Improvement of late gadolinium enhancement image quality using a deep learning-based reconstruction algorithm and its influence on myocardial scar quantification

Prognostic value of late gadolinium enhancement cardiac MRI for ICD therapy in non-ischaemic cardiomyopathy : A 5-year cohort study

AIM

To evaluate the impact of the 2023 Dutch national guidelines for primary prevention implantable cardioverter-defibrillator (ICD) implantation on outcomes in non-ischaemic cardiomyopathy (NICM) patients and to assess the role of late gadolinium enhancement cardiac magnetic resonance imaging (LGE-CMR) in predicting ICD therapy.

METHODS

This retrospective, single-centre observational exploratory cohort study included patients with NICM who received a primary prevention single-chamber, dual-chamber or subcutaneous ICD between January 2008 and April 2022 and underwent LGE-CMR prior to implantation. Patients were classified into LGE+ and LGE- groups based on the presence of late enhancement detected by CMR. The primary endpoint was time to first appropriate ICD therapy. The secondary endpoint was all-cause mortality.

RESULTS

Of the 258 NICM patients in the database, a total of 85 patients were included, of whom 41 had LGE on CMR. After a 5-year follow-up period, appropriate ICD therapy occurred in 20% of the patients in the LGE+ group and 14% of patients in the LGE- group (p = 0.37). All-cause mortality was 7% in the LGE+ group and 14% in the LGE- group (p = 0.46). Multivariable analysis showed no parameters significantly associated with appropriate ICD therapy.

CONCLUSION

Applying the 2023 national guidelines retrospectively on a population of NICM patients with a primary prevention ICD indication demonstrated no significant association between LGE on CMR and appropriate ICD therapy over a follow-up period of 5 years. These findings underscore the need for further research and randomised trials to refine risk stratification and ICD implantation guidelines in NICM, ideally leveraging a multicentre approach to address current limitations in sample size and enhance the generalisability of the results.

T2-weighted imaging of rectal cancer using a 3D fast spin echo sequence with and without deep learning reconstruction: A reader study

PURPOSE

To compare image quality and clinical utility of a T2-weighted (T2W) 3-dimensional (3D) fast spin echo (FSE) sequence using deep learning reconstruction (DLR) versus conventional reconstruction for rectal magnetic resonance imaging (MRI).

METHODS

The study included 50 patients with rectal cancer who underwent rectal MRI consecutively between July 7, 2020 and January 20, 2021 using a T2W 3D FSE sequence with DLR and conventional reconstruction. Three radiologists reviewed the two sets of images, scoring overall SNR, motion artifacts, and overall image quality on a 3-point scale and indicating clinical preference for DLR or conventional reconstruction based on those three criteria as well as image characterization of bowel wall layer definition, tumor invasion of muscularis propria, residual disease, fibrosis, nodal margin, and extramural venous invasion.

RESULTS

Image quality was rated as moderate or good for both DLR and conventional reconstruction for most cases. DLR was preferred over conventional reconstruction in all of the categories except for bowel wall layer definition.

CONCLUSION

Both conventional reconstruction and DLR provide acceptable image quality for T2W 3D FSE imaging of rectal cancer. DLR was clinically preferred over conventional reconstruction in almost all categories.

Deep learning image enhancement for confident diagnosis of TMJ osteoarthritis in zero-TE MR imaging

OBJECTIVES

This study aimed to evaluate the effectiveness of deep learning method for denoising and artefact reduction (AR) in zero echo time MRI (ZTE-MRI). Also, clinical applicability was evaluated by comparing image diagnosis to the temporomandibular joint (TMJ) cone-beam CT (CBCT).

METHODS

CBCT and routine ZTE-MRI data were collected for 30 patients, along with an additional ZTE-MRI obtained with reduced scan time. Scan time-reduced image sets were processed into denoised and AR images based on a deep learning technique. The image quality of the routine sequence, denoised, and AR image sets was compared quantitatively using the signal-to-noise ratio (SNR) and qualitatively using a 3-point grading system (0: poor, 1: good, 2: excellent). The presence of osteoarthritis was assessed in each imaging protocol. Diagnostic accuracy of each protocol was compared against the CBCT results, which served as the reference standard. The SNR and the qualitative scores were compared using analysis of variance test and Kruskal-Wallis test, respectively. The diagnostic accuracy was assessed using Cohen's κ (<0.5 = poor; 0.5 to <0.75 = moderate; 0.75 to <0.9 = good; ≥0.9 = excellent).

RESULTS

Both the denoised and AR protocols resulted in significantly enhanced SNR compared to the routine protocol, with the AR protocol showing a higher SNR than the denoised one. The qualitative assessment also showed highest grade in AR protocol with statistical significance. The osteoarthritis diagnosis showed enhanced agreement with CBCT in denoised (κ = 0.928) and AR images (κ = 0.929) than routine images (κ = 0.707).

CONCLUSIONS

A newly developed deep learning technique for both denoising and artefact reduction in ZTE-MRI presented clinical usefulness. Specifically, AR protocol showed significantly improved image quality and comparable diagnostic accuracy comparable to CBCT. It can be expected that this novel technique would help overcome the current limitation of ZTE-MRI for replacing CBCT in bone imaging of TMJ.

Lumos: Software for Multi-level Multi-reader Comparison of Cardiovascular Magnetic Resonance Late Gadolinium Enhancement Scar Quantification

Cardiovascular magnetic resonance imaging (CMR) offers state-of-the-art myocardial tissue differentiation. The CMR technique late gadolinium enhancement (LGE) currently provides the noninvasive gold standard for the detection of myocardial fibrosis. Typically, thresholding methods are used for fibrotic scar tissue quantification. A major challenge for standardized CMR assessment is large variations in the estimated scar for different methods. The aim was to improve quality assurance for LGE scar quantification, a multi-reader comparison tool "Lumos" was developed to support quality control for scar quantification methods. The thresholding methods and an exact rasterization approach were implemented, as well as a graphical user interface (GUI) with statistical and case-specific tabs. Twenty LGE cases were considered with half of them including artifacts and clinical results for eight scar quantification methods computed. Lumos was successfully implemented as a multi-level multi-reader comparison software, and differences between methods can be seen in the statistical results. Histograms visualize confounding effects of different methods. Connecting the statistical level with the case level allows for backtracking statistical differences to sources of differences in the threshold calculation. Being able to visualize the underlying groundwork for the different methods in the myocardial histogram gives the opportunity to identify causes for different thresholds. Lumos showed the differences in the clinical results between cases with artifacts and cases without artifacts. A video demonstration of Lumos is offered as supplementary material 1. Lumos allows for a multi-reader comparison for LGE scar quantification that offers insights into the origin of reader differences.

Super-resolution synthetic MRI using deep learning reconstruction for accurate diagnosis of knee osteoarthritis

OBJECTIVE

To assess the accuracy of deep learning reconstruction (DLR) technique on synthetic MRI (SyMRI) including T2 measurements and diagnostic performance of DLR synthetic MRI (SyMRIDL) in patients with knee osteoarthritis (KOA) using conventional MRI as standard reference.

MATERIALS AND METHODS

This prospective study recruited 36 volunteers and 70 patients with suspected KOA from May to October 2023. DLR and non-DLR synthetic T2 measurements (T2-SyMRIDL, T2-SyMRI) for phantom and in vivo knee cartilage were compared with multi-echo fast-spin-echo (MESE) sequence acquired standard T2 values (T2MESE). The inter-reader agreement on qualitative evaluation of SyMRIDL and the positive percent agreement (PPA) and negative percentage agreement (NPA) were analyzed using routine images as standard diagnosis.

RESULTS

DLR significantly narrowed the quantitative differences between T2-SyMRIDL and T2MESE for 0.8 ms with 95% LOA [-5.5, 7.1]. The subjective assessment between DLR synthetic MR images and conventional MRI was comparable (all p > 0.05); Inter-reader agreement for SyMRIDL and conventional MRI was substantial to almost perfect with values between 0.62 and 0.88. SyMRIDL MOAKS had substantial inter-reader agreement and high PPA/NPA values (95%/99%) using conventional MRI as standard reference. Moreover, T2-SyMRIDL measurements instead of non-DLR ones significantly differentiated normal-appearing from injury-visible cartilages.

CONCLUSION

DLR synthetic knee MRI provided both weighted images for clinical diagnosis and accurate T2 measurements for more confidently identifying early cartilage degeneration from normal-appearing cartilages.

CRITICAL RELEVANCE STATEMENT

One-acquisition synthetic MRI based on deep learning reconstruction provided an accurate quantitative T2 map and morphologic images in relatively short scan time for more confidently identifying early cartilage degeneration from normal-appearing cartilages compared to the conventional morphologic knee sequences.

