To prospectively compare image quality and infarct sizing methods between magnetization-transfer "flow-independent dark-blood delayed enhancement" (MT-FIDDLE) and standard "bright-blood"-late gadolinium enhancement (LGE) cardiac-magnetic-resonance (CMR) sequence.

"Bright-blood"-LGE and MT-FIDDLE sequence were acquired in 110 patients at 4 days (n = 33), 4 months (n = 39) and 12 months (n = 38) after acute ST-elevation myocardial infarction (STEMI). Subjective image quality, including confidence in infarct segmentation and blood-pool bordering, were each rated on a 4-point Likert scale. Objective image quality was assessed by the detectability index (DI). Infarct volumes derived via full-width at half-maximum (FWHM) and different number of standard deviations ("n-SD") methods on MT-FIDDLE images were compared with FWHM and reference 5-SD results from "bright-blood-LGE images.

Overall subjective median image quality was excellent for both LGE sequences. Qualitative analysis revealed a significantly higher confidence in infarct segmentation and in blood-pool bordering for MT-FIDDLE as compared to "bright-blood"-LGE (all p < 0.001). Infarct volumes assessed by the FWHM technique on MT-FIDDLE and "bright-blood"-LGE showed excellent agreement overall (Concordance correlation coefficient, CCC = 0.96). The 3-SD technique for MT-FIDDLE showed the best agreement with the 5-SD method for "bright-blood"-LGE overall (CCC = 0.94), as well as in the subgroup with excellent confidence in infarct segmentation on "bright-blood"-LGE (CCC = 0.96). DI of scar versus LV blood-pool was higher for MT-FIDDLE (8.9 ± 5.5) compared to "bright-blood"-LGE sequence (2.0 ± 1.5; p < 0.001).

MT-FIDDLE significantly optimizes the discrimination between myocardial infarction and adjacent blood-pool in STEMI patients. As compared to the established 5-SD technique on "bright-blood"-LGE, the 3-SD method on MT-FIDDLE results in consistent infarct volumes.

Question Does magnetization-transfer "flow-independent dark-blood delayed enhancement" (MT-FIDDLE) offer any benefits over standard "bright-blood"-late gadolinium enhancement (LGE) cardiac-magnetic-resonance (CMR) for identifying STEMI infarct borders? Findings MT-FIDDLE image quality was higher than LGE CMR and measured infarct volume comparability to the standard 5-SD-threshold-technique. Clinical relevance MT-FIDDLE facilitates the assessment of myocardial infarctions at the subendocardial border, improving the discrimination between myocardial infarction and adjacent blood-pool in STEMI patients.

To develop a novel, free-breathing, 3D jointT 1 $$ {T}_1 $$ /T 1 ρ $$ {T}_{1\rho } $$ /T 2 $$ {T}_2 $$ mapping sequence with Dixon encoding to provide co-registered 3DT 1 $$ {T}_1 $$ ,T 1 ρ $$ {T}_{1\rho } $$ , andT 2 $$ {T}_2 $$ maps and water-fat volumes with isotropic spatial resolution in a single≈ 7 $$ \approx 7 $$ min scan for comprehensive contrast-agent-free myocardial tissue characterization and simultaneous evaluation of the whole-heart anatomy.

An interleaving sequence over 5 heartbeats is proposed to provideT 1 $$ {T}_1 $$ ,T 1 ρ $$ {T}_{1\rho } $$ , andT 2 $$ {T}_2 $$ encoding, with 3D data acquired with Dixon gradient-echo readout and 2D image navigators to enable100 % $$ 100\% $$ respiratory scan efficiency. Images were reconstructed with a non-rigid motion-corrected, low-rank patch-based reconstruction, and maps were generated through dictionary matching. The proposed sequence was compared against conventional 2D techniques in phantoms, 10 healthy subjects, and 1 patient.

The proposed 3DT 1 $$ {T}_1 $$ ,T 1 ρ $$ {T}_{1\rho } $$ , andT 2 $$ {T}_2 $$ measurements showed excellent correlation with 2D reference measurements in phantoms. For healthy subjects, the mapping values of septal myocardial tissue wereT 1 = 1060 ± 48 ms $$ {T}_1=1060\pm 48\kern0.2778em \mathrm{ms} $$ ,T 1 ρ = 48 . 1 ± 3 . 9 ms $$ {T}_{1\rho }=48.1\pm 3.9\kern0.2778em \mathrm{ms} $$ , andT 2 = 44 . 2 ± 3 . 2 ms $$ {T}_2=44.2\pm 3.2\kern0.2778em \mathrm{ms} $$ for the proposed sequence, againstT 1 = 959 ± 15 ms $$ {T}_1=959\pm 15\kern0.2778em \mathrm{ms} $$ ,T 1 ρ = 56 . 4 ± 1 . 9 ms $$ {T}_{1\rho }=56.4\pm 1.9\kern0.2778em \mathrm{ms} $$ , andT 2 = 47 . 3 ± 1 . 5 ms $$ {T}_2=47.3\pm 1.5\kern0.2778em \mathrm{ms} $$ for 2D MOLLI, 2DT 1 ρ $$ {T}_{1\rho } $$ -prep bSSFP and 2DT 2 $$ {T}_2 $$ -prep bSSFP, respectively. Promising results were obtained when comparing the proposed mapping to 2D references in 1 patient with active myocarditis.

The proposed approach enables simultaneous 3D whole-heart jointT 1 $$ {T}_1 $$ /T 1 ρ $$ {T}_{1\rho } $$ /T 2 $$ {T}_2 $$ mapping and water/fat imaging in≈ $$ \approx $$ 7 min scan time, demonstrating good agreement with conventional mapping techniques in phantoms and healthy subjects and promising results in 1 patient with suspected cardiovascular disease.

To propose and validate an unsupervised MRI reconstruction method that does not require fully sampled k-space data.

The proposed method, deep image prior with structured sparsity (DISCUS), extends the deep image prior (DIP) by introducing group sparsity to frame-specific code vectors, enabling the discovery of a low-dimensional manifold for capturing temporal variations. DISCUS was validated using four studies: (I) simulation of a dynamic Shepp-Logan phantom to demonstrate its manifold discovery capabilities, (II) comparison with compressed sensing and DIP-based methods using simulated single-shot late gadolinium enhancement (LGE) image series from six distinct digital cardiac phantoms in terms of normalized mean square error (NMSE) and structural similarity index measure (SSIM), (III) evaluation on retrospectively undersampled single-shot LGE data from eight patients, and (IV) evaluation on prospectively undersampled single-shot LGE data from eight patients, assessed via blind scoring from two expert readers.

DISCUS outperformed competing methods, demonstrating superior reconstruction quality in terms of NMSE and SSIM (Studies I-III) and expert reader scoring (Study IV).

An unsupervised image reconstruction method is presented and validated on simulated and measured data. These developments can benefit applications where acquiring fully sampled data is challenging.

The purpose of this study is to perform image registration and averaging of multiple free-breathing single-shot cardiac images, where the individual images may have a low signal-to-noise ratio (SNR).

