Cine Magnetic Resonance Imaging Research Articles

Cine magnetic resonance imaging sagittal plane, slow motion, postoperative. The powerful impact of the cardiac impulse was shown by the fact that even the stable aortic wall was dented (arrow).

Cine magnetic resonance imaging sagittal plane, slow motion, postoperative. The powerful impact of the cardiac impulse was shown by the fact that even the stable aortic wall was dented (arrow).

MR-guided beam gating: Residual motion, gating efficiency and dose reconstruction for stereotactic treatments of the liver and lung

PurposeThis study aims to investigate the efficiency and the geometric as well as the dosimetric benefit of magnetic-resonance guided beam gating for stereotactic treatments in moving organs. MethodPatients treated with MR-guided (MRIdian system) SBRT for lung (n = 10) and liver (n = 10) targets were analyzed. Breath-hold gating was performed based on lesion tracking in sagittal cine MRI images. The target offset from the geometric center of the gating window with and without gating was evaluated. A dose reconstruction workflow based on convolution of these 2D position-probability maps and the daily 3D dose distribution was used to estimate the daily delivered dose including motion. The dose to the clinical target volume (CTV) and to a 2-cm ring structure around the planning target volume were evaluated. ResultsThe applied gating protocol resulted in a mean (±standard deviation) gating efficiency of 55%±16%. Over all patients, the mean target offset (2D-root-mean-square error) was 8.3 ± 4.3 mm, which reduced to 2.4 ± 0.6 mm during gating. The dose reconstruction showed a mean deviation in CTV coverage (D95) from the static plans of −1.7%±1.8% with gating and −12.0%±8.4% if no gating would have been used. The mean dose (Dmean) in the ring structure, with respect to the static plans, showed mean deviations of −0.1%±0.3% with gating and −1.6%±1.8% without gating. ConclusionThe MRIdian system enables gating based on the inner anatomy and the implemented dose reconstruction workflow demonstrated geometric robust delivery of the planned radiation doses.

Cardiac Disease Classification Using Two-Dimensional Thickness and Few-Shot Learning Based on Magnetic Resonance Imaging Image Segmentation.

Cardiac cine magnetic resonance imaging (MRI) is a widely used technique for the noninvasive assessment of cardiac functions. Deep neural networks have achieved considerable progress in overcoming various challenges in cine MRI analysis. However, deep learning models cannot be used for classification because limited cine MRI data are available. To overcome this problem, features from cine image settings are derived by handcrafting and addition of other clinical features to the classical machine learning approach for ensuring the model fits the MRI device settings and image parameters required in the analysis. In this study, a novel method was proposed for classifying heart disease (cardiomyopathy patient groups) using only segmented output maps. In the encoder–decoder network, the fully convolutional EfficientNetB5-UNet was modified to perform the semantic segmentation of the MRI image slice. A two-dimensional thickness algorithm was used to combine the segmentation outputs for the 2D representation of images of the end-diastole (ED) and end-systole (ES) cardiac volumes. The thickness images were subsequently used for classification by using a few-shot model with an adaptive subspace classifier. Model performance was verified by applying the model to the 2017 MICCAI Medical Image Computing and Computer-Assisted Intervention dataset. High segmentation performance was achieved as follows: the average Dice coefficients of segmentation were 96.24% (ED) and 89.92% (ES) for the left ventricle (LV); the values for the right ventricle (RV) were 92.90% (ED) and 86.92% (ES). The values for myocardium were 88.90% (ED) and 90.48% (ES). An accuracy score of 92% was achieved in the classification of various cardiomyopathy groups without clinical features. A novel rapid analysis approach was proposed for heart disease diagnosis, especially for cardiomyopathy conditions using cine MRI based on segmented output maps.

Automatic myocardium strain quantification in MR synthetic images with Deep Leaning.

Accurate quantification of myocardium strain in magnetic resonance images is important to correctly diagnose and monitor cardiac diseases. Currently, available methods to estimate motion are based on tracking brightness pattern differences between images. In cine-MR images, the myocardium interior presents an inhered homogeneity, which reduces the accuracy in estimated motion, and consequently strain. Neural networks have recently been shown to be an important tool for a variety of applications, including motion estimation. In this work, we investigate the feasibility of quantifying myocardium strain in cardiac resonance synthetic images using motion generated by a compact and powerful network called Pyramid, Warping, and Cost Volume (PWC). Using the motion generated by the neural network, the radial myocardium strain obtained presents a mean average error of 12.30% +- 6.50%, and in the circumferential direction 1.20% +-0.61 %, better than the two classical methods evaluated. Clinical Relevance- This work demonstrates the feasibility of estimating myocardium strain using motion estimated by a convolutional neural network.

