Compensation for respiratory motion-induced signal loss and phase corruption in free-breathing self-navigated cine DENSE using deep learning

In vivo human knee varus-valgus loading apparatus for analysis of MRI-based intratissue strain and relaxometry

The diagnosis of early-stage osteoarthritis remains as an unmet challenge in medicine and a roadblock to evaluating the efficacy of disease-modifying treatments. Recent studies demonstrate that unique patterns of intratissue cartilage deformation under cyclic loading can serve as potential biomarkers to detect early disease pathogenesis. However, a workflow to obtain deformation, strain maps, and quantitative MRI metrics due to the loading of articular cartilage in vivo has not been fully developed. In this study, we characterize and demonstrate an apparatus that is capable of applying a varus-valgus load to the human knee in vivo within an MRI environment to enable the measurement of cartilage structure and mechanical function. The apparatus was first tested in a lab environment, then the functionality and utility of the apparatus were examined during varus loading in a clinical 3T MRI system for human imaging. We found that the device enables quantitative MRI metrics for biomechanics and relaxometry data acquisition during joint loading leading to compression of the medial knee compartment. Integration with spiral DENSE MRI during cyclic loading provided time-dependent displacement and strain maps within the tibiofemoral cartilage. The results from these procedures demonstrate that the performance of this loading apparatus meets the design criteria and enables a simple and practical workflow for future studies of clinical cohorts, and the identification and validation of imaging-based biomechanical biomarkers.

Accelerated four-dimensional free-breathing whole-liver water-fat magnetic resonance imaging with deep dictionary learning and chemical shift modeling

Background

Multi-echo chemical-shift-encoded magnetic resonance imaging (MRI) has been widely used for fat quantification and fat suppression in clinical liver examinations. Clinical liver water-fat imaging typically requires breath-hold acquisitions, with the free-breathing acquisition method being more desirable for patient comfort. However, the acquisition for free-breathing imaging could take up to several minutes. The purpose of this study is to accelerate four-dimensional free-breathing whole-liver water-fat MRI by jointly using high-dimensional deep dictionary learning and model-guided (MG) reconstruction.

Methods

A high-dimensional model-guided deep dictionary learning (HMDDL) algorithm is proposed for the acceleration. The HMDDL combines the powers of the high-dimensional dictionary learning neural network (hdDLNN) and the chemical shift model. The neural network utilizes the prior information of the dynamic multi-echo data in spatial respiratory motion, and echo dimensions to exploit the features of images. The chemical shift model is used to guide the reconstruction of field maps, R 2 ∗ maps, water images, and fat images. Data acquired from ten healthy subjects and ten subjects with clinically diagnosed nonalcoholic fatty liver disease (NAFLD) were selected for training. Data acquired from one healthy subject and two NAFLD subjects were selected for validation. Data acquired from five healthy subjects and five NAFLD subjects were selected for testing. A three-dimensional (3D) blipped golden-angle stack-of-stars multi-gradient-echo pulse sequence was designed to accelerate the data acquisition. The retrospectively undersampled data were used for training, and the prospectively undersampled data were used for testing. The performance of the HMDDL was evaluated in comparison with the compressed sensing-based water-fat separation (CS-WF) algorithm and a parallel non-Cartesian recurrent neural network (PNCRNN) algorithm.

Results

Four-dimensional water-fat images with ten motion states for whole-liver are demonstrated at several R values. In comparison with the CS-WF and PNCRNN, the HMDDL improved the mean peak signal-to-noise ratio (PSNR) of images by 9.93 and 2.20 dB, respectively, and improved the mean structure similarity (SSIM) of images by 0.058 and 0.009, respectively, at R=10. The paired t-test shows that there was no significant difference between HMDDL and ground truth for proton-density fat fraction (PDFF) and R 2 ∗ values at R up to 10.

Conclusions

The proposed HMDDL enables features of water images and fat images from the highly undersampled multi-echo data along spatial, respiratory motion, and echo dimensions, to improve the performance of accelerated four-dimensional (4D) free-breathing water-fat imaging.