To address low SNR encountered in single-shot imaging, especially at low field strengths, we propose a fast deep learning (DL)-based image registration method, called Averaging Morph with Edge Detection (AiM-ED). AiM-ED jointly registers multiple noisy source images to a noisy target image and utilizes a noise-robust pre-trained edge detector to define the training loss. We validate AiM-ED using synthetic late gadolinium enhanced (LGE) images from the MR extended cardiac-torso (MRXCAT) phantom and free-breathing single-shot LGE images from healthy subjects (24 slices) and patients (5 slices) under various levels of added noise. Additionally, we demonstrate the clinical feasibility of AiM-ED by applying it to data from patients (6 slices) scanned on a 0.55T scanner.

Compared with a traditional energy-minimization-based image registration method and DL-based VoxelMorph, images registered using AiM-ED exhibit higher values of recovery SNR and three perceptual image quality metrics. An ablation study shows the benefit of both jointly processing multiple source images and using an edge map in AiM-ED.

For single-shot LGE imaging, AiM-ED outperforms existing image registration methods in terms of image quality. With fast inference, minimal training data requirements, and robust performance at various noise levels, AiM-ED has the potential to benefit single-shot CMR applications.

The purpose of this study was to introduce and evaluate a novel 2D wideband black-blood (BB) LGE sequence, incorporating wideband inversion recovery, wideband T2 preparation, and non-rigid motion correction (MOCO) reconstruction, to improve myocardial scar detection and address artifacts associated with implantable cardioverter defibrillators (ICDs).

The wideband MOCO free-breathing BB-LGE sequence was tested on a sheep with ischemic scar and in 22 patients with cardiac disease, including 15 with cardiac implants, at 1.5T. Wideband MOCO free-breathing BB-LGE sequence was compared with conventional and wideband breath-held PSIR-LGE and conventional and wideband breath-held BB-LGE techniques. Image sharpness, entropy, and scar-to-blood, scar-to-myocardium, and blood-to-myocardium contrast were analyzed and reconstruction times were measured. Two expert readers assessed the image quality, ICD artifact severity, and the diagnostic confidence with scar extent. Finally, for the animal study, a histology of the heart was performed to confirm the presence and localization of scar tissue.

In the animal, wideband MOCO free-breathing BB-LGE were reconstructed in 0.6 s and demonstrated a 200 % improvement in scar-to-blood contrast compared to wideband breath-held PSIR-LGE, with significant improvement in image sharpness and reduction in entropy. It also effectively minimized ICD artifacts and accurately detected scars. In patients, wideband MOCO free-breathing BB-LGE were reconstructed in 1.5 ± 0.4 (standard deviation) s per slice. Seventeen patients (17/22; 77%) with myocardial scars were confidently diagnosed with wideband MOCO free-breathing BB-LGE, compared to 11 (11/22; 50 %) with wideband breath-held PSIR-LGE (P < 0.01).

Free-breathing wideband T2-prepared black-blood LGE imaging, combined with motion-corrected reconstruction, offers a promising diagnostic approach for the evaluation of myocardial lesions in patients with ICDs.

Papillary muscle infarction (PMI) has been linked to significantly increased mortality and is associated with ventricular arrhythmias and mitral regurgitation. Reference bright-blood late gadolinium enhancement (LGE) imaging provides poor scar-to-blood contrast, making PMI visualization challenging. Black-blood LGE imaging overcomes this limitation by improving the blood-scar contrast.

To evaluate a recent co-registered bright- (papillary muscle localization) and black-blood (PMI visualization) sequence (Scar-specific imaging with Preserved myOcardial visualizaTion: SPOT) to improve PMI visualization compared to a reference standard phase-sensitive inversion recovery (PSIR) sequence, and to enable automated PMI detection (auto-PMI).

Retrospective.

198 patients with ischemic heart disease were divided into an optimization dataset (N = 127) and a testing dataset (N = 71).

2D SPOT and PSIR balanced steady-state free precession sequences at 1.5 T.

Auto-PMI included: image acquisition, slice selection, endocardial segmentation, blood pool preprocessing, and PMI detection. Three radiologists (8, 5 and 2 years of MRI experience) assessed PMI in SPOT and PSIR images independently. A consensus reading regarding all assessments of both sequences was established. The number of patients with PMI in SPOT and PSIR acquisitions was compared. The diagnostic performances of visual (SPOT and PSIR) and auto-PMI (SPOT) detection were evaluated. Inter- and intra-observer reproducibility of the visual PMI detection was assessed.

McNemar test, p-value < 0.05 was considered statistically significant.

In the testing dataset, significantly more patients with PMI were detected using SPOT compared to PSIR in each session (37 vs. 27, 36 vs. 29, 41 vs. 31, 42 vs. 25). Sensitivity ranges for visual PMI detection were significantly higher using SPOT (89%-100% vs. 61%-82%). SPOT vs. PSIR inter- and intra-observer reproducibility ranges were 77%-80% vs. 71%-77%, and 97% vs. 88%, respectively. Auto-PMI sensitivity was 87%.

Co-registered bright- and black-blood SPOT imaging improved visual PMI detection and facilitated automated PMI assessment.

3. Technical Efficacy: Stage 2.

Hypertrophic cardiomyopathy sometimes complicates left ventricular (LV) outflow tract obstruction. Alcohol septal ablation (ASA) is indicated for drug-refractory hypertrophic obstructive cardiomyopathy (HOCM). Moreover, with an aging population, aortic valve stenosis (AS) is increasing, and surgical aortic valve replacement (SAVR) is indicated in these cases. Both AS and HOCM have stenosis at the exit of the LV and there is a difference in valvular and/or muscular stenosis. However, it is not clear how the release of stenosis affects blood flow. We investigate the influence of ASA and SAVR on blood flow using four-dimensional flow phase-contrast magnetic resonance imaging (4D flow MRI).

In this single-center retrospective observational study, we evaluated the blood flow of eight patients (five patients with HOCM and three patients with AS) before and after the intervention using 4D flow MRI.

The LV-aortic pressure gradient (PG) significantly improved from 79.4 ± 3.9 to 23.0 ± 2.0 mmHg (p < 0.001) by SAVR in the patients with AS. However, turbulent kinetic energy value (TKE) loss was not improved. However, the intra-LV PG in patients with HOCM improved from 79.0 ± 54.2 to 8.7 ± 4.0 mmHg (p < 0.05) by ASA. TKE loss improved from 7.0 ± 2.0 to 5.0 ± 0.1 mW (p < 0.05) and New York Heart Association functional class significantly improved from 2.2 ± 0.5 to 1.1 ± 0.3 (p < 0.001) by ASA.

The release of valvular or muscular stenosis has different effects on intra-LV blood flow. ASA reduced TKE loss and 4D flow MRI is useful to evaluate the efficacy of therapeutic interventions.

Late gadolinium enhancement (LGE) images, reconstructed using magnitude (MAG) or phase-sensitive inversion recovery (PSIR) sequences, differ in signal intensities because of their handling of longitudinal magnetization. These differences influence LGE quantification, which typically uses full-width at half maximum (FWHM) or standard deviation (n-SD) thresholding when predicting cardiac events.

This study assessed the impact of FWHM and n-SD on MAG- and PSIR-derived scar characteristics.

Patients with ischemic cardiomyopathy undergoing LGE imaging were retrospectively studied. Two reconstruction techniques (MAG vs PSIR) and 2 thresholding methods (FWHM vs n-SD) were evaluated. LGE images were postprocessed with commercially available software, using scar thresholds of 40%-60% of the maximum signal intensity for FWHM and 2-5 SDs above the mean for n-SD. Scar quantification was compared between patients with primary and secondary prevention implantable cardioverter-defibrillator.