Intrafraction motion during radiotherapy of breast tumor, breast tumor bed, and individual axillary lymph nodes on cine magnetic resonance imaging.

In (ultra-)hypofractionation, the contribution of intrafraction motion to treatment accuracy becomes increasingly important. Our purpose was to evaluate intrafraction motion and resulting geometric uncertainties for breast tumor (bed) and individual axillary lymph nodes, and to compare prone and supine position for the breast tumor (bed). During 1-3min of free breathing, we acquired transverse/sagittal interleaved 1.5T cine magnetic resonance imaging (MRI) of the breast tumor (bed) in prone and supine position and coronal/sagittal cine MRI of individual axillary lymph nodes in supine position. A total of 31 prone and 23 supine breast cine MRI (in 23 women) and 52 lymph node cine MRI (in 24 women) were included. Maximum displacement, breathing amplitude, and drift were analyzed using deformable image registration. Geometric uncertainties were calculated for all displacements and for breathing motion only. Median maximum displacements (range over the three orthogonal orientations) were 1.1-1.5mm for the breast tumor (bed) in prone and 1.8-3.0mm in supine position, and 2.2-2.4mm for lymph nodes. Maximum displacements were significantly smaller in prone than in supine position, mainly due to smaller breathing amplitude: 0.6-0.9mm in prone vs. 0.9-1.4mm in supine. Systematic and random uncertainties were 0.1-0.4mm in prone position and 0.2-0.8mm in supine position for the tumor (bed), and 0.4-0.6mm for the lymph nodes. Intrafraction motion of breast tumor (bed) and individual lymph nodes was small. Motion of the tumor (bed) was smaller in prone than in supine position.

Association Between Internal Organ/Liver Tumor and External Surface Motion From Cine MR Images on an MRI-Linac.

Purposes/ObjectivesHistorically, motion correlation between internal tumor and external surrogates have been based on limited sets of X-ray or magnetic resonance (MR) images. With the recent clinical implementation of MR-guided linear accelerators, a vast quantity of continuous planar real-time MR imaging data is acquired. In this study, information was extracted from MR cine imaging during liver cancer treatments to establish associations between internal tumor/diaphragm and external surface/skin movement.Methods and MaterialsThis retrospective study used 305,644 MR image frames acquired over 118 treatment/imaging sessions of the first 23 liver cancer patients treated on an MRI-linac. 9 features were automatically determined on each MR image frame: Lung_Area, the posterior (Dia_Post), dome (Dia_Dome), and anterior (Dia_Ant) points of a diaphragmatic curve and the diaphragm curve point (Dia_Max), the chest (Chest) and the belly (Belly) skin points experiencing the maximum motion ranges; the superior-interior (SI) and posterior-anterior (PA) positions of a target. For every session, correlation analyses were performed twice among the 9 features: 1) over a breath-hold (BH) set and 2) on a pseudo free-breathing (PFB) generated by removing breath-holding frames.Results303,123 frames of images were successfully analyzed. For BH set analysis, correlation coefficients were as follows: 0.94 ± 0.07 between any two features among Dia_Post, Dia_Dome, Dia_Max, and Lung_Area; 0.95 ± 0.06 between SI and any feature among Dia_Post, Dia_Dome, Dia_Max, or Lung_Area; 0.76 ± 0.29 between SI and Belly (with 50% of correlations ≥ 0.87). The PFB set had 142,862 frames of images. For this set, correlation coefficients were 0.96 ± 0.06 between any two features among Dia_Post, Dia_Dome, Dia_Max, and Lung_Area; 0.95 ± 0.06 between SI and any feature among Dia_Post, Dia_Dome, Dia_Max, or Lung_Area; 0.80 ± 0.26 between SI and Belly (with 50% of correlations ≥ 0.91).ConclusionDiaphragmatic motion as assessed by cine MR imaging is highly correlated with liver tumor motion. Belly vertical motion is highly correlated with liver tumor longitudinal motion in approximately half of the cases. More detailed analyses of those cases displaying weak correlations are in progress.

Cine magnetic resonance imaging detects shorter cardiac rest periods in postcapillary pulmonary hypertension.