KEY POINTS

Deep learning reconstruction (DLR) synthetic knee cartilage T2 values showed no difference from conventional ones. DLR synthetic T1-, proton density-, STIR-weighted images had high positive percent agreement and negative percentage agreement using MRI OA Knee Score features. DLR synthetic T2 measurements could identify early cartilage degeneration from normal-appearing ones.

Evaluating performance and quality of a fast multi-contrast scan in routine brain MRI

BACKGROUND AND PURPOSE

Neuromix is a fast, motion robust multi-contrast sequence capable of providing all diagnostic contrasts in ∼3.5 minutes. However, more evaluation is needed across the various contrasts compared to gold standard, optimized sequences routinely used in the clinic. The goal of this study was to prospectively determine how NeuroMix performs in the clinical setting compared to routine clinical MRI.

METHODS

NeuroMix and routine clinical MRI sequences were acquired on a 3 Tesla clinical scanner for 39 patients clinically indicated for brain MRI. Three radiologists were asked to assess the diagnostic confidence of NeuroMix compared to the routine MRI using a series of questions. Signal-to-noise and contrast-to-noise ratios (SNR and CNR) were assessed for NeuroMix. Fleiss' free-marginal multirater kappa was calculated for the qualitative assessment performed by the radiologists.

RESULTS

Radiologists were comfortable substituting or reading some of the NeuroMix sequences in place of the corresponding conventional sequence for some contrasts, including diffusion-weighted imaging, single-shot T2, and susceptibility-weighted imaging. The image quality, SNR, and CNR allowed the radiologists to visualize anatomy and pathology on NeuroMix images. There was no significant difference between coefficient of variation for the apparent diffusion coefficient maps (p = .084).

CONCLUSIONS

Analysis revealed both positives and some pitfalls of NeuroMix. However, these results indicate Neuromix as having the capability to be a backup sequence in case artifacts are present in routine sequences, or potentially a replacement for some contrasts altogether.

Late gadolinium enhancement cardiovascular magnetic resonance with generative artificial intelligence

BACKGROUND

Late gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR) imaging enables imaging of scar/fibrosis and is a cornerstone of most CMR imaging protocols. CMR imaging can benefit from image acceleration; however, image acceleration in LGE remains challenging due to its limited signal-to-noise ratio. In this study, we sought to evaluate a rapid two-dimensional (2D) LGE imaging protocol using a generative artificial intelligence (AI) algorithm with inline reconstruction.

METHODS

A generative AI-based image enhancement was used to improve the sharpness of 2D LGE images acquired with low spatial resolution in the phase-encode direction. The generative AI model is an image enhancement technique built on the enhanced super-resolution generative adversarial network. The model was trained using balanced steady-state free-precession cine images, readily used for LGE without additional training. The model was implemented inline, allowing the reconstruction of images on the scanner console. We prospectively enrolled 100 patients (55 ± 14 years, 72 males) referred for clinical CMR at 3T. We collected three sets of LGE images in each subject, with in-plane spatial resolutions of 1.5 × 1.5-3-6 mm2. The generative AI model enhanced in-plane resolution to 1.5 × 1.5 mm2 from the low-resolution counterparts. Images were compared using a blur metric, quantifying the perceived image sharpness (0 = sharpest, 1 = blurriest). LGE image sharpness (using a 5-point scale) was assessed by three independent readers.

RESULTS

The scan times for the three imaging sets were 15 ± 3, 9 ± 2, and 6 ± 1 s, with inline generative AI-based images reconstructed time of ∼37 ms. The generative AI-based model improved visual image sharpness, resulting in lower blur metric compared to low-resolution counterparts (AI-enhanced from 1.5 × 3 mm2 resolution: 0.3 ± 0.03 vs 0.35 ± 0.03, P < 0.01). Meanwhile, AI-enhanced images from 1.5 × 3 mm2 resolution and original LGE images showed similar blur metric (0.30 ± 0.03 vs 0.31 ± 0.03, P = 1.0) Additionally, there was an overall 18% improvement in image sharpness between AI-enhanced images from 1.5 × 3 mm2 resolution and original LGE images in the subjective blurriness score (P < 0.01).

CONCLUSION

The generative AI-based model enhances the image quality of 2D LGE images while reducing the scan time and preserving imaging sharpness. Further evaluation in a large cohort is needed to assess the clinical utility of AI-enhanced LGE images for scar evaluation, as this proof-of-concept study does not provide evidence of an impact on diagnosis.

Application of deep learning reconstruction in abdominal magnetic resonance cholangiopancreatography for image quality improvement and acquisition time reduction

PURPOSE

To compare deep learning (DL)-based and conventional reconstruction through subjective and objective analysis and ascertain whether DL-based reconstruction improves the quality and acquisition speed of clinical abdominal magnetic resonance imaging (MRI).

METHODS

The 124 patients who underwent abdominal MRI between January and July 2021 were retrospectively studied. For each patient, two-dimensional axial T2-weighted single-shot fast spin-echo MRI images with or without fat saturation were reconstructed using DL-based and conventional methods. The subjective image quality scores and objective metrics, including signal-to-noise ratios (SNRs) and contrast-to-noise ratios (CNRs) of the images were analysed. An explorative analysis was performed to compare 20 patients' MRI images with site routine settings, high-resolution settings and high-speed settings. Paired t tests and Wilcoxon signed-rank tests were used for subjective and objective comparisons.

RESULTS

A total of 144 patients were evaluated (mean age, 62.2 ± 14.1 years; 83 men). The MRI images reconstructed using DL-based methods had higher SNRs and CNRs than did those reconstructed using conventional methods (all p < 0.01). The subjective scores of the images reconstructed using DL-based methods were higher than those of the images reconstructed using conventional methods (p < 0.01), with significantly lower variation (p < 0.01). Exploratory analysis revealed that the DL-based reconstructions with thin slice thickness and higher temporal resolution had the highest image quality and were associated with the shortest scan times.

CONCLUSIONS

DL-based reconstruction methods can be used to improve the quality with higher stability and accelerate the acquisition of abdominal MRI.

Enhanced image quality and lesion detection in FLAIR MRI of white matter hyperintensity through deep learning-based reconstruction

OBJECTIVE

To delve deeper into the study of degenerative diseases, it becomes imperative to investigate whether deep-learning reconstruction (DLR) can improve the evaluation of white matter hyperintensity (WMH) on 3.0T scanners, and compare its lesion detection capabilities with conventional reconstruction (CR).

METHODS

A total of 131 participants (mean age, 46 years ±17; 46 men) were included in the study. The images of these participants were evaluated by readers blinded to clinical data. Two readers independently assessed subjective image indicators on a 4-point scale. The severity of WMH was assessed by four raters using the Fazekas scale. To evaluate the relative detection capabilities of each method, we employed the Wilcoxon signed rank test to compare scores between the DLR and the CR group. Additionally, we assessed interrater reliability using weighted k statistics and intraclass correlation coefficient to test consistency among the raters.

RESULTS

In terms of subjective image scoring, the DLR group exhibited significantly better scores compared to the CR group (P < 0.001). Regarding the severity of WMH, the DL group demonstrated superior performance in detecting lesions. Majority readers agreed that the DL group provided clearer visualization of the lesions compared to the conventional group.

CONCLUSION

DLR exhibits notable advantages over CR, including subjective image quality, lesion detection sensitivity, and inter reader reliability.

The role of artificial intelligence in cardiovascular magnetic resonance imaging

Cardiovascular magnetic resonance (CMR) imaging is the gold standard test for myocardial tissue characterization and chamber volumetric and functional evaluation. However, manual CMR analysis can be time-consuming and is subject to intra- and inter-observer variability. Artificial intelligence (AI) is a field that permits automated task performance through the identification of high-level and complex data relationships. In this review, we review the rapidly growing role of AI in CMR, including image acquisition, sequence prescription, artifact detection, reconstruction, segmentation, and data reporting and analysis including quantification of volumes, function, myocardial infarction (MI) and scar detection, and prediction of outcomes. We conclude with a discussion of the emerging challenges to widespread adoption and solutions that will allow for successful, broader uptake of this powerful technology.

Artificial Intelligence for Cardiovascular Care-Part 1: Advances: JACC Review Topic of the Week

Recent artificial intelligence (AI) advancements in cardiovascular care offer potential enhancements in diagnosis, treatment, and outcomes. Innovations to date focus on automating measurements, enhancing image quality, and detecting diseases using novel methods. Applications span wearables, electrocardiograms, echocardiography, angiography, genetics, and more. AI models detect diseases from electrocardiograms at accuracy not previously achieved by technology or human experts, including reduced ejection fraction, valvular heart disease, and other cardiomyopathies. However, AI's unique characteristics necessitate rigorous validation by addressing training methods, real-world efficacy, equity concerns, and long-term reliability. Despite an exponentially growing number of studies in cardiovascular AI, trials showing improvement in outcomes remain lacking. A number are currently underway. Embracing this rapidly evolving technology while setting a high evaluation benchmark will be crucial for cardiology to leverage AI to enhance patient care and the provider experience.