Of the 80 patients, 32 (40%) had an implantable cardioverter-defibrillator for primary prevention. PSIR imaging showed significantly larger scar metrics than did MAG using FWHM and n-SD thresholding, including larger border zone (16.43 ± 8.15 g vs 21.42 ± 10.72 g; P<.001) and conduction corridor (CC) characteristics. MAG-based analysis revealed significant differences in scar and CC metrics. For PSIR, scar metrics were consistent across FWHM and n-SD. MAG-based analysis showed larger border zone and CC length in patients with primary prevention, with similar trends for PSIR.

This study demonstrates significant differences in myocardial scar metrics based on reconstruction and thresholding techniques. PSIR consistently provided robust scar characterization across methods, emphasizing its clinical potential to standardize LGE-cardiac magnetic resonance workflows and improve ventricular arrhythmia risk stratification.

to assess the relation of focal and diffuse left ventricular (LV) fibrosis to left bundle branch block (LBBB).

60 patients with dilated cardiomyopathy and LBBB (DCM-LBBB), 50 DCM-nonLBBB patients, 15 patients with LBBB and structurally normal heart (idiopathic LBBB) and 10 healthy volunteers (HV) underwent cardiovascular magnetic resonance (CMR) with late gadolinium enhancement (LGE). LGE LV images were post-proceeded for core scar (CS) and gray zone (GZ) calculation. Diffuse LV fibrosis was estimated on LGE-CMR images with the diffuse intensity ratio (DIR). Endomyocardial biopsy (EMB) was performed in 15(24.6 %) DCM-LBBB and 16 (32 %) non-LBBB DCM patients and allowed the quantification of collagen volume fraction (CVF).

The percentage of CVF correlated with the DIR value in the same segment (r = 0.66, p < 0.001). The value of CVF in EMB and frequency of LGE in both DCM groups was comparable (p = 0.8). In DCM-nonLBBB patients the percentage of CS was significantly higher (4.0[1.6; 11.7]% versus 1.4[0.1;8.5]% in DCM-LBBB patients, p = 0.047), whereas percentage of GZ and total fibrosis in both DCM groups was comparable. DIR value was higher in patients with idiopathic LBBB than in HV (0.54±0.09 versus 0.34±0.1, р<0,001).

Neither focal nor interstitial fibrosis is associated with LBBB in DCM patients. Diffuse inflammation in DCM-LBBB patients may contribute to the progression of systolic dysfunction but is not a cause of LBBB. The increased value of interstitial fibrosis in patients with idiopathic LBBB may reflect latent diffuse process in myocardium inexorably leading to DCM development.

Myocardial T1-rho (T1ρ) mapping is a promising method for identifying and quantifying myocardial injuries without contrast agents, but its clinical use is hindered by the lack of dedicated analysis tools.

To explore the feasibility of clinically integrated artificial intelligence-driven analysis for efficient and automated myocardial T1ρ mapping.

Retrospective.

Five hundred seventy-three patients divided into a training (N = 500) and a test set (N = 73) including ischemic and nonischemic cases.

Single-shot bSSFP T1ρ mapping sequence at 1.5 T.

The automated process included: left ventricular (LV) wall segmentation, right ventricular insertion point detection and creation of a 16-segment model for segmental T1ρ value analysis. Two radiologists (20 and 7 years of MRI experience) provided ground truth annotations. Interobserver variability and segmentation quality were assessed using the Dice coefficient with manual segmentation as reference standard. Global and segmental T1ρ values were compared. Processing times were measured.

Intraclass correlation coefficients (ICCs) and Bland-Altman analysis (bias ±2SD); Paired Student's t-tests and one-way ANOVA. A P value <0.05 was considered significant.

The automated approach significantly reduced processing time (3 seconds vs. 1 minute 51 seconds ± 22 seconds). In the test set, automated LV wall segmentation closely matched manual results (Dice 81.9% ± 9.0) and closely aligned with interobserver segmentation (Dice 82.2% ± 6.5). Excellent ICCs were achieved on a patient basis (0.94 [95% CI: 0.91 to 0.96]) with bias of -0.93 cm2 ± 6.60. There was no significant difference in global T1ρ values between manual (54.9 msec ± 4.6; 95% CI: 53.8 to 56.0 msec, range: 46.6-70.9 msec) and automated processing (55.4 msec ± 5.1; 95% CI: 54.2 to 56.6 msec; range: 46.4-75.1 msec; P = 0.099). The pipeline demonstrated a high level of agreement with manual-derived T1ρ values at the patient level (ICC = 0.85; bias +0.52 msec ± 5.18). No significant differences in myocardial T1ρ values were found between methods across the 16 segments (P = 0.75).

Automated myocardial T1ρ mapping shows promise for the rapid and noninvasive assessment of heart disease.

3 TECHNICAL EFFICACY: Stage 1.

This study aimed to evaluate the influence of reader training and experience on the detection of (small) myocardial infarctions (MIs) and the assessment of ischemic scar transmurality using dark-blood late gadolinium enhancement (LGE) and bright-blood LGE magnetic resonance imaging. It was hypothesized that dark-blood LGE simplifies the detection of (small) MIs for less experienced readers, compared with bright-blood LGE imaging.

One hundred patients referred for cardiac magnetic resonance imaging for suspected ischemic scar were retrospectively included. Dark-blood LGE was performed first, followed by bright-blood LGE. Nine clinicians, grouped into three levels based on their training and experience, assessed the LGE images for the presence of MI and ischemic scar transmurality. Their assessment was subsequently compared with a European Association of Cardiovascular Imaging level 3 consultant. Reader confidence was evaluated with a 4-point Likert scale. Multilevel logistic regression was used to compare the 2 LGE methods and assess differences in myocardial infarction detection and transmurality among the 3 experience levels. Wilcoxon signed rank tests were performed to compare the reader confidence between the 2 LGE methods, whereas Friedman omnibus tests were conducted to assess differences in reader confidence among the 3 experience levels.

Dark-blood LGE resulted in increased correct detection of MIs compared with bright-blood LGE for both level 1 (87.3% vs 82.7%, odds ratio [OR]: 1.55 [95% confidence interval (CI): 0.94-2.54], P = 0.083) and level 2 readers (89.7% vs 83.0%, OR: 2.05 [95% CI: 1.20-3.51], P = 0.009). There was no significant difference observed between dark-blood LGE and bright-blood LGE for level 3 readers (88.7% vs 87.0%, OR: 1.20 [95% CI: 0.70-2.06], P = 0.495). Level 2 readers significantly detected more small MIs correctly when using dark-blood LGE compared with bright-blood LGE (66.7% vs 50.8%, OR: 2.40 [95% CI: 1.03-5.60], P = 0.042). All experience levels showed significantly increased confidence in presence of ischemic scar and transmurality with dark-blood LGE.

Readily available dark-blood LGE can assist less experienced readers in correctly detecting and assessing (small) MIs compared with conventional bright-blood LGE. Regardless of experience level, dark-blood LGE improves reader confidence in assessing the presence and transmurality of MIs.