A shorter cardiac rest period within a cardiac cycle is usually thought to be a result of a fast heart rate, and its clinical relevance has long been ignored. The aim of the present study was to test the hypothesis that the length of cardiac rest periods is altered in postcapillary pulmonary hypertension (PH). Twenty-six patients with postcapillary PH and 20 healthy controls were recruited for cardiac magnetic resonance imaging (MRI) scans. All participants had a heart rate no higher than 80 beats/minute. Cine magnetic resonance imaging (MRI, acquired at a four-chamber view) was analyzed to determine the length of cardiac rest periods at end-systole and mid-to-late diastole. PH patients had a shorter rest period at mid-to-late diastole than controls (17.5 ± 8.7% vs. 24.2 ± 4.2%, P = 0.003). Receiver operating characteristic (ROC) curves showed that the proportion of the rest period in diastole (defined as the length of diastasis/diastole) can discriminate PH patients from controls [area under the curve (AUC) = 0.83, 95% confidence interval (CI): 0.71-0.96]. The existence of postcapillary PH was a significant contributor (β = -5.537, P = 0.023) to shorter cardiac rest periods at mid-to-late diastole after adjusting for potential confounders, including age, sex, heart rate, and blood pressure. Postcapillary PH is independently associated with shorter cardiac rest periods at mid-to-late diastole. The length of cardiac rest periods has the potential to become a novel quantitative imaging biomarker for indicating cardiovascular health.

A Study of Left Ventricular (LV) Segmentation on Cardiac Cine-MR Images

Left ventricular segmentation from cardiac images has high impact to have early diagnosis of various cardiovascular disorders. However, it is really a challenging task to segment left ventricular images from magnetic resonance image (MRI). In this paper, we explore several state-of-the-art segmentation algorithms applied on left ventricular (LV) segmentation on cardiac cine-MR images. Both adaptive and global thresholding algorithms along with region-based segmentation algorithm have been explored. Edge-based segmentation is disregard due to the absence of edge information in the employed dataset. For evaluation, we explored a benchmark dataset that was used for the MICCAI 3D segmentation challenge. We found that the cardiac MRI global thresholding has proved to be much efficient and robust than the adaptive thresholding. We achieved more than 92% accuracy for global thresholding, whereas, about 78% accuracy for the adaptive thresholding approach. The use of entropy or histogram to characterize segmentation in place of the intensity value of the pixel has a vital effect on segmentation efficiency. It is evident that the intensity information is corrupted by acquisition procedure, as well as the structure of organs. Due to the lack of boundary information in cardiac cine-MRI, clustering and region-based segmentation have produced more than 93% segmentation accuracy. For the case of soft clustering, the increased accuracy is found as 96%. However, more explorations are required, specially based on deep learning approaches on very large datasets.

Myocardial Strain Evaluation with Cardiovascular MRI: Physics, Principles, and Clinical Applications.

Myocardial strain is a measure of myocardial deformation, which is a more sensitive imaging biomarker of myocardial disease than the commonly used ventricular ejection fraction. Although myocardial strain is commonly evaluated by using speckle-tracking echocardiography, cardiovascular MRI (CMR) is increasingly performed for this purpose. The most common CMR technique is feature tracking (FT), which involves postprocessing of routinely acquired cine MR images. Other CMR strain techniques require dedicated sequences, including myocardial tagging, strain-encoded imaging, displacement encoding with stimulated echoes, and tissue phase mapping. The complex systolic motion of the heart can be resolved into longitudinal strain, circumferential strain, radial strain, and torsion. Myocardial strain metrics include strain, strain rate, displacement, velocity, torsion, and torsion rate. Wide variability exists in the reference ranges for strain dependent on the imaging technique, analysis software, operator, patient demographics, and hemodynamic factors. In anticancer therapy cardiotoxicity, CMR myocardial strain can help identify left ventricular dysfunction before the decline of ejection fraction. CMR myocardial strain is also valuable for identifying patients with left ventricle dyssynchrony who will benefit from cardiac resynchronization therapy. CMR myocardial strain is also useful in ischemic heart disease, cardiomyopathies, pulmonary hypertension, and congenital heart disease. The authors review the physics, principles, and clinical applications of CMR strain techniques. Online supplemental material is available for this article. ©RSNA, 2022.

Sleep Endoscopy and Cine Magnetic Resonance Imaging Evaluation of Children With Persistent Obstructive Sleep Apnea.