Evaluation of deep learning-based reconstruction late gadolinium enhancement images for identifying patients with clinically unrecognized myocardial infarction

BACKGROUND

The presence of infarction in patients with unrecognized myocardial infarction (UMI) is a critical feature in predicting adverse cardiac events. This study aimed to compare the detection rate of UMI using conventional and deep learning reconstruction (DLR)-based late gadolinium enhancement (LGEO and LGEDL, respectively) and evaluate optimal quantification parameters to enhance diagnosis and management of suspected patients with UMI.

METHODS

This prospective study included 98 patients (68 men; mean age: 55.8 ± 8.1 years) with suspected UMI treated at our hospital from April 2022 to August 2023. LGEO and LGEDL images were obtained using conventional and commercially available inline DLR algorithms. The myocardial signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and percentage of enhanced area (Parea) employing the signal threshold versus reference mean (STRM) approach, which correlates the signal intensity (SI) within areas of interest with the average SI of normal regions, were analyzed. Analysis was performed using the standard deviation (SD) threshold approach (2SD-5SD) and full width at half maximum (FWHM) method. The diagnostic efficacies based on LGEDL and LGEO images were calculated.

RESULTS

The SNRDL and CNRDL were two times better than the SNRO and CNRO, respectively (P < 0.05). Parea-DL was elevated compared to Parea-O using the threshold methods (P < 0.05); however, no intergroup difference was found based on the FWHM method (P > 0.05). The Parea-DL and Parea-O also differed except between the 2SD and 3SD and the 4SD/5SD and FWHM methods (P < 0.05). The receiver operating characteristic curve analysis revealed that each SD method exhibited good diagnostic efficacy for detecting UMI, with the Parea-DL having the best diagnostic efficacy based on the 5SD method (P < 0.05). Overall, the LGEDL images had better image quality. Strong diagnostic efficacy for UMI identification was achieved when the STRM was ≥ 4SD and ≥ 3SD for the LGEDL and LGEO, respectively.

CONCLUSIONS

STRM selection for LGEDL magnetic resonance images helps improve clinical decision-making in patients with UMI. This study underscored the importance of STRM selection for analyzing LGEDL images to enhance diagnostic accuracy and clinical decision-making for patients with UMI, further providing better cardiovascular care.

Thin-slice elbow MRI with deep learning reconstruction: Superior diagnostic performance of elbow ligament pathologies

PURPOSE

With the slice thickness routinely used in elbow MRI, small or subtle lesions may be overlooked or misinterpreted as insignificant. To compare 1 mm slice thickness MRI (1 mm MRI) with deep learning reconstruction (DLR) to 3 mm slice thickness MRI (3 mm MRI) without/with DLR, and 1 mm MRI without DLR regarding image quality and diagnostic performance for elbow tendons and ligaments.

METHODS

This retrospective study included 53 patients between February 2021 and January 2022, who underwent 3 T elbow MRI, including T2-weighted fat-saturated coronal 3 mm and 1 mm MRI without/with DLR. Two radiologists independently assessed four MRI scans for image quality and artefacts, and identified the pathologies of the five elbow tendons and ligaments. In 19 patients underwent elbow surgery after elbow MRI, diagnostic performance was evaluated using surgical records as a reference standard.

RESULTS

For both readers, 3 mm MRI with DLR had significant higher image quality scores than 3 mm MRI without DLR and 1 mm MRI with DLR (all P < 0.01). For common extensor tendon and elbow ligament pathologies, 1 mm MRI with DLR showed the highest number of pathologies for both readers. The 1 mm MRI with DLR had the highest kappa values for all tendons and ligaments. For reader 1, 1 mm MRI with DLR showed superior diagnostic performance than 3 mm MRI without/with DLR. For reader 2, 1 mm MRI with DLR showed the highest diagnostic performance; however, there was no significant difference.

CONCLUSIONS

One mm MRI with DLR showed the highest diagnostic performance for evaluating elbow tendon and ligament pathologies, with similar subjective image qualities and artefacts.

Optimized 3D brachial plexus MR neurography using deep learning reconstruction

OBJECTIVE

To evaluate whether 'fast,' unilateral, brachial plexus, 3D magnetic resonance neurography (MRN) acquisitions with deep learning reconstruction (DLR) provide similar image quality to longer, 'standard' scans without DLR.

MATERIALS AND METHODS

An IRB-approved prospective cohort of 30 subjects (13F; mean age = 50.3 ± 17.8y) underwent clinical brachial plexus 3.0 T MRN with 3D oblique-coronal STIR-T2-weighted-FSE. 'Standard' and 'fast' scans (time reduction = 23-48%, mean = 33%) were reconstructed without and with DLR. Evaluation of signal-to-noise ratio (SNR) and edge sharpness was performed for 4 image stacks: 'standard non-DLR,' 'standard DLR,' 'fast non-DLR,' and 'fast DLR.' Three raters qualitatively evaluated 'standard non-DLR' and 'fast DLR' for i) bulk motion (4-point scale), ii) nerve conspicuity of proximal and distal suprascapular and axillary nerves (5-point scale), and iii) nerve signal intensity, size, architecture, and presence of a mass (binary). ANOVA or Wilcoxon signed rank test compared differences. Gwet's agreement coefficient (AC2) assessed inter-rater agreement.

RESULTS

Quantitative SNR and edge sharpness were superior for DLR versus non-DLR (SNR by + 4.57 to + 6.56 [p < 0.001] for 'standard' and + 4.26 to + 4.37 [p < 0.001] for 'fast;' sharpness by + 0.23 to + 0.52/pixel for 'standard' [p < 0.018] and + 0.21 to + 0.25/pixel for 'fast' [p < 0.003]) and similar between 'standard non-DLR' and 'fast DLR' (SNR: p = 0.436-1, sharpness: p = 0.067-1). Qualitatively, 'standard non-DLR' and 'fast DLR' had similar motion artifact, as well as nerve conspicuity, signal intensity, size and morphology, with high inter-rater agreement (AC2: 'standard' = 0.70-0.98, 'fast DLR' = 0.69-0.97).

CONCLUSION

DLR applied to faster, 3D MRN acquisitions provides similar image quality to standard scans. A faster, DL-enabled protocol may replace currently optimized non-DL protocols.

Thin-slice Two-dimensional T2-weighted Imaging with Deep Learning-based Reconstruction: Improved Lesion Detection in the Brain of Patients with Multiple Sclerosis

PURPOSE

Brain MRI with high spatial resolution allows for a more detailed delineation of multiple sclerosis (MS) lesions. The recently developed deep learning-based reconstruction (DLR) technique enables image denoising with sharp edges and reduced artifacts, which improves the image quality of thin-slice 2D MRI. We, therefore, assessed the diagnostic value of 1 mm-slice-thickness 2D T2-weighted imaging (T2WI) with DLR (1 mm T2WI with DLR) compared with conventional MRI for identifying MS lesions.

METHODS

Conventional MRI (5 mm T2WI, 2D and 3D fluid-attenuated inversion recovery) and 1 mm T2WI with DLR (imaging time: 7 minutes) were performed in 42 MS patients. For lesion detection, two neuroradiologists counted the MS lesions in two reading sessions (conventional MRI interpretation with 5 mm T2WI and MRI interpretations with 1 mm T2WI with DLR). The numbers of lesions per region category (cerebral hemisphere, basal ganglia, brain stem, cerebellar hemisphere) were then compared between the two reading sessions.

RESULTS

For the detection of MS lesions by 2 neuroradiologists, the total number of detected MS lesions was significantly higher for MRI interpretation with 1 mm T2WI with DLR than for conventional MRI interpretation with 5 mm T2WI (765 lesions vs. 870 lesions at radiologist A, < 0.05). In particular, of the 33 lesions in the brain stem, radiologist A detected 21 (63.6%) additional lesions by 1 mm T2WI with DLR.

CONCLUSION

Using the DLR technique, whole-brain 1 mm T2WI can be performed in about 7 minutes, which is feasible for routine clinical practice. MRI with 1 mm T2WI with DLR enabled increased MS lesion detection, particularly in the brain stem.

Quality assurance of late gadolinium enhancement cardiac magnetic resonance images: a deep learning classifier for confidence in the presence or absence of abnormality with potential to prompt real-time image optimization

BACKGROUND

Late gadolinium enhancement (LGE) of the myocardium has significant diagnostic and prognostic implications, with even small areas of enhancement being important. Distinguishing between definitely normal and definitely abnormal LGE images is usually straightforward, but diagnostic uncertainty arises when reporters are not sure whether the observed LGE is genuine or not. This uncertainty might be resolved by repetition (to remove artifact) or further acquisition of intersecting images, but this must take place before the scan finishes. Real-time quality assurance by humans is a complex task requiring training and experience, so being able to identify which images have an intermediate likelihood of LGE while the scan is ongoing, without the presence of an expert is of high value. This decision-support could prompt immediate image optimization or acquisition of supplementary images to confirm or refute the presence of genuine LGE. This could reduce ambiguity in reports.