Cardiac MRI (CMR) is an important imaging modality in the evaluation of cardiovascular diseases. CMR image acquisition is technically challenging, which in some circumstances is associated with artifacts, both general as well as sequence specific. Recognizing imaging artifacts, understanding their causes, and applying effective approaches for artifact mitigation are critical for successful CMR. Balanced steady-state free precession (bSSFP), the most common CMR sequence, is associated with band and flow artifacts, which are amplified at 3-T imaging. This can be mitigated by targeted shimming, by short repetition time, or by using a frequency-scout sequence. In patients with cardiac arrhythmias or poor breath hold, the quality of cine imaging can be improved with a non-electrocardiographically gated free-breathing real-time sequence. Motion artifacts on late gadolinium enhancement (LGE) images can be mitigated by using single-shot technique with motion compensation and signal averaging. LGE images are also prone to partial-volume averaging and incomplete myocardial nulling. In phase-contrast imaging, aliasing artifact is seen when the velocity of blood is higher than the encoded velocity. Aliasing can be mitigated by increasing the encoded velocity or using postprocessing software. In first-pass perfusion imaging, a dark rim artifact due to Gibbs ringing can be distinguished from a true perfusion defect based on earlier appearance and fading after a few cardiac cycles. With implanted cardiac devices, artifactual high signal intensity mimicking scar is seen on LGE images, which can be mitigated using a wide-band sequence. With devices and metallic artifacts, traditional gradient-recalled echo sequence has fewer artifacts than bSSFP. CMR at 3 T requires adaptation of sequences to minimize artifacts. ©RSNA, 2025 Supplemental material is available for this article.

Polarity-corrected inversion time preparation (PCTIP), a myocardial T1 mapping technique, is expected to reduce measurement underestimation in the modified Look-Locker inversion recover method. However, measurement precision is reduced, especially for heart rate variability. We devised an analysis using a recurrence formula to overcome this problem and showed that it improved the measurement accuracy, especially at high heart rates. Therefore, this study aimed to determine the effect of this analysis on the accuracy and precision of T1 measurements for irregular heart rate variability.

A PCTIP scan using a 3T MRI scanner was performed in phantom experiment. We generated the simulated R-waves required for electrocardiogram (ECG)-gated acquisition using a signal generator set to 30 combinations. T1 map was generated using the signal train of the PCTIP images by nonlinear curve fitting using conventional and recurrence formulas. Accuracy against reference T1 and precision of heart rate variability were evaluated. To evaluate the fitting accuracy of both analyses, the relative fitting error was calculated.

For the longer T1, the fitting error was larger than the short T1, with the conventional analysis showing 10.1±2.0%. The recurrence formula analysis showed a small fitting error less than 1%, which was consistent for all heart rate variability patterns. In the conventional analysis, the accuracy, especially for longer T1, showed a large underestimation of the measurements and poor linearity. However, in the recurrence formula analysis, the accuracy improved at a long T1, and linearity also improved. The Bland-Altman plot showed that it varied greatly depending on the heart rate variability pattern for the longer T1 in the conventional analysis, whereas the recurrence formula analysis suppressed this variation.

T1 analysis of PCTIP using the recurrence formula analysis achieved accurate and precise T1 measurements, even for irregular heart rate variability.

Cardiovascular magnetic resonance imaging is a powerful diagnostic tool for assessing cardiac structure and function. Traditional breath-held imaging protocols, however, pose challenges for patients with arrhythmias or limited breath-holding capacity. We introduce Motion-Guided Deep Image prior (M-DIP), a novel unsupervised reconstruction framework for accelerated real-time cardiac MRI. M-DIP employs a spatial dictionary to synthesize a time-dependent template image, which is further refined using time-dependent deformation fields that model cardiac and respiratory motion. Unlike prior DIP-based methods, M-DIP simultaneously captures physiological motion and frame-to-frame content variations, making it applicable to a wide range of dynamic applications. We validate M-DIP using simulated MRXCAT cine phantom data as well as free-breathing real-time cine and single-shot late gadolinium enhancement data from clinical patients. Comparative analyses against state-of-the-art supervised and unsupervised approaches demonstrate M-DIP's performance and versatility. M-DIP achieved better image quality metrics on phantom data, as well as higher reader scores for in-vivo patient data.

Evaluate the feasibility of quantification of Relaxation Along a Fictitious Field in the 2nd rotating frame (RAFF2) relaxation times in the human myocardium at 3 T.

T RAFF 2 mapping was performed using a breath-held ECG-gated acquisition of five images: one without preparation, three preceded by RAFF2 trains of varying duration, and one preceded by a saturation prepulse. Pixel-wiseT RAFF 2 maps were obtained after three-parameter exponential fitting. The repeatability ofT RAFF 2 ,T 1 , andT 2 was assessed in phantom via the coefficient of variation (CV) across three repetitions. In seven healthy subjects,T RAFF 2 was tested for precision, reproducibility, inter-subject variability, and image quality (IQ) on a Likert scale (1 = Nondiagnostic, 5 = Excellent). Additionally,T RAFF 2 mapping was performed in three patients with suspected cardiovascular disease, comparing it to late gadolinium enhancement (LGE), nativeT 1 ,T 2 , and ECV mapping.

In phantom,T RAFF 2 showed good repeatability (CV < 1.5%) while showing no ( R 2 = 0.09 ) and high ( R 2 = 0.99 ) correlation withT 1 andT 2 , respectively. MyocardialT RAFF 2 maps exhibited overall acceptable image quality (IQ = 3.0 ± 1.0) with moderate artifact levels, stemming from off-resonances near the coronary sinus. AverageT RAFF 2 time across subjects and repetitions was 79.1 ± 7.3 ms. Good precision (7.6 ± 1.4%), reproducibility (1.0 ± 0.6%), and low inter-subject variability (10.0 ± 1.8%) were obtained. In patients, visual agreement of the infarcted area was observed in theT RAFF 2 map and LGE.

MyocardialT RAFF 2 quantification at 3 T was successfully achieved in a single breath-hold with acceptable image quality, albeit with residual off-resonance artifacts. Nonetheless, preliminary clinical data indicate potential sensitivity ofT RAFF 2 mapping to myocardial infarction detection without the need for contrast agents, but off-resonance artifacts mitigation warrants further investigation.

Cardiovascular magnetic resonance imaging (CMR) has received extensive validation for the assessment of valvular heart disease (VHD) and offers an accurate and direct method for the quantification of aortic regurgitation (AR). According to the current guidelines, CMR represents a useful second-line investigation in patients with poor acoustic windows or when echocardiography is inconclusive, for example, in cases of multiple or eccentric aortic jets. Without ionizing radiation exposure, CMR provides in-depth information not only on the severity degree of AR, providing a precise quantification of regurgitant volume and fraction, but also on cardiac structure and function, being recognized as the gold standard for the assessment of heart chamber size and systolic function. CMR allows a free choice of cardiac imaging planes and provides further information on the myocardium, thanks to the tissue characterization ability offered by several sequences, such as the late gadolinium enhancement technique. The possibilities offered by CMR become even more interesting in the context of mixed and multiple VHD, where the echocardiographic assessments often encounter difficulties in the quantification of each single valve lesion. The current scientific data support a greater expansion of CMR in this field, thanks to its additional advantages for the diagnosis, risk stratification, and to guide treatment. This review investigates the current CMR techniques and protocols in AR, with special insights into the evaluation of mixed aortic valve disease and multiple VHD including AR.

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.

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.

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).

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.

Although quantitative myocardial T1 and T2 mappings are clinically used to evaluate myocardial diseases, their application needs a minimum of six breath-holds to cover three short-axis slices. The purpose of this work is to simultaneously quantify multislice myocardial T1 and T2 across three short-axis slices in one breath-hold by combining simultaneous multislice (SMS) with multimapping.

