to perform in subjects who move because subject motion disrupts blood oxygenation level dependent (BOLD) signal measurement. It has become common to use retrospective framewise motion detection and censoring in fMRI studies to eliminate artifacts arising from motion. Data censoring results in significant loss of data and statistical power unless the data acquisition is extended to acquire more data not corrupted by motion. Acquiring more data than is necessary leads to longer than necessary scan duration, which is more expensive and may lead to additional subject non-compliance. Therefore, it is well established that real-time prospective motion monitoring is crucial to ensure data quality and reduce imaging costs. In addition, real-time monitoring of motion allows for feedback to the operator and the subject during the acquisition, to enable intervention to reduce the subject motion. The most widely used form of motion monitoring for fMRI is based on volume-to-volume registration (VVR), which quantifies motion as the misalignment between subsequent volumes. However, motion is not constrained to occur only at the boundaries of volume acquisition, but instead may occur at any time. Consequently, each slice of an fMRI acquisition may be displaced by motion, and assessment of whole volume to volume motion may be insensitive to both intra-volume and inter-volume motion that is revealed by displacement of the slices. We developed the first slice-by-slice self-navigated motion monitoring system for fMRI by developing a real-time slice-to-volume registration (SVR) algorithm. Our real-time SVR algorithm, which is the core of the system, uses a local image patch-based matching criterion along with a Levenberg-Marquardt optimizer, all accelerated via symmetric multi-processing, with interleaved and simultaneous multi-slice acquisition schemes. Extensive experimental results on real motion data demonstrated that our fast motion monitoring system, named S lice L ocalization I ntegrated M RI M onitoring (SLIMM), provides more accurate motion measurements than a VVR based approach. Therefore, SLIMM offers improved online motion monitoring which is particularly important in fMRI for challenging patient populations. Real-time motion monitoring is crucial for online data quality control and assurance, for enabling feedback to the subject and the operator to act to mitigate motion, and in adaptive acquisition strategies that aim to ensure enough data of sufficient quality is acquired without acquiring excess data.

show abstract

Super‐resolution for upper abdominal MRI: Acquisition and post‐processing protocol optimization using brain MRI control data and expert reader validation

Cited by 11 publications

References 39 publications

Generative adversarial network-based post-processed image super-resolution technology for accelerating brain MRI: comparison with compressed sensing

Generative adversarial network-based post-processed image super-resolution technology for accelerating brain MRI: comparison with compressed sensing

Has the non-resection rate decreased during the last two decades among patients undergoing surgical exploration for pancreatic adenocarcinoma?

SLIMM: Slice localization integrated MRI monitoring

Contact Info

Product

Resources

About