METHODS

Short-axis, phase-sensitive inversion recovery late gadolinium images were extracted from our clinical cardiac magnetic resonance (CMR) database and shuffled. Two, independent, blinded experts scored each individual slice for "LGE likelihood" on a visual analog scale, from 0 (absolute certainty of no LGE) to 100 (absolute certainty of LGE), with 50 representing clinical equipoise. The scored images were split into two classes-either "high certainty" of whether LGE was present or not, or "low certainty." The dataset was split into training, validation, and test sets (70:15:15). A deep learning binary classifier based on the EfficientNetV2 convolutional neural network architecture was trained to distinguish between these categories. Classifier performance on the test set was evaluated by calculating the accuracy, precision, recall, F1-score, and area under the receiver operating characteristics curve (ROC AUC). Performance was also evaluated on an external test set of images from a different center.

RESULTS

One thousand six hundred and forty-five images (from 272 patients) were labeled and split at the patient level into training (1151 images), validation (247 images), and test (247 images) sets for the deep learning binary classifier. Of these, 1208 images were "high certainty" (255 for LGE, 953 for no LGE), and 437 were "low certainty". An external test comprising 247 images from 41 patients from another center was also employed. After 100 epochs, the performance on the internal test set was accuracy = 0.94, recall = 0.80, precision = 0.97, F1-score = 0.87, and ROC AUC = 0.94. The classifier also performed robustly on the external test set (accuracy = 0.91, recall = 0.73, precision = 0.93, F1-score = 0.82, and ROC AUC = 0.91). These results were benchmarked against a reference inter-expert accuracy of 0.86.

CONCLUSION

Deep learning shows potential to automate quality control of late gadolinium imaging in CMR. The ability to identify short-axis images with intermediate LGE likelihood in real-time may serve as a useful decision-support tool. This approach has the potential to guide immediate further imaging while the patient is still in the scanner, thereby reducing the frequency of recalls and inconclusive reports due to diagnostic indecision.

Quantitative double echo steady state T2 mapping of upper extremity peripheral nerves and muscles

Introduction

T2 mapping can characterize peripheral neuropathy and muscle denervation due to axonal damage. Three-dimensional double echo steady-state (DESS) can simultaneously provide 3D qualitative information and T2 maps with equivalent spatial resolution. However, insufficient signal-to-noise ratio may bias DESS-T2 values. Deep learning reconstruction (DLR) techniques can reduce noise, and hence may improve quantitation of high-resolution DESS-T2. This study aims to (i) evaluate the effect of DLR methods on DESS-T2 values, and (ii) to evaluate the feasibility of using DESS-T2 maps to differentiate abnormal from normal nerves and muscles in the upper extremities, with abnormality as determined by electromyography.

Methods and results

Analysis of images from 25 subjects found that DLR decreased DESS-T2 values in abnormal muscles (DLR = 37.71 ± 9.11 msec, standard reconstruction = 38.56 ± 9.44 msec, p = 0.005) and normal muscles (DLR: 27.18 ± 6.34 msec, standard reconstruction: 27.58 ± 6.34 msec, p < 0.001) consistent with a noise reduction bias. Mean DESS-T2, both with and without DLR, was higher in abnormal nerves (abnormal = 75.99 ± 38.21 msec, normal = 35.10 ± 9.78 msec, p < 0.001) and muscles (abnormal = 37.71 ± 9.11 msec, normal = 27.18 ± 6.34 msec, p < 0.001). A higher DESS-T2 in muscle was associated with electromyography motor unit recruitment (p < 0.001).

Discussion

These results suggest that quantitative DESS-T2 is improved by DLR and can differentiate the nerves and muscles involved in peripheral neuropathies from those uninvolved.

Evaluation of Extra-Prostatic Extension on Deep Learning-Reconstructed High-Resolution Thin-Slice T2-Weighted Images in Patients with Prostate Cancer

The aim of this study was to compare diagnostic performance for extra-prostatic extension (EPE) and image quality among three image datasets: conventional T2-weighted images (T2WIconv, slice thickness, 3 mm) and high-resolution thin-slice T2WI (T2WIHR, 2 mm), with and without deep learning reconstruction (DLR) in patients with prostatic cancer (PCa). A total of 88 consecutive patients (28 EPE-positive and 60 negative) diagnosed with PCa via radical prostatectomy who had undergone 3T-MRI were included. Two independent reviewers performed a crossover review in three sessions, in which each reviewer recorded five-point confidence scores for the presence of EPE and image quality using a five-point Likert scale. Pathologic topographic maps served as the reference standard. For both reviewers, T2WIconv showed better diagnostic performance than T2WIHR with and without DLR (AUCs, in order, for reviewer 1, 0.883, 0.806, and 0.772, p = 0.0006; for reviewer 2, 0.803, 0.762, and 0.745, p = 0.022). The image quality was also the best in T2WIconv, followed by T2WIHR with DLR and T2WIHR without DLR for both reviewers (median, in order, 3, 4, and 5, p < 0.0001). In conclusion, T2WIconv was optimal in regard to image quality and diagnostic performance for the evaluation of EPE in patients with PCa.

Fast high-quality MRI protocol of the lumbar spine with deep learning-based algorithm: an image quality and scanning time comparison with standard protocol

OBJECTIVE

The objective of this study is to prospectively compare quantitative and subjective image quality, scanning time, and diagnostic confidence between a new deep learning-based reconstruction(DLR) algorithm and standard MRI protocol of lumbar spine.

MATERIALS AND METHODS

Eighty healthy volunteers underwent 1.5T MRI examination of lumbar spine from September 2021 to May 2023. Protocol acquisition comprised sagittal T1- and T2-weighted fast spin echo and short-tau inversion recovery images and axial multislices T2-weighted fast spin echo images. All sequences were acquired with both DLR algorithm and standard protocols. Two radiologists, blinded to the reconstruction technique, performed quantitative and qualitative image quality analysis in consensus reading; diagnostic confidence was also assessed. Quantitative image quality analysis was assessed by calculating signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR). Qualitative image quality analysis and diagnostic confidence were assessed with a five-point Likert scale. Scanning times were also compared.

RESULTS

DLR SNR was higher in all sequences (all p<0.001). CNR of the DLR was superior to conventional dataset only for axial and sagittal T2-weighted fast spin echo images (p<0.001). Qualitative analysis showed DLR had higher overall quality in all sequences (all p<0.001), with an inter-rater agreement of 0.83 (0.78-0.86). DLR total protocol scanning time was lower compared to standard protocol (6:26 vs 12:59 min, p<0.001). Diagnostic confidence for DLR algorithm was not inferior to standard protocol.

CONCLUSION

DLR applied to 1.5T MRI is a feasible method for lumbar spine imaging providing morphologic sequences with higher image quality and similar diagnostic confidence compared with standard protocol, enabling a remarkable time saving (up to 50%).

Ultrafast Brain MRI Protocol at 1.5 T Using Deep Learning and Multi-shot EPI

RATIONALE AND OBJECTIVES

To evaluate clinical feasibility and image quality of a comprehensive ultrafast brain MRI protocol with multi-shot echo planar imaging and deep learning-enhanced reconstruction at 1.5T.

MATERIALS AND METHODS

Thirty consecutive patients who underwent clinically indicated MRI at a 1.5 T scanner were prospectively included. A conventional MRI (c-MRI) protocol, including T1-, T2-, T2*-, T2-FLAIR, and diffusion-weighted images (DWI)-weighted sequences were acquired. In addition, ultrafast brain imaging with deep learning-enhanced reconstruction and multi-shot EPI (DLe-MRI) was performed. Subjective image quality was evaluated by three readers using a 4-point Likert scale. To assess interrater agreement, Fleiss' kappa (ϰ) was determined. For objective image analysis, relative signal intensity levels for grey matter, white matter, and cerebrospinal fluid were calculated.

RESULTS

Time of acquisition (TA) of c-MRI protocols added up to 13:55 minutes, whereas the TA of DLe-MRI-based protocol added up to 3:04 minutes, resulting in a time reduction of 78%. All DLe-MRI acquisitions yielded diagnostic image quality with good absolute values for subjective image quality. C-MRI demonstrated slight advantages for DWI in overall subjective image quality (c-MRI: 3.93 [+/- 0.25] vs DLe-MRI: 3.87 [+/- 0.37], P = .04) and diagnostic confidence (c-MRI: 3.93 [+/- 0.25] vs DLe-MRI: 3.83 [+/- 3.83], P = .01). For most evaluated quality scores, moderate interobserver agreement was found. Objective image evaluation revealed comparable results for both techniques.