An SMS-Multimapping sequence with multiband radiofrequency (RF) excitations and Cartesian fast low-angle shot readouts was developed for data acquisition. When 3 slices are simultaneously acquired, the acceleration rate is around 12-fold, causing a highly ill-conditioned reconstruction problem. To mitigate image artifacts and noise caused by the ill-conditioning, a reconstruction algorithm based on locally low-rank and sparsity (LLRS) constraints was developed. Validation was performed in phantoms and in vivo imaging, with 20 healthy subjects and 4 patients, regarding regional mean, precision, and scan-rescan reproducibility.

The phantom imaging shows that SMS-Multimapping with locally low-rank (LLRS) accurately reconstructed multislice T1 and T2 maps despite a six-fold acceleration of scan time. Healthy subject imaging shows that the proposed LLRS algorithm substantially improved image quality relative to split slice-generalized autocalibrating partially parallel acquisition. Compared with modified look-locker inversion recovery (MOLLI), SMS-Multimapping exhibited higher T1 mean (1118 ± 43 ms vs 1190 ± 49 ms, P < 0.01), lower precision (67 ± 17 ms vs 90 ± 17 ms, P < 0.01), and acceptable scan-rescan reproducibility measured by 2 scans 10-min apart (bias = 1.4 ms for MOLLI and 9.0 ms for SMS-Multimapping). Compared with balanced steady-state free precession (bSSFP) T2 mapping, SMS-Multimapping exhibited similar T2 mean (43.5 ± 3.3 ms vs 43.0 ± 3.5 ms, P = 0.64), similar precision (4.9 ± 2.1 ms vs 5.1 ± 1.0 ms, P = 0.93), and acceptable scan-rescan reproducibility (bias = 0.13 ms for bSSFP T2 mapping and 0.55 ms for SMS-Multimapping). In patients, SMS-Multimapping clearly showed the abnormality in a similar fashion as the reference methods despite using only one breath-hold.

SMS-Multimapping with the proposed LLRS reconstruction can measure multislice T1 and T2 maps in one breath-hold with good accuracy, reasonable precision, and acceptable reproducibility, achieving a six-fold reduction of scan time and an improvement of patient comfort.

Cardiac balanced steady state free precession (bSSFP) cine imaging suffers from banding and flow artifacts induced by off-resonance. The work aimed to develop a twofold phase cycling sequence with a neural network-based reconstruction (2P-SSFP+Network) for a joint suppression of banding and flow artifacts in cardiac cine imaging.

A dual-encoder neural network was trained on 1620 pairs of phase-cycled left ventricular (LV) cine images collected from 18 healthy subjects. Twenty healthy subjects and 25 patients were prospectively scanned using the proposed 2P-SSFP sequence. bSSFP cine of a single RF phase increment (1P-SSFP), bSSFP cine of a single radiofrequency (RF) phase increment with a network-based artifact reduction (1P-SSFP+Network), the averaging of the two phase-cycled images (2P-SSFP+Average), and the proposed method were mutually compared, in terms of artifact suppression performance in the LV, generalizability over altered scan parameters and scanners, suppression of large-area banding artifacts in the left atrium (LA), and accuracy of downstream segmentation tasks.

In the healthy subjects, 2P-SSFP+Network showed robust suppressions of artifacts across a range of phase combinations. Compared with 1P-SSFP and 2P-SSFP+Average, 2P-SSFP+Network improved banding artifacts (3.85 ± 0.67 and 4.50 ± 0.45 vs 5.00 ± 0.00, P < 0.01 and P = 0.02, respectively), flow artifacts (3.35 ± 0.78 and 2.10 ± 0.77 vs 4.90 ± 0.20, both P < 0.01), and overall image quality (3.25 ± 0.51 and 2.30 ± 0.60 vs 4.75 ± 0.25, both P < 0.01). 1P-SSFP+Network and 2P-SSFP+Network achieved a similar artifact suppression performance, yet the latter had fewer hallucinations (two-chamber, 4.25 ± 0.51 vs 4.85 ± 0.45, P = 0.04; four-chamber, 3.45 ± 1.21 vs 4.65 ± 0.50, P = 0.03; and left atrium (LA), 3.35 ± 1.00 vs 4.65 ± 0.45, P < 0.01). Furthermore, in the pulmonary veins and LA, 1P-SSFP+Network could not eliminate banding artifacts since they occupied a large area, whereas 2P-SSFP+Network reliably suppressed the artifacts. In the downstream automated myocardial segmentation task, 2P-SSFP+Network achieved more accurate segmentations than 1P-SSFP with different phase increments.

2P-SSFP+Network jointly suppresses banding and flow artifacts while manifesting a good generalizability against variations of anatomy and scan parameters. It provides a feasible solution for robust suppression of the two types of artifacts in bSSFP cine imaging.

Cardiovascular magnetic resonance imaging (MRI) in patients with cardiac implants, such as pacemakers and defibrillators, has gained importance in recent years with the development of modern cardiac implantable electronic devices. The increasing clinical need to perform MRI examinations in patients with cardiac implants has driven the development of new advanced MRI sequences to mitigate image artifacts associated with cardiac implants. More specifically, advances in imaging techniques, such as wideband late gadolinium enhancement imaging, wideband T1 mapping, and wideband perfusion, have been designed to improve image quality and examinations in patients with cardiac implants, enabling a comprehensive and more reliable diagnosis, which was previously unattainable in these patients. This review article explores recent developments and applications of wideband techniques in the field of cardiovascular MRI, offering insights into their transformative potential. Clinical applications of wideband cardiovascular MRI are highlighted, particularly in assessing myocardial viability, guiding ventricular tachycardia ablation, and characterizing myocardial tissue.

The diagnosis of myocarditis by cardiovascular magnetic resonance (CMR) requires the use of T2 and T1 weighted imaging, ideally incorporating parametric mapping. Current two-dimensional (2D) mapping sequences are acquired sequentially and involve multiple breath-holds resulting in prolonged scan times and anisotropic image resolution. We developed an isotropic free-breathing three-dimensional (3D) whole-heart sequence that allows simultaneous T1 and T2 mapping and validated it in patients with suspected myocarditis.

Eighteen healthy volunteers and 28 patients with suspected myocarditis underwent conventional 2D T1 and T2 mapping with whole-heart coverage and 3D joint T1/T2 mapping on a 1.5T scanner. Acquisition time, image quality, and diagnostic performance were compared. Qualitative analysis was performed using a 4-point Likert scale. Bland-Altman plots were used to assess the quantitative agreement between 2D and 3D sequences.

The 3D T1/T2 sequence was acquired in 8 min 26 s under free breathing, whereas 2D T1 and T2 sequences were acquired with breath-holds in 11 min 44 s (p = 0.0001). All 2D images were diagnostic. For 3D images, 89% (25/28) of T1 and 96% (27/28) of T2 images were diagnostic with no significant difference in the proportion of diagnostic images for the 3D and 2D T1 (p = 0.2482) and T2 maps (p = 1.0000). Systematic bias in T1 was noted with biases of 102, 115, and 152 ms for basal-apical segments, with a larger bias for higher T1 values. Good agreement between T2 values for 3D and 2D techniques was found (bias of 1.8, 3.9, and 3.6 ms for basal-apical segments). The sensitivity and specificity of the 3D sequence for diagnosing acute myocarditis were 74% (95% confidence interval [CI] 49%-91%) and 83% (36%-100%), respectively, with a c-statistic (95% CI) of 0.85 (0.79-0.91) and no statistically significant difference between the 2D and 3D sequences for the detection of acute myocarditis for T1 (p = 0.2207) or T2 (p = 1.0000).