To compare findings of same-day cine magnetic resonance imaging (MRI) and drug-induced sleep endoscopy (DISE) and examine how each technique uniquely contributes to the evaluation of persistent obstructive sleep apnea following adenotonsillectomy. Retrospective cohort study. Quaternary care center. Chart review was performed for consecutive patients who underwent same-day cine MRI and DISE between 2015 and 2020. Descriptive statistics are reported, and Cohen kappa coefficients were calculated to evaluate the agreement between cine MRI and DISE for obstruction at the adenoids, lingual tonsils, and tongue base. There were 137 patients, the mean age was 10.4 years (95% CI, 3.2-16.7), and 62.8% were male. The most common sites of obstruction on DISE were the tongue base (86.9%), velum (78.7%), epiglottis (74.5%), inferior turbinate (68.6%), and lingual tonsil (61.3%). The most common sites of obstruction on cine MRI were the hypopharynx (56.3%), tongue base (44.8%), lingual tonsil (38.0%), and macroglossia (37.6%). There was moderate agreement for adenoid hypertrophy (κ = 0.53) and poor agreement for lingual tonsil hypertrophy (κ = 0.15) and tongue base obstruction (κ = 0.09). DISE identified more instances of multilevel obstruction when compared with cine MRI (94.9% vs 48.2%). DISE offered a better examination of nasal and supraglottic obstruction and is sensitive to partial vs complete collapse, while cine MRI offered better soft tissue resolution for lymphoid tissue hypertrophy and provided a global view of primary and secondary airway obstruction. Cine MRI and DISE are complementary modalities in the evaluation of children with persistent obstructive sleep apnea.

CMRI2SPEC: CINE MRI SEQUENCE TO SPECTROGRAM SYNTHESIS VIA A PAIRWISE HETEROGENEOUS TRANSLATOR.

Multimodal representation learning using visual movements from cine magnetic resonance imaging (MRI) and their acoustics has shown great potential to learn shared representation and to predict one modality from another. Here, we propose a new synthesis framework to translate from cine MRI sequences to spectrograms with a limited dataset size. Our framework hinges on a novel fully convolutional heterogeneous translator, with a 3D CNN encoder for efficient sequence encoding and a 2D transpose convolution decoder. In addition, a pairwise correlation of the samples with the same speech word is utilized with a latent space representation disentanglement scheme. Furthermore, an adversarial training approach with generative adversarial networks is incorporated to provide enhanced realism on our generated spectrograms. Our experimental results, carried out with a total of 63 cine MRI sequences alongside speech acoustics, show that our framework improves synthesis accuracy, compared with competing methods. Our framework thereby has shown the potential to aid in better understanding the relationship between the two modalities.

Assessment of the relationship between regional wall motion abnormality score revealed by parametric imaging and the extent of LGE with CMR

The aim of this study was to assess the relationship between left ventricular (LV) regional myocardial wall motion abnormality (WMA), revealed by visual interpretation of cardiac magnetic resonance (CMR) cine images together with the computed wall motion parametric image, and the transmural scar extent, as assessed by Late gadolinium Enhancement (LGE), in 40 patients. Each cine CMR short-axis loop was processed to compute a parametric image where each pixel represents the amplitude of the Hilbert transform of videointensity over time. Two expert radiologists blindly interpreted the cine CMR images in combination with the corresponding parametric image to assign a WMA score for each of the 16 myocardial sectors in which the LV myocardium was subdivided. Such score was compared per sector to the level of transmural scar extent obtained by LGE images.A total of 592 myocardial segments were analyzed. A significant decrease in regional wall motion was observed in sectors with LGE transmural hyperenhancement > 75% of tissue, as well as a correlation between parametric image amplitude and peak radial and circumferential strain, computed by feature tracking. The results showed a reduction in prediction error Lambda of WMA from LGE of 65%, and of LGE from WMA of 63%. In particular, the estimated probability of correct prediction of WMA from LGE was 76%, while that of LGE from WMA was 75%.The interpretation of myocardial viability by LGE images combined with the WMA information, derived from cine CMR and parametric images, could improve the clinical decision making process.

Right ventricular strain and volume analyses through deep learning-based fully automatic segmentation based on radial long-axis reconstruction of short-axis cine magnetic resonance images.

We propose a deep learning-based fully automatic right ventricle (RV) segmentation technique that targets radially reconstructed long-axis (RLA) images of the center of the RV region in routine short axis (SA) cardiovascular magnetic resonance (CMR) images. Accordingly, the purpose of this study is to compare the accuracy of deep learning-based fully automatic segmentation of RLA images with the accuracy of conventional deep learning-based segmentation in SA orientation in terms of the measurements of RV strain parameters. We compared the accuracies of the above-mentioned methods in RV segmentations and in measuring RV strain parameters by Dice similarity coefficients (DSCs) and correlation coefficients. DSC of RV segmentation of the RLA method exhibited a higher value than those of the conventional SA methods (0.84 vs. 0.61). Correlation coefficient with respect to manual RV strain measurements in the fully automatic RLA were superior to those in SA measurements (0.5-0.7 vs. 0.1-0.2). Our proposed RLA realizes accurate fully automatic extraction of the entire RV region from an available CMR cine image without any additional imaging. Our findings overcome the complexity of image analysis in CMR without the limitations of the RV visualization in echocardiography.