CONCLUSION

DLe-MRI is feasible and allows for highly accelerated comprehensive brain MRI within 3minutes at 1.5 T with good image quality. This technique may potentially strengthen the role of MRI in neurological emergencies.

Clinical feasibility of deep learning reconstruction in liver diffusion-weighted imaging: Improvement of image quality and impact on apparent diffusion coefficient value

PURPOSE

Diffusion-weighted imaging (DWI) of the liver suffers from low resolution, noise, and artifacts. This study aimed to investigate the effect of deep learning reconstruction (DLR) on image quality and apparent diffusion coefficient (ADC) quantification of liver DWI at 3 Tesla.

METHOD

In this prospective study, images of the liver obtained at DWI with b-values of 0 (DWI0), 50 (DWI50) and 800 s/mm2 (DWI800) from consecutive patients with liver lesions from February 2022 to February 2023 were reconstructed with and without DLR (non-DLR). Image quality was assessed qualitatively using Likert scoring system and quantitatively using signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR) and liver/parenchyma boundary sharpness from region-of-interest (ROI) analysis. ADC value of lesion were measured. Phantom experiment was also performed to investigate the factors that determine the effect of DLR on ADC value. Qualitative score, SNR, CNR, boundary sharpness, and apparent diffusion coefficients (ADCs) for DWI were compared using paired t-test and Wilcoxon signed rank test. P < 0.05 was considered statistically significant.

RESULTS

A total of 85 patients with 170 lesions were included. DLR group showed a higher qualitative score than the non-DLR group. for example, with DWI800 the score was 4.77 ± 0.52 versus 4.30 ± 0.63 (P < 0.001). DLR group also showed higher SNRs, CNRs and boundary sharpness than the non-DLR group. DLR reduced the ADC of malignant tumors (1.105[0.904, 1.340] versus 1.114[0.904, 1.320]) (P < 0.001), but there was no significant difference in the diagnostic value of malignancy for DLR and non-DLR groups (P = 57.3). The phantom study confirmed a reduction of ADC in images with low resolution, and a stronger reduction of ADC in heterogeneous structures than in homogeneous ones (P < 0.001).

CONCLUSIONS

DLR improved image quality of liver DWI. DLR reduced the ADC value of lesions, but did not affect the diagnostic performance of ADC in distinguishing malignant tumors on a 3.0-T MRI system.

Evaluation of a Deep Learning Reconstruction for High-Quality T2-Weighted Breast Magnetic Resonance Imaging

Deep learning (DL) reconstruction techniques to improve MR image quality are becoming commercially available with the hope that they will be applicable to multiple imaging application sites and acquisition protocols. However, before clinical implementation, these methods must be validated for specific use cases. In this work, the quality of standard-of-care (SOC) T2w and a high-spatial-resolution (HR) imaging of the breast were assessed both with and without prototype DL reconstruction. Studies were performed using data collected from phantoms, 20 retrospectively collected SOC patient exams, and 56 prospectively acquired SOC and HR patient exams. Image quality was quantitatively assessed via signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and edge sharpness. Qualitatively, all in vivo images were scored by either two or four radiologist readers using 5-point Likert scales in the following categories: artifacts, perceived sharpness, perceived SNR, and overall quality. Differences in reader scores were tested for significance. Reader preference and perception of signal intensity changes were also assessed. Application of the DL resulted in higher average SNR (1.2-2.8 times), CNR (1.0-1.8 times), and image sharpness (1.2-1.7 times). Qualitatively, the SOC acquisition with DL resulted in significantly improved image quality scores in all categories compared to non-DL images. HR acquisition with DL significantly increased SNR, sharpness, and overall quality compared to both the non-DL SOC and the non-DL HR images. The acquisition time for the HR data only required a 20% increase compared to the SOC acquisition and readers typically preferred DL images over non-DL counterparts. Overall, the DL reconstruction demonstrated improved T2w image quality in clinical breast MRI.

Evaluation of late gadolinium enhancement cardiac MRI using deep learning reconstruction

BACKGROUND

Deep learning (DL)-based methods have been used to improve the imaging quality of magnetic resonance imaging (MRI) by denoising.

PURPOSE

To assess the effects of DL-based MR reconstruction (DLR) method on late gadolinium enhancement (LGE) image quality.

MATERIAL AND METHODS

A total of 85 patients who underwent cardiovascular magnetic resonance (CMR) examination, including LGE imaging using conventional construction and DLR with varying levels of noise reduction (NR) levels, were included. Both magnitude LGE (MLGE) and phase-sensitive LGE (PSLGE) images were reviewed independently by double-blinded observers who used a 5-point Likert scale for multiple measures regarding image quality. Meanwhile, the signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and edge sharpness of images were calculated and compared between conventional LGE imaging and DLR LGE imaging.

RESULTS

Both MLGE and PSLGE with DLR at 50% and 75% noise reduction levels received significantly higher scores than conventional imaging for overall imaging quality (all P < 0.01). In addition, the SNR, CNR, and edge sharpness of all DLR LGE imaging are higher than conventional imaging (all P < 0.01). The highest subjective score and best image quality is obtained when the DLR noise reduction level is at 75%.

CONCLUSION

DLR reduced image noise while improving image contrast and sharpness in the cardiovascular LGE imaging.

Usefulness of T2-Weighted Images with Deep-Learning-Based Reconstruction in Nasal Cartilage

OBJECTIVE

This study aims to evaluate the feasibility of visualizing nasal cartilage using deep-learning-based reconstruction (DLR) fast spin-echo (FSE) imaging in comparison to three-dimensional fast spoiled gradient-echo (3D FSPGR) images.

MATERIALS AND METHODS

This retrospective study included 190 set images of 38 participants, including axial T1- and T2-weighted FSE images using DLR (T1WIDL and T2WIDL, belong to FSEDL) and without using DLR (T1WIO and T2WIO, belong to FSEO) and 3D FSPGR images. Subjective evaluation (overall image quality, noise, contrast, artifacts, and identification of anatomical structures) was independently conducted by two radiologists. Objective evaluation including signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) was conducted using manual region-of-interest (ROI)-based analysis. Coefficient of variation (CV) and Bland-Altman plots were used to demonstrate the intra-rater repeatability of measurements for cartilage thickness on five different images.

RESULTS

Both qualitative and quantitative results confirmed superior FSEDL to 3D FSPGR images (both p < 0.05), improving the diagnosis confidence of the observers. Lower lateral cartilage (LLC), upper lateral cartilage (ULC), and septal cartilage (SP) were relatively well delineated on the T2WIDL, while 3D FSPGR showed poorly on the septal cartilage. For the repeatability of cartilage thickness measurements, T2WIDL showed the highest intra-observer (%CV = 8.7% for SP, 9.5% for ULC, and 9.7% for LLC) agreements. In addition, the acquisition time for T1WIDL and T2WIDL was respectively reduced by 14.2% to 29% compared to 3D FSPGR (both p < 0.05).

CONCLUSIONS

Two-dimensional equivalent-thin-slice T1- and T2-weighted images using DLR showed better image quality and shorter scan time than 3D FSPGR and conventional construction images in nasal cartilages. The anatomical details were preserved without losing clinical performance on diagnosis and prognosis, especially for pre-rhinoplasty planning.

A Feasibility Study on Deep Learning Reconstruction to Improve Image Quality With PROPELLER Acquisition in the Setting of T2-Weighted Gynecologic Pelvic Magnetic Resonance Imaging

OBJECTIVES

Evaluate deep learning (DL) to improve the image quality of the PROPELLER (Periodically Rotated Overlapping Parallel Lines with Enhanced Reconstruction technique) for 3 T magnetic resonance imaging of the female pelvis.

METHODS

Three radiologists prospectively and independently compared non-DL and DL PROPELLER sequences from 20 patients with a history of gynecologic malignancy. Sequences with different noise reduction factors (DL 25%, DL 50%, and DL 75%) were blindly reviewed and scored based on artifacts, noise, relative sharpness, and overall image quality. The generalized estimating equation method was used to assess the effect of methods on the Likert scales. Quantitatively, the contrast-to-noise ratio and signal-to-noise ratio (SNR) of the iliac muscle were calculated, and pairwise comparisons were performed based on a linear mixed model. P values were adjusted using the Dunnett method. Interobserver agreement was assessed using the κ statistic. P value was considered statistically significant at less than 0.05.