Free-breathing whole-heart 3D joint T1/T2 mapping was comparable to 2D mapping sequences with respect to diagnostic performance, but with the added advantages of free breathing and shorter scan times. Further work is required to address the bias noted at high T1 values, but this did not significantly impact diagnostic accuracy.

To develop and share a deep learning method that can accurately identify optimal inversion time (TI) from multi-vendor, multi-institutional and multi-field strength inversion scout (TI scout) sequences for late gadolinium enhancement cardiac MRI.

Retrospective multicentre study conducted on 1136 1.5-T and 3-T cardiac MRI examinations from four centres and three scanner vendors. Deep learning models, comprising a convolutional neural network (CNN) that provides input to a long short-term memory (LSTM) network, were trained on TI scout pixel data from centres 1 to 3 to identify optimal TI, using ground truth annotations by two readers. Accuracy within 50 ms, mean absolute error (MAE), Lin's concordance coefficient (LCCC) and reduced major axis regression (RMAR) were used to select the best model from validation results, and applied to holdout test data. Robustness of the best-performing model was also tested on imaging data from centre 4.

The best model (SE-ResNet18-LSTM) produced accuracy of 96.1%, MAE 22.9 ms and LCCC 0.47 compared to ground truth on the holdout test set and accuracy of 97.3%, MAE 15.2 ms and LCCC 0.64 when tested on unseen external (centre 4) data. Differences in vendor performance were observed, with greatest accuracy for the most commonly represented vendor in the training data.

A deep learning model was developed that can identify optimal inversion time from TI scout images on multi-vendor data with high accuracy, including on previously unseen external data. We make this model available to the scientific community for further assessment or development.

A robust automated inversion time selection tool for late gadolinium-enhanced imaging allows for reproducible and efficient cross-vendor inversion time selection.

• A model comprising convolutional and recurrent neural networks was developed to extract optimal TI from TI scout images. • Model accuracy within 50 ms of ground truth on multi-vendor holdout and external data of 96.1% and 97.3% respectively was achieved. • This model could improve workflow efficiency and standardise optimal TI selection for consistent LGE imaging.

Current echocardiographic risk factors for prognosis in cardiac amyloidosis (CA) do not distinguish between the two main subtypes: transthyretin cardiomyopathy (TTR) and immunoglobulin light chain cardiomyopathy (AL), each of which require distinct diagnostic and therapeutic approaches. Additionally, only traditional parameters have been studied with little data on advanced techniques. Accordingly, we sought to determine whether differences exist in 2D transthoracic echocardiography (2DE) predictors of survival between the CA subtypes using a comprehensive approach.

220 patients (72±12 years) with confirmed CA (AL=89, TTR=131) who underwent 2DE at the time of CA diagnosis were enrolled. Left ventricular (LV) dimensions, indexed mass (LVMi), global longitudinal strain (LVGLS), apical-sparing ratio (LVASR), diastology, right ventricular (RV) size and function indices including tricuspid annular systolic excursion (TAPSE), RV free-wall (RVFWS) and global (RVGLS) strain, indexed left (LA) and right atrial volumes (LAVi and RAVi), LA strain (reservoir and booster) and RV systolic pressure (RVSP) were measured. A propensity-score weighted stepwise variable selection Cox proportional hazards model derived from NYHA class and renal impairment status at diagnosis was used to determine the associations between 2DE parameters and mortality specific to CA subtype over a median follow-up of 36-months.

After adjusting for age, atrial fibrillation and treatment, parameters associated with survival were RVFWS (p=0.003, HR 1.15, 95% CI[1.053,1.245]) and RVSP (p=0.03, HR 1.03, 95% CI[1.004,1.063]) in AL and LVASR (p=0.007, HR 6.68, 95% CI[1.75,25.492]) and RAVi (p=0.049, HR 1.03, 95% CI[1.000,1.052]) in TTR.

Echocardiographic prognosticators for survival are specific to cardiac amyloid subtype. These results potentially provide information critical for clinical decision-making and follow-up in these patients.

Left ventricular (LV) hypertrophy occurs in both aortic stenosis (AS) and systemic hypertension (HTN) in response to wall stress. However, differentiation of hypertrophy due to these 2 etiologies is lacking. The aim was to study the 3-dimensional geometric remodeling pattern in severe AS pre- and postsurgical aortic valve replacement and to compare with HTN and healthy controls.

Ninety-one subjects (36 severe AS, 19 HTN, and 36 healthy controls) underwent cine cardiac magnetic resonance. Cardiac magnetic resonance was repeated 8 months post-aortic valve replacement (n=18). Principal component analysis was performed on the 3-dimensional meshes reconstructed from 109 cardiac magnetic resonance scans of 91 subjects at end-diastole. Principal component analysis modes were compared across experimental groups together with conventional metrics of shape, strain, and scar.

A unique AS signature was identified by wall thickness linked to a LV left-right axis shift and a decrease in short-axis eccentricity. HTN was uniquely linked to increased septal thickness. Combining these 3 features had good discriminative ability between AS and HTN (area under the curve, 0.792). The LV left-right axis shift was not reversible post-aortic valve replacement, did not associate with strain, age, or sex, and was predictive of postoperative LV mass regression (R2=0.339, P=0.014).

Unique remodeling signatures might differentiate the etiology of LV hypertrophy. Preliminary findings suggest that LV axis shift is characteristic in AS, is not reversible post-aortic valve replacement, predicts mass regression, and may be interpreted to be an adaptive mechanism.

Mitral valve prolapse is a common valve disorder that usually has a benign prognosis unless there is significant regurgitation or LV impairment. However, a subset of patients are at an increased risk of ventricular arrhythmias and sudden cardiac death, which has led to the recognition of "arrhythmic mitral valve prolapse" as a clinical entity. Emerging risk factors include mitral annular disjunction and myocardial fibrosis. While echocardiography remains the primary method of evaluation, cardiac magnetic resonance has become crucial in managing this condition. Cine magnetic resonance sequences provide accurate characterization of prolapse and annular disjunction, assessment of ventricular volumes and function, identification of early dysfunction and remodeling, and quantitative assessment of mitral regurgitation when integrated with flow imaging. However, the unique strength of magnetic resonance lies in its ability to identify tissue changes. T1 mapping sequences identify diffuse fibrosis, in turn related to early ventricular dysfunction and remodeling. Late gadolinium enhancement sequences detect replacement fibrosis, an independent risk factor for ventricular arrhythmias and sudden cardiac death. There are consensus documents and reviews on the use of cardiac magnetic resonance specifically in arrhythmic mitral valve prolapse. However, in this article, we propose an algorithm for the broader use of cardiac magnetic resonance in managing this condition in various scenarios. Future advancements may involve implementing techniques for tissue characterization and flow analysis, such as 4D flow imaging, to identify patients with ventricular dysfunction and remodeling, increased arrhythmic risk, and more accurate grading of mitral regurgitation, ultimately benefiting patient selection for surgical therapy.