Comparison between diaphragmatic-navigated and self-navigated coronary magnetic resonance angiography at 3T in pediatric patients with congenital coronary artery anomalies

Background

Coronary magnetic resonance angiography (CMRA) is being increasingly used in pediatric patients with congenital coronary artery anomalies (CAAs). However, the data on the free-breathing self-navigation technique, which has the potential to simplify the acquisition plan with a high success rate at 3T, remain scarce. This study investigated the clinical application value of self-navigated (sNAV) CMRA at 3T in pediatric patients with suspected CAAs and compared it to conventional diaphragmatic-navigated (dNAV) CMRA.

Methods

From April 2019 to March 2022, we enrolled 65 pediatric patients (38 males and 27 females; mean age 8.5±4.4 years) with suspected CAAs in this prospective study. All patients underwent both dNAV and sNAV sequences in random order with gradient recalled echo (GRE) sequence during free breathing, with 39 (20 males and 19 females; mean age 10.2±3.6 years) of them additionally undergoing coronary computed tomography angiography (CCTA) or invasive coronary angiography (ICA). We measured and compared the success rate, scan time, visual score of the 9 main coronary artery segments, vessel sharpness, and vessel length between the two sequences. The diagnostic accuracy was compared using CCTA or ICA as a reference.

Results

The success rate of sNAV-CMRA (65/65, 100%) was higher than that of dNAV-CMRA (61/65, 93.8%) (P<0.001), and the scan time of sNAV-CMRA (7.3±2.5 min) was significantly shorter than that of dNAV-CMRA (9.1±3.6 min) (P=0.002). The acquisition efficiency of dNAV-CMRA was 40.5%±12.9%, while for sNAV-CMRA, 100% acquisition efficiency was achieved. There was no significant difference in vessel length of any of the coronary arteries, visual score, or vessel sharpness of the left circumflex coronary artery (LCX) between the two sequences (all P values >0.050). The visual score and vessel sharpness of the right coronary artery and left anterior descending coronary artery (LAD) were significantly improved in dNAV-CMRA compared with sNAV-CMRA (all P values <0.050). The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for the detection of CAAs were not significantly different between the two sequences (all P values >0.050).

Conclusions

Our findings demonstrated that both sNAV and dNAV in CMRA provide clinical application value in pediatric patients with CAAs and have similar diagnostic performance. Although the image quality of sNAV-CMRA is slightly inferior compared to that of dNAV-CMRA, sNAV-CMRA allows for a simpler scanning procedure.

StrainNet: Improved Myocardial Strain Analysis of Cine MRI by Deep Learning from DENSE

Purpose

To develop a three-dimensional (two dimensions + time) convolutional neural network trained with displacement encoding with stimulated echoes (DENSE) data for displacement and strain analysis of cine MRI.

Materials and Methods

In this retrospective multicenter study, a deep learning model (StrainNet) was developed to predict intramyocardial displacement from contour motion. Patients with various heart diseases and healthy controls underwent cardiac MRI examinations with DENSE between August 2008 and January 2022. Network training inputs were a time series of myocardial contours from DENSE magnitude images, and ground truth data were DENSE displacement measurements. Model performance was evaluated using pixelwise end-point error (EPE). For testing, StrainNet was applied to contour motion from cine MRI. Global and segmental circumferential strain (Ecc) derived from commercial feature tracking (FT), StrainNet, and DENSE (reference) were compared using intraclass correlation coefficients (ICCs), Pearson correlations, Bland-Altman analyses, paired t tests, and linear mixed-effects models.

Results

The study included 161 patients (110 men; mean age, 61 years ± 14 [SD]), 99 healthy adults (44 men; mean age, 35 years ± 15), and 45 healthy children and adolescents (21 males; mean age, 12 years ± 3). StrainNet showed good agreement with DENSE for intramyocardial displacement, with an average EPE of 0.75 mm ± 0.35. The ICCs between StrainNet and DENSE and FT and DENSE were 0.87 and 0.72, respectively, for global Ecc and 0.75 and 0.48, respectively, for segmental Ecc. Bland-Altman analysis showed that StrainNet had better agreement than FT with DENSE for global and segmental Ecc.

Conclusion

StrainNet outperformed FT for global and segmental Ecc analysis of cine MRI.Keywords: Image Postprocessing, MR Imaging, Cardiac, Heart, Pediatrics, Technical Aspects, Technology Assessment, Strain, Deep Learning, DENSE Supplemental material is available for this article. © RSNA, 2023.