RESULTS

Qualitatively, DL 50 and DL 75 were ranked as the best sequences in 86% of cases. Images generated by the DL method were significantly better than non-DL images ( P < 0.0001). Iliacus muscle SNR on DL 50 and DL 75 was significantly better than non-DL images ( P < 0.0001). There was no difference in contrast-to-noise ratio between the DL and non-DL techniques in the iliac muscle. There was a high percent agreement (97.1%) in terms of DL sequences' superior image quality (97.1%) and sharpness (100%) relative to non-DL images.

CONCLUSION

The utilization of DL reconstruction improves the image quality of PROPELLER sequences with improved SNR quantitatively.

Dark-Blood Late Gadolinium Enhancement MRI Is Noninferior to Bright-Blood LGE in Non-Ischemic Cardiomyopathies

(1) Background and Objectives: Dark-blood late gadolinium enhancement has been shown to be a reliable cardiac magnetic resonance (CMR) method for assessing viability and depicting myocardial scarring in ischemic cardiomyopathy. The aim of this study was to evaluate dark-blood LGE imaging compared with conventional bright-blood LGE for the detection of myocardial scarring in non-ischemic cardiomyopathies. (2) Materials and Methods: Patients with suspected non-ischemic cardiomyopathy were prospectively enrolled in this single-centre study from January 2020 to March 2023. All patients underwent 1.5 T CMR with both dark-blood and conventional bright-blood LGE imaging. Corresponding short-axis stacks of both techniques were analysed for the presence, distribution, pattern, and localisation of LGE, as well as the quantitative scar size (%). (3) Results: 343 patients (age 44 ± 17 years; 124 women) with suspected non-ischemic cardiomyopathy were examined. LGE was detected in 123 of 343 cases (36%) with excellent inter-reader agreement (κ 0.97-0.99) for both LGE techniques. Dark-blood LGE showed a sensitivity of 99% (CI 98-100), specificity of 99% (CI 98-100), and an accuracy of 99% (CI 99-100) for the detection of non-ischemic scarring. No significant difference in total scar size (%) was observed. Dark-blood imaging with mean 5.35 ± 4.32% enhanced volume of total myocardial volume, bright-blood with 5.24 ± 4.28%, p = 0.84. (4) Conclusions: Dark-blood LGE imaging is non-inferior to conventional bright-blood LGE imaging in detecting non-ischemic scarring. Therefore, dark-blood LGE imaging may become an equivalent method for the detection of both ischemic and non-ischemic scars.

High-Resolution Single-Shot Fast Spin-Echo MR Imaging with Deep Learning Reconstruction Algorithm Can Improve Repeatability and Reproducibility of Follicle Counting

OBJECTIVE

To investigate the diagnostic performance of high-resolution single-shot fast spin-echo (SSFSE) imaging with deep learning (DL) reconstruction algorithm on follicle counting and compare it with original SSFSE images and conventional fast spin-echo (FSE) images.

METHODS

This study included 20 participants (40 ovaries) with clinically confirmed polycystic ovary syndrome (PCOS) who underwent high-resolution ovary MRI, including three-plane T2-weighted FSE sequences and slice-matched T2-weighted SSFSE sequences. A DL reconstruction algorithm was applied to the SSFSE sequences to generate SSFSE-DL images, and the original SSFSE images were also saved. Subjective evaluations such as the blurring artifacts, subjective noise, and clarity of the follicles on the SSFSE-DL, SSFSE, and conventional FSE images were independently conducted by two observers. Intra-class correlation coefficients and Bland-Altman plots were used to present the repeatability and reproducibility of the follicle number per ovary (FNPO) based on the three types of images.

RESULTS

SSFSE-DL images showed less blurring artifact, subjective noise, and better clarity of the follicles than SSFSE and FSE (p < 0.05). For the repeatability of the FNPO, SSFSE-DL showed the highest intra-observer (ICC = 0.930; 95% CI: 0.878-0.962) and inter-observer (ICC = 0.914; 95% CI: 0.843-0.953) agreements. The inter-observer 95% limits of agreement (LOA) for SSFSE-DL, SSFSE, and FSE ranged from -3.7 to 4.5, -4.4 to 7.0, and -7.1 to 7.6, respectively. The intra-observer 95% LOA for SSFSE-DL, SSFSE, and FSE ranged from -3.5 to 4.0, -5.1 to 6.1, and -5.7 to 4.2, respectively. The absolute values of intra-observer and inter-observer differences for SSFSE-DL were significantly lower than those for SSFSE and FSE (p < 0.05).

CONCLUSIONS

Compared with the original SSFSE images and the conventional FSE images, high-resolution SSFSE images with DL reconstruction algorithm can better display follicles, thus improving FNPO assessment.

Comparison of deep learning-based reconstruction of PROPELLER Shoulder MRI with conventional reconstruction

OBJECTIVE

To compare the image quality and agreement among conventional and accelerated periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER) MRI with both conventional reconstruction (CR) and deep learning-based reconstruction (DLR) methods for evaluation of shoulder.

MATERIALS AND METHODS

We included patients who underwent conventional (acquisition time, 8 min) and accelerated (acquisition time, 4 min and 24 s; 45% reduction) PROPELLER shoulder MRI using both CR and DLR methods between February 2021 and February 2022 on a 3 T MRI system. Quantitative evaluation was performed by calculating the signal-to-noise ratio (SNR). Two musculoskeletal radiologists compared the image quality using conventional sequence with CR as the reference standard. Interobserver agreement between image sets for evaluating shoulder was analyzed using weighted/unweighted kappa statistics.

RESULTS

Ninety-two patients with 100 shoulder MRI scans were included. Conventional sequence with DLR had the highest SNR (P < .001), followed by accelerated sequence with DLR, conventional sequence with CR, and accelerated sequence with CR. Comparison of image quality by both readers revealed that conventional sequence with DLR (P = .003 and P < .001) and accelerated sequence with DLR (P = .016 and P < .001) had better image quality than the conventional sequence with CR. Interobserver agreement was substantial to almost perfect for detecting shoulder abnormalities (κ = 0.600-0.884). Agreement between the image sets was substantial to almost perfect (κ = 0.691-1).

CONCLUSION

Accelerated PROPELLER with DLR showed even better image quality than conventional PROPELLER with CR and interobserver agreement for shoulder pathologies comparable to that of conventional PROPELLER with CR, despite the shorter scan time.

Deep Learning-Based Reconstruction for Cardiac MRI: A Review

Cardiac magnetic resonance (CMR) is an essential clinical tool for the assessment of cardiovascular disease. Deep learning (DL) has recently revolutionized the field through image reconstruction techniques that allow unprecedented data undersampling rates. These fast acquisitions have the potential to considerably impact the diagnosis and treatment of cardiovascular disease. Herein, we provide a comprehensive review of DL-based reconstruction methods for CMR. We place special emphasis on state-of-the-art unrolled networks, which are heavily based on a conventional image reconstruction framework. We review the main DL-based methods and connect them to the relevant conventional reconstruction theory. Next, we review several methods developed to tackle specific challenges that arise from the characteristics of CMR data. Then, we focus on DL-based methods developed for specific CMR applications, including flow imaging, late gadolinium enhancement, and quantitative tissue characterization. Finally, we discuss the pitfalls and future outlook of DL-based reconstructions in CMR, focusing on the robustness, interpretability, clinical deployment, and potential for new methods.

Deep learning reconstruction in pediatric brain MRI: comparison of image quality with conventional T2-weighted MRI

INTRODUCTION

Deep learning-based MRI reconstruction has recently been introduced to improve image quality. This study aimed to evaluate the performance of deep learning reconstruction in pediatric brain MRI.

METHODS

A total of 107 consecutive children who underwent 3.0 T brain MRI were included in this study. T2-weighted brain MRI was reconstructed using the three different reconstruction modes: deep learning reconstruction, conventional reconstruction with an intensity filter, and original T2 image without a filter. Two pediatric radiologists independently evaluated the following image quality parameters of three reconstructed images on a 5-point scale: overall image quality, image noisiness, sharpness of gray-white matter differentiation, truncation artifact, motion artifact, cerebrospinal fluid and vascular pulsation artifacts, and lesion conspicuity. The subjective image quality parameters were compared among the three reconstruction modes. Quantitative analysis of the signal uniformity using the coefficient of variation was performed for each reconstruction.

RESULTS

The overall image quality, noisiness, and gray-white matter sharpness were significantly better with deep learning reconstruction than with conventional or original reconstruction (all P < 0.001). Deep learning reconstruction had significantly fewer truncation artifacts than the other two reconstructions (all P < 0.001). Motion and pulsation artifacts showed no significant differences among the three reconstruction modes. For 36 lesions in 107 patients, lesion conspicuity was better with deep learning reconstruction than original reconstruction. Deep learning reconstruction showed lower signal variation compared to conventional and original reconstructions.

CONCLUSION

Deep learning reconstruction can reduce noise and truncation artifacts and improve lesion conspicuity and overall image quality in pediatric T2-weighted brain MRI.