This study investigated the impact of ComBat harmonization on the reproducibility of radiomic features extracted from magnetic resonance images (MRI) acquired on different scanners, using various data acquisition parameters and multiple image pre-processing techniques using a dedicated MRI phantom. Four scanners were used to acquire an MRI of a nonanatomic phantom as part of the TCIA RIDER database. In fast spin-echo inversion recovery (IR) sequences, several inversion durations were employed, including 50, 100, 250, 500, 750, 1000, 1500, 2000, 2500, and 3000 ms. In addition, a 3D fast spoiled gradient recalled echo (FSPGR) sequence was used to investigate several flip angles (FA): 2, 5, 10, 15, 20, 25, and 30 degrees. Nineteen phantom compartments were manually segmented. Different approaches were used to pre-process each image: Bin discretization, Wavelet filter, Laplacian of Gaussian, logarithm, square, square root, and gradient. Overall, 92 first-, second-, and higher-order statistical radiomic features were extracted. ComBat harmonization was also applied to the extracted radiomic features. Finally, the Intraclass Correlation Coefficient (ICC) and Kruskal-Wallis's (KW) tests were implemented to assess the robustness of radiomic features. The number of non-significant features in the KW test ranged between 0-5 and 29-74 for various scanners, 31-91 and 37-92 for three times tests, 0-33 to 34-90 for FAs, and 3-68 to 65-89 for IRs before and after ComBat harmonization, with different image pre-processing techniques, respectively. The number of features with ICC over 90% ranged between 0-8 and 6-60 for various scanners, 11-75 and 17-80 for three times tests, 3-83 to 9-84 for FAs, and 3-49 to 3-63 for IRs before and after ComBat harmonization, with different image pre-processing techniques, respectively. The use of various scanners, IRs, and FAs has a great impact on radiomic features. However, the majority of scanner-robust features is also robust to IR and FA. Among the effective parameters in MR images, several tests in one scanner have a negligible impact on radiomic features. Different scanners and acquisition parameters using various image pre-processing might affect radiomic features to a large extent. ComBat harmonization might significantly impact the reproducibility of MRI radiomic features.

Artificial intelligence (AI) integration in cardiac magnetic resonance imaging presents new and exciting avenues for advancing patient care, automating post-processing tasks, and enhancing diagnostic precision and outcomes. The use of AI significantly streamlines the examination workflow through the reduction of acquisition and postprocessing durations, coupled with the automation of scan planning and acquisition parameters selection. This has led to a notable improvement in examination workflow efficiency, a reduction in operator variability, and an enhancement in overall image quality. Importantly, AI unlocks new possibilities to achieve spatial resolutions that were previously unattainable in patients. Furthermore, the potential for low-dose and contrast-agent-free imaging represents a stride toward safer and more patient-friendly diagnostic procedures. Beyond these benefits, AI facilitates precise risk stratification and prognosis evaluation by adeptly analysing extensive datasets. This comprehensive review article explores recent applications of AI in the realm of cardiac magnetic resonance imaging, offering insights into its transformative potential in the field.

Joint bright- and black-blood MRI techniques provide improved scar localization and contrast. Black-blood contrast is obtained after the visual selection of an optimal inversion time (TI) which often results in uncertainties, inter- and intra-observer variability and increased workload. In this work, we propose an artificial intelligence-based algorithm to enable fully automated TI selection and simplify myocardial scar imaging.

The proposed algorithm first localizes the left ventricle using a U-Net architecture. The localized left cavity centroid is extracted and a squared region of interest ("focus box") is created around the resulting pixel. The focus box is then propagated on each image and the sum of the pixel intensity inside is computed. The smallest sum corresponds to the image with the lowest intensity signal within the blood pool and healthy myocardium, which will provide an ideal scar-to-blood contrast. The image's corresponding TI is considered optimal. The U-Net was trained to segment the epicardium in 177 patients with binary cross-entropy loss. The algorithm was validated retrospectively in 152 patients, and the agreement between the algorithm and two magnetic resonance (MR) operators' prediction of TI values was calculated using the Fleiss' kappa coefficient. Thirty focus box sizes, ranging from 2.3mm2 to 20.3cm2, were tested. Processing times were measured.

The U-Net's Dice score was 93.0 ± 0.1%. The proposed algorithm extracted TI values in 2.7 ± 0.1 s per patient (vs. 16.0 ± 8.5 s for the operator). An agreement between the algorithm's prediction and the MR operators' prediction was found in 137/152 patients (κ= 0.89), for an optimal focus box of size 2.3cm2.

The proposed fully-automated algorithm has potential of reducing uncertainties, variability, and workload inherent to manual approaches with promise for future clinical implementation for joint bright- and black-blood MRI.

Modern MRI scanners utilize one or more arrays of small receive-only coils to collect k-space data. The sensitivity maps of the coils, when estimated using traditional methods, differ from the true sensitivity maps, which are generally unknown. Consequently, the reconstructed MR images exhibit undesired spatial variation in intensity. These intensity variations can be at least partially corrected using pre-scan data. In this work, we propose an intensity correction method that utilizes pre-scan data. For demonstration, we apply our method to a digital phantom, as well as to cardiac MRI data collected from a commercial scanner by Siemens Healthineers. The code is available at https://github.com/OSU-MR/SCC.

High-quality training data are not always available in dynamic MRI. To address this, we propose a self-supervised deep learning method called deep image prior with structured sparsity (DISCUS) for reconstructing dynamic images. DISCUS is inspired by deep image prior (DIP) and recovers a series of images through joint optimization of network parameters and input code vectors. However, DISCUS additionally encourages group sparsity on frame-specific code vectors to discover the low-dimensional manifold that describes temporal variations across frames. Compared to prior work on manifold learning, DISCUS does not require specifying the manifold dimensionality. We validate DISCUS using three numerical studies. In the first study, we simulate a dynamic Shepp-Logan phantom with frames undergoing random rotations, translations, or both, and demonstrate that DISCUS can discover the dimensionality of the underlying manifold. In the second study, we use data from a realistic late gadolinium enhancement (LGE) phantom to compare DISCUS with compressed sensing (CS) and DIP, and to demonstrate the positive impact of group sparsity. In the third study, we use retrospectively undersampled single-shot LGE data from five patients to compare DISCUS with CS reconstructions. The results from these studies demonstrate that DISCUS outperforms CS and DIP, and that enforcing group sparsity on the code vectors helps discover true manifold dimensionality and provides additional performance gain.

The acquisition of images minutes or even hours after intravenous extracellular gadolinium-based contrast agents (GBCA) administration ("Late/Delayed Gadolinium Enhancement" imaging; in this review, further termed LGE) has gained significant prominence in recent years in magnetic resonance imaging. The major limitation of LGE is the long examination time; thus, it becomes necessary to understand when it is worth waiting time after the intravenous injection of GBCA and which additional information comes from LGE. LGE can potentially be applied to various anatomical sites, such as heart, arterial vessels, lung, brain, abdomen, breast, and the musculoskeletal system, with different pathophysiological mechanisms. One of the most popular clinical applications of LGE regards the assessment of myocardial tissue thanks to its ability to highlight areas of acute myocardial damage and fibrotic tissues. Other frequently applied clinical contexts involve the study of the urinary tract with magnetic resonance urography and identifying pathological abdominal processes characterized by high fibrous stroma, such as biliary tract tumors, autoimmune pancreatitis, or intestinal fibrosis in Crohn's disease. One of the current areas of heightened research interest revolves around the possibility of non-invasively studying the dynamics of neurofluids in the brain (the glymphatic system), the disruption of which could underlie many neurological disorders.