Accelerated Synthetic MRI with Deep Learning-Based Reconstruction for Pediatric Neuroimaging

BACKGROUND AND PURPOSE

Synthetic MR imaging is a time-efficient technique. However, its rather long scan time can be challenging for children. This study aimed to evaluate the clinical feasibility of accelerated synthetic MR imaging with deep learning-based reconstruction in pediatric neuroimaging and to investigate the impact of deep learning-based reconstruction on image quality and quantitative values in synthetic MR imaging.

MATERIALS AND METHODS

This study included 47 children 2.3-14.7 years of age who underwent both standard and accelerated synthetic MR imaging at 3T. The accelerated synthetic MR imaging was reconstructed using a deep learning pipeline. The image quality, lesion detectability, tissue values, and brain volumetry were compared among accelerated deep learning and accelerated and standard synthetic data sets.

RESULTS

The use of deep learning-based reconstruction in the accelerated synthetic scans significantly improved image quality for all contrast weightings (P < .001), resulting in image quality comparable with or superior to that of standard scans. There was no significant difference in lesion detectability between the accelerated deep learning and standard scans (P > .05). The tissue values and brain tissue volumes obtained with accelerated deep learning and the other 2 scans showed excellent agreement and a strong linear relationship (all, R 2 > 0.9). The difference in quantitative values of accelerated scans versus accelerated deep learning scans was very small (tissue values, <0.5%; volumetry, -1.46%-0.83%).

CONCLUSIONS

The use of deep learning-based reconstruction in synthetic MR imaging can reduce scan time by 42% while maintaining image quality and lesion detectability and providing consistent quantitative values. The accelerated deep learning synthetic MR imaging can replace standard synthetic MR imaging in both contrast-weighted and quantitative imaging.

Artificial intelligence based image quality enhancement in liver MRI: a quantitative and qualitative evaluation

Purpose

To compare liver MRI with AIR Recon Deep Learning™(ARDL) algorithm applied and turned-off (NON-DL) with conventional high-resolution acquisition (NAÏVE) sequences, in terms of quantitative and qualitative image analysis and scanning time.

Material and methods

This prospective study included fifty consecutive volunteers (31 female, mean age 55.5 ± 20 years) from September to November 2021. 1.5 T MRI was performed and included three sets of images: axial single-shot fast spin-echo (SSFSE) T2 images, diffusion-weighted images(DWI) and apparent diffusion coefficient(ADC) maps acquired with both ARDL and NAÏVE protocol; the NON-DL images, were also assessed. Two radiologists in consensus drew fixed regions of interest in liver parenchyma to calculate signal-to-noise-ratio (SNR) and contrast to-noise-ratio (CNR). Subjective image quality was assessed by two other radiologists independently with a five-point Likert scale. Acquisition time was recorded.

Results

SSFSE T2 objective analysis showed higher SNR and CNR for ARDL vs NAÏVE, ARDL vs NON-DL(all P < 0.013). Regarding DWI, no differences were found for SNR with ARDL vs NAÏVE and, ARDL vs NON-DL (all P > 0.2517).CNR was higher for ARDL vs NON-DL(P = 0.0170), whereas no differences were found between ARDL and NAÏVE(P = 1). No differences were observed for all three comparisons, in terms of SNR and CNR, for ADC maps (all P > 0.32).

Qualitative analysis for all sequences showed better overall image quality for ARDL with lower truncation artifacts, higher sharpness and contrast (all P < 0.0070) with excellent inter-rater agreement (k ≥ 0.8143). Acquisition time was lower in ARDL sequences compared to NAÏVE (SSFSE T2 = 19.08 ± 2.5 s vs. 24.1 ± 2 s and DWI = 207.3 ± 54 s vs. 513.6 ± 98.6 s, all P < 0.0001).

Conclusion

ARDL applied on upper abdomen showed overall better image quality and reduced scanning time compared with NAÏVE protocol.

Compact pediatric cardiac magnetic resonance imaging protocols

Cardiac MRI is in many respects an ideal modality for pediatric cardiovascular imaging, enabling a complete noninvasive assessment of anatomy, morphology, function and flow in one radiation-free and potentially non-contrast exam. Nonetheless, traditionally lengthy and complex imaging acquisition strategies have often limited its broader use beyond specialized centers. In this review, the author presents practical cardiac MRI imaging protocols to facilitate the performance of succinct yet successful exams that provide the most salient clinical data for the majority of congenital and acquired pediatric cardiac disease. In addition, the author reviews newer and evolving techniques that permit more rapid but similarly diagnostic MRI, including compressed sensing and artificial intelligence/machine learning reconstruction, four-dimensional flow acquisition and blood pool contrast agents. With the modern armamentarium of cardiac MRI methods, the goal of compact yet comprehensive exams in children can now be realized.

The Applications of Artificial Intelligence in Cardiovascular Magnetic Resonance-A Comprehensive Review

Cardiovascular disease remains an integral field on which new research in both the biomedical and technological fields is based, as it remains the leading cause of mortality and morbidity worldwide. However, despite the progress of cardiac imaging techniques, the heart remains a challenging organ to study. Artificial intelligence (AI) has emerged as one of the major innovations in the field of diagnostic imaging, with a dramatic impact on cardiovascular magnetic resonance imaging (CMR). AI will be increasingly present in the medical world, with strong potential for greater diagnostic efficiency and accuracy. Regarding the use of AI in image acquisition and reconstruction, the main role was to reduce the time of image acquisition and analysis, one of the biggest challenges concerning magnetic resonance; moreover, it has been seen to play a role in the automatic correction of artifacts. The use of these techniques in image segmentation has allowed automatic and accurate quantification of the volumes and masses of the left and right ventricles, with occasional need for manual correction. Furthermore, AI can be a useful tool to directly help the clinician in the diagnosis and derivation of prognostic information of cardiovascular diseases. This review addresses the applications and future prospects of AI in CMR imaging, from image acquisition and reconstruction to image segmentation, tissue characterization, diagnostic evaluation, and prognostication.

Evaluation of deep learning reconstructed high-resolution 3D lumbar spine MRI

OBJECTIVES

To compare interobserver agreement and image quality of 3D T2-weighted fast spin echo (T2w-FSE) L-spine MRI images processed with a deep learning reconstruction (DLRecon) against standard-of-care (SOC) reconstruction, as well as against 2D T2w-FSE images. The hypothesis was that DLRecon 3D T2w-FSE would afford improved image quality and similar interobserver agreement compared to both SOC 3D and 2D T2w-FSE.

METHODS

Under IRB approval, patients who underwent routine 3-T lumbar spine (L-spine) MRI from August 17 to September 17, 2020, with both isotropic 3D and 2D T2w-FSE sequences, were retrospectively included. A DLRecon algorithm, with denoising and sharpening properties was applied to SOC 3D k-space to generate 3D DLRecon images. Four musculoskeletal radiologists blinded to reconstruction status evaluated randomized images for motion artifact, image quality, central/foraminal stenosis, disc degeneration, annular fissure, disc herniation, and presence of facet joint cysts. Inter-rater agreement for each graded variable was evaluated using Conger's kappa (κ).

RESULTS

Thirty-five patients (mean age 58 ± 19, 26 female) were evaluated. 3D DLRecon demonstrated statistically significant higher median image quality score (2.0/2) when compared to SOC 3D (1.0/2, p < 0.001), 2D axial (1.0/2, p < 0.001), and 2D sagittal sequences (1.0/2, p value < 0.001). κ ranges (and 95% CI) for foraminal stenosis were 0.55-0.76 (0.32-0.86) for 3D DLRecon, 0.56-0.73 (0.35-0.84) for SOC 3D, and 0.58-0.71 (0.33-0.84) for 2D. Mean κ (and 95% CI) for central stenosis at L4-5 were 0.98 (0.96-0.99), 0.97 (0.95-0.99), and 0.98 (0.96-0.99) for 3D DLRecon, 3D SOC and 2D, respectively.

CONCLUSIONS

DLRecon 3D T2w-FSE L-spine MRI demonstrated higher image quality and similar interobserver agreement for graded variables of interest when compared to 3D SOC and 2D imaging.

KEY POINTS

• 3D DLRecon T2w-FSE isotropic lumbar spine MRI provides improved image quality when compared to 2D MRI, with similar interobserver agreement for clinical evaluation of pathology. • 3D DLRecon images demonstrated better image quality score (2.0/2) when compared to standard-of-care (SOC) 3D (1.0/2), p value < 0.001; 2D axial (1.0/2), p value < 0.001; and 2D sagittal sequences (1.0/2), p value < 0.001. • Interobserver agreement for major variables of interest was similar among all sequences and reconstruction types. For foraminal stenosis, κ ranged from 0.55 to 0.76 (95% CI 0.32-0.86) for 3D DLRecon, 0.56-0.73 (95% CI 0.35-0.84) for standard-of-care (SOC) 3D, and 0.58-0.71 (95% CI 0.33-0.84) for 2D.