Visualizing left atrial anatomy including the pulmonary veins (PVs) is important for planning the procedure of pulmonary vein isolation with ablation in patients with atrial fibrillation (AF). The aims of our study are to investigate the feasibility of the 3D whole-heart bright-blood and black-blood phase-sensitive (BOOST) inversion recovery sequence in patients with AF scheduled for ablation or electro-cardioversion, and to analyze the correlation between image quality and heart rate and rhythm of patients.

BOOST was performed for assessing PVs both with T2 preparation pre-pulse (T2prep) and magnetization transfer preparation (MTC) in 45 patients with paroxysmal or permanent AF scheduled for ablation or electro-cardioversion. Image quality analyses were performed by two independent observers. Qualitative assessment was made using the Likert scale; for quantitative analysis, signal to noise ratios (SNR) and contrast to noise ratios (CNR) were calculated for each PV. Heart rate and rhythm were analyzed based on standard 12-lead ECGs.

All MTC-BOOST acquisitions achieved diagnostic quality in the PVs, while a significant proportion of T2prep-BOOST images were not suitable for assessing PVs. SNR and CNR values of the MTC-BOOST bright-blood images were higher if patients had sinus rhythm. We found a significant or nearly significant negative correlation between heart rate and the SNR and CNR values of MTC-BOOST bright-blood images.

3D whole-heart MTC-BOOST bright-blood imaging is suitable for visualizing the PVs in patients with AF, producing diagnostic image quality in 100% of cases. However, image quality was influenced by heart rate and rhythm.

The novel 3D whole-heart BOOST CMR sequence needs no contrast administration and is performed during free-breathing; therefore, it is easy to use for a wide range of patients and is suitable for visualizing the PVs in patients with AF.

• The applicability of the novel 3D whole-heart bright-blood and black-blood phase-sensitive sequence to pulmonary vein imaging in clinical practice is unknown. • Magnetization transfer-bright-blood and black-blood phase-sensitive imaging is suitable for visualizing the pulmonary veins in patients with atrial fibrillation with excellent or good image quality. • Bright-blood and black-blood phase-sensitive cardiac magnetic resonance sequence is easy to use for a wide range of patients as it needs no contrast administration and is performed during free-breathing.

To identify clinical correlates of myocardial T1ρ and to examine how myocardial T1ρ values change under various clinical scenarios.

A total of 66 patients (26% female, median age 57 years [Q1-Q3, 44-65 years]) with known structural heart disease and 44 controls (50% female, median age 47 years [28-57 years]) underwent cardiac magnetic resonance imaging at 1.5 T, including T1ρ mapping, T2 mapping, native T1 mapping, late gadolinium enhancement, and extracellular volume (ECV) imaging. In controls, T1ρ positively related with T2 (P = 0.038) and increased from basal to apical levels (P < 0.001). As compared with controls and remote myocardium, T1ρ significantly increased in all patients' sub-groups and all types of myocardial injuries: acute and chronic injuries, focal and diffuse tissue abnormalities, as well as ischaemic and non-ischaemic aetiologies (P < 0.05). T1ρ was independently associated with T2 in patients with acute injuries (P = 0.004) and with native T1 and ECV in patients with chronic injuries (P < 0.05). Myocardial T1ρ mapping demonstrated good intra- and inter-observer reproducibility (intraclass correlation coefficient = 0.86 and 0.83, respectively).

Myocardial T1ρ mapping appears to be reproducible and equally sensitive to acute and chronic myocardial injuries, whether of ischaemic or non-ischaemic origins. It may thus be a contrast-agent-free biomarker for gaining new and quantitative insight into myocardial structural disorders. These findings highlight the need for further studies through prospective and randomized trials.

While single-shot late gadolinium enhancement (LGE) is useful for imaging patients with arrhythmia and/or dyspnea, it produces low spatial resolution. One approach to improve spatial resolution is to accelerate data acquisition using compressed sensing (CS). Our previous work described a single-shot, multi-inversion time (TI) LGE pulse sequence using radial k-space sampling and CS, but over-regularization resulted in significant image blurring that muted the benefits of data acceleration. The purpose of the present study was to improve the spatial resolution of the single-shot, multi-TI LGE pulse sequence by incorporating view sharing (VS) and k-space weighted contrast (KWIC) filtering into a GRASP-Pro reconstruction. In 24 patients (mean age = 61 ± 16 years; 9/15 females/males), we compared the performance of our improved multi-TI LGE and standard multi-TI LGE, where clinical standard LGE was used as a reference. Two clinical raters independently graded multi-TI images and clinical LGE images visually on a five-point Likert scale (1, nondiagnostic; 3, clinically acceptable; 5, best) for three categories: the conspicuity of myocardium or scar, artifact, and noise. The summed visual score (SVS) was defined as the sum of the three scores. Myocardial scar volume was quantified using the full-width at half-maximum method. The SVS was not significantly different between clinical breath-holding LGE (median 13.5, IQR 1.3) and multi-TI LGE (median 12.5, IQR 1.6) (P = 0.068). The myocardial scar volumes measured from clinical standard LGE and multi-TI LGE were strongly correlated (coefficient of determination, R2  = 0.99) and in good agreement (mean difference = 0.11%, lower limit of the agreement = -2.13%, upper limit of the agreement = 2.34%). The inter-rater agreement in myocardial scar volume quantification was strong (intraclass correlation coefficient = 0.79). The incorporation of VS and KWIC into GRASP-Pro improved spatial resolution. Our improved 25-fold accelerated, single-shot LGE sequence produces clinically acceptable image quality, multi-TI reconstruction, and accurate myocardial scar volume quantification.

A deep learning (DL) model that automatically detects cardiac pathologies on cardiac MRI may help streamline the diagnostic workflow. To develop a DL model to detect cardiac pathologies on cardiac MRI T1-mapping and late gadolinium phase sensitive inversion recovery (PSIR) sequences were used.

Subjects in this study were either diagnosed with cardiac pathology (n = 137) including acute and chronic myocardial infarction, myocarditis, dilated cardiomyopathy, and hypertrophic cardiomyopathy or classified as normal (n = 63). Cardiac MR imaging included T1-mapping and PSIR sequences. Subjects were split 65/15/20% for training, validation, and hold-out testing. The DL models were based on an ImageNet pretrained DenseNet-161 and implemented using PyTorch and fastai. Data augmentation with random rotation and mixup was applied. Categorical cross entropy was used as the loss function with a cyclic learning rate (1e-3). DL models for both sequences were developed separately using similar training parameters. The final model was chosen based on its performance on the validation set. Gradient-weighted class activation maps (Grad-CAMs) visualized the decision-making process of the DL model.

The DL model achieved a sensitivity, specificity, and accuracy of 100%, 38%, and 88% on PSIR images and 78%, 54%, and 70% on T1-mapping images. Grad-CAMs demonstrated that the DL model focused its attention on myocardium and cardiac pathology when evaluating MR images.

The developed DL models were able to reliably detect cardiac pathologies on cardiac MR images. The diagnostic performance of T1 mapping alone is particularly of note since it does not require a contrast agent and can be acquired quickly.