Impact of Deep Learning Reconstruction Combined With a Sharpening Filter on Single-Shot Fast Spin-Echo T2-Weighted Magnetic Resonance Imaging of the Uterus

OBJECTIVE

This study aimed to evaluate the effects of deep learning (DL) reconstruction and a postprocessing sharpening filter on the image quality of single-shot fast spin-echo (SSFSE) T2-weighted imaging (T2WI) of the uterus.

MATERIALS AND METHODS

Fifty consecutive patients who underwent pelvic magnetic resonance imaging were included. Parasagittal T2WI with a slice thickness of 4 mm was obtained with the periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER) and SSFSE sequences (mean scan time, 204 and 22 seconds, respectively). The following 3 types of SSFSE images were reconstructed, and the signal-to-noise ratio (SNR) and tissue contrast were assessed: conventional reconstruction (SSFSE-C), DL reconstruction (SSFSE-DL), and DL with a sharpening filter (SSFSE-DLF). Three radiologists independently assessed image quality, and area under the visual grading characteristics curve (AUCVGC) analysis was performed to compare the SSFSE and PROPELLER images.

RESULTS

Compared with that of the PROPELLER images, the SNR of the SSFSE-C, SSFSE-DL, and SSFSE-DLF images was significantly lower (P < 0.05), significantly higher (P < 0.05), and equivalent, respectively. The SSFSE-DL images exhibited significantly lower contrast between the junctional zone and myometrium than those obtained with the other sequences (P < 0.05). In qualitative comparisons with the PROPELLER images, all 3 SSFSE sequences, SSFSE-DL, and SSFSE-DLF demonstrated significantly higher scores for artifacts, noise, and sharpness, respectively (P < 0.01). The overall image quality of SSFSE-C (mean AUCVGC, 0.03; P < 0.01) and SSFSE-DL (mean AUCVGC, 0.23; P < 0.01) was rated as significantly inferior, whereas that of SSFSE-DLF (mean AUCVGC, 0.69) was equivalent or significantly higher (P < 0.01).

CONCLUSION

Using a combination of DL and a sharpening filter markedly increases the image quality of SSFSE of the uterus to the level of the PROPELLER sequence.

Linear late gadolinium enhancement in the basal anterior septum and lateral wall may represent the contrast enhancement of vessels: A CMR and CCTA comparison study

BACKGROUND

The purpose of this paper was to verify that the linear high-intensity signal on late gadolinium enhancement-cardiac magnetic resonance (LGE-CMR) may represent the contrast enhancement of vessels rather than scars or fibrosis, and to assess whether this linear high-intensity signal will affect the quantification of myocardial fibrosis in patients with hypertrophic cardiomyopathy (HCM).

METHODS

A total of 58 patients who underwent both coronary computed tomography angiography (CCTA) and LGE-CMR in our hospital were ultimately enrolled. The definitions of positive linear LGE (LLGE+) were as follows: (1) LLGE in the basal anterior septum or lateral wall, and (2) LLGE observable at 10 mm or more. All other patients were regarded as negative LLGE (LLGE-). In LLGE+ patients, the length of the LLGE located in the anterior septum and lateral wall was compared with the length of the septal perforator artery and the circumflex artery on CCTA, respectively. For nine patients with HCM, the LGE% was measured before and after removal of LLGE.

RESULTS

Among the 58 patients, 40 showed LLGE+ and 18 showed LLGE-. For patients with LLGE in the anterior septum, there was a strong correlation between LLGE and anterior septal perforator arteries in length (r=0.887, p<0.001). For patients with LLGE in the lateral wall, LLGE also correlated well with the circumflex arteries in length (r=0.962, p<0.001). In nine patients with HCM, the LGE% decreased significantly after the removal of LLGE [9.50 (7.70 - 17.35)% vs. 8.80 (6.20 - 15.55)%, p<0.05].

CONCLUSIONS

The LLGE in the anterior septum and lateral wall may represent contrast enhancement of the anterior septal perforator artery and the circumflex artery, respectively. This LLGE may overestimate the extent of myocardial fibrosis in patients with HCM.

Feasibility of late gadolinium enhancement (LGE) in ischemic cardiomyopathy using 2D-multisegment LGE combined with artificial intelligence reconstruction deep learning noise reduction algorithm

BACKGROUND

Despite the low spatial resolution of 2D-multisegment late gadolinium enhancement (2D-MSLGE) sequences, it may be useful in uncooperative patients instead of standard 2D single segmented inversion recovery gradient echo late gadolinium enhancement sequences (2D-SSLGE). The aim of the study is to assess the feasibility and comparison of 2D-MSLGE reconstructed with artificial intelligence reconstruction deep learning noise reduction (NR) algorithm compared to standard 2D-SSLGE in consecutive patients with ischemic cardiomyopathy (ICM).

METHODS

Fifty-seven patients with known ICM referred for a clinically indicated CMR were enrolled in this study. 2D-MSLGE were reconstructed using a growing level of NR (0%,25%,50%,75%and 100%). Subjective image quality, signal to noise ratio (SNR) and contrast to noise ratio (CNR) were evaluated in each dataset and compared to standard 2D-SSLGE. Moreover, diagnostic accuracy, LGE mass and scan time were compared between 2D-MSLGE with NR and 2D-SSLGE.

RESULTS

The application of NR reconstruction ≥50% to 2D-MSLGE provided better subjective image quality, CNR and SNR compared to 2D-SSLGE (p < 0.01). The best compromise in terms of subjective and objective image quality was observed for values of 2D-MSLGE 75%, while no differences were found in terms of LGE quantification between 2D-MSLGE versus 2D-SSLGE, regardless the NR applied. The sensitivity, specificity, negative predictive value, positive predictive value and accuracy of 2D-MSLGE NR 75% were 87.77%,96.27%,96.13%,88.16% and 94.22%, respectively. Time of acquisition of 2D-MSLGE was significantly shorter compared to 2D-SSLGE (p < 0.01).

CONCLUSION

When compared to standard 2D-SSLGE, the application of NR reconstruction to 2D-MSLGE provides superior image quality with similar diagnostic accuracy.

Image Quality and Diagnostic Performance of Accelerated Shoulder MRI With Deep Learning-Based Reconstruction

Background: Shoulder MRI using standard multiplanar sequences requires long scan times. Accelerated sequences have tradeoffs in noise and resolution. Deep learning-based reconstruction (DLR) may allow reduced scan time with preserved image quality. Objectives: To compare standard shoulder MRI sequences and accelerated sequences without and with DLR in terms of image quality and diagnostic performance. Methods: This retrospective study included 105 patients (45 men, 60 women; mean age 57.6±10.9 years) who underwent a total of 110 3-T shoulder MRI examinations. Examinations included standard sequences (scan time, 9 minutes 23 seconds) and accelerated sequences (3 minutes 5 seconds; 67% reduction), both including fast spin echo sequences in three planes. Standard sequences were reconstructed using the conventional pipeline; accelerated sequences were reconstructed using both conventional pipeline and a commercially available DLR pipeline. Two radiologists independently assessed three image sets (standard, accelerated without DLR, accelerated with DLR) for subjective image quality and artifacts using 4-point scales (4=highest quality), and identified pathologies of subscapularis tendon, supraspinatus-infraspinatus tendon, biceps brachii long head tendon, and glenoid labrum. Interobserver and inter-image set agreement for the evaluated pathologies was assessed using weighted kappa statistics. In 27 patients who underwent arthroscopy, diagnostic performance was calculated using arthroscopic findings as reference. Results: Mean subjective image quality for readers 1 and 2 was 10.6±1.2 and 10.5±1.4 for standard, 8.1±1.3 and 7.2±1.1 for accelerated without DLR, and 10.7±1.2 and 10.5±1.6 for accelerated with DLR. Mean artifact score for readers 1 and 2 was 9.3±1.2 and 10.0±1.0 for standard, 7.3±1.3 and 9.1±0.8 for accelerated without DLR, and 9.4±1.2 and 9.8±0.8 for accelerated with DLR. Interobserver agreement ranged from kappa=0.813-0.951 except for accelerated without DLR for SST-IST (κ=0.673). Inter-image set agreement ranged from kappa=0.809-0.957 except for reader 1 for SST-IST (κ=0.663-0.700). Accuracy, sensitivity, and specificity for tears of the four structures was not different (p>.05) among image sets. Conclusions: Accelerated sequences with DLR provide 67% scan time reduction with similar subjective image quality, artifacts, and diagnostic performance as standard sequences. Clinical impact: Accelerated sequences with DLR may provide an alternative to standard sequences for clinical shoulder MRI.