Development of a novel airbag system of abdominal compression for reducing respiratory motion: preliminary results in healthy volunteers.

This study used cine-magnetic resonance imaging (cine-MRI) to evaluate the safety and efficacy of a novel airbag system combined with a shell-type body fixation system in reducing respiratory motion in normal volunteers. The airbag system consists of a six-sided polygon inflatable airbag, a same shape plate, a stiff air supply tube, an air-supply pump and a digital pressure load cell monitor. Piezoelectric sensors were installed in the plate to detect compression pressure load changes; pressure load data were transferred to the digital pressure load cell monitor through Bluetooth. Five volunteers underwent cine-MRI with and without airbag compression to detect differences in the respiratory motion of the organs. The volunteers’ physiologic signs were stable during the experiment. The maximum inspiration pressure load was 4.48 ± 0.86 kgf (range, 4.00–6.00 kgf), while the minimum expiration pressure load was 3.69 ± 0.95 kgf (range, 2.8–5.3 kgf). Under airbag compression, the right diaphragm movement was reduced from 19.50 ± 6.43 mm to 9.60 ± 3.61 mm (P < 0.05) in the coronal plane and 23.12 ± 6.30 mm to 11.00 ± 3.69 mm (P < 0.05) in the sagittal plane. The left diaphragm, pancreas and liver in the coronal plane and the right kidney and liver in the sagittal plane also showed significant movement reduction. This novel airbag abdominal compression system was found to be safe during the experiment and successful in the reduction of internal organ respiratory motion and promises to be a convenient and efficient tool for clinical radiotherapy.

Significance of aortoseptal angle anomalies to left ventricular hemodynamics and subaortic stenosis: A numerical study

Discrete subaortic stenosis (DSS) is an obstructive cardiac disease caused by a membranous lesion in the left ventricular (LV) outflow tract (LVOT). Although its etiology is unknown, the higher prevalence of DSS in LVOT anatomies featuring a steep aortoseptal angle (AoSA) suggests a potential role for hemodynamics. Therefore, the objective of this study was to quantify the impact of AoSA steepening on the LV three-dimensional (3D) hemodynamic stress environment. A 3D LV model reconstructed from cardiac cine-magnetic resonance imaging was connected to four LVOT geometrical variations spanning the clinical AoSA range (115°-160°). LV hemodynamic stresses were characterized in terms of cycle-averaged pressure, temporal shear magnitude (TSM), and oscillatory shear index. The wall shear stress (WSS) topological skeleton was further analyzed by computing the scaled divergence of the WSS vector field. AoSA steepening caused an increasingly perturbed subaortic flow marked by LVOT flow skewness and complex 3D secondary flow patterns. These disturbances generated WSS overloads (>45% increase in TSM vs. 160° model) on the inferior LVOT wall, and increased WSS contraction (>66% decrease in WSS divergence vs. 160° model) in regions prone to DSS membrane formation. AoSA steepening generated substantial hemodynamic stress abnormalities in LVOT regions prone to DSS formation. Further studies are needed to assess the possible impact of such mechanical abnormalities on the tissue and cellular responses.

Reference Values for Pediatric Atrial Volumes Assessed by Steady-State Free-Precession Magnetic Resonance Imaging Using Monoplane and Biplane Area-Length Methods.

Measurement of atrial volumes by MRI is becoming increasingly important in pediatric cardiac disorders. However, MRI normal values for atrial volumes in children are lacking. To establish pediatric reference values for atrial volumes. Retrospective. A total of 155 healthy children from two large institutions (103 male, age 13.9± 2.8 years, range 4-18 years). A 1.5 T; balanced steady-state free precession (bSSFP) sequence. The monoplane and biplane area-length methods were used to measure minimal and maximal left and right atrial volumes (LAmin , LAmax , RAmin , and RAmax ) from four-chamber (4ch) and two-chamber (2ch) MR cine images. Centile charts and tables for atrial volumes were created. Descriptive statistics, lambda-mu-sigma (LMS)-method of Cole and Green, univariable and multivariable linear regression models. A P value < 0.05 was considered to be statistically significant. In the multivariable linear model, body surface area was significantly associated with all atrial volumes and sex was significantly associated with RA volumes, LA volumes measured in the 2ch-view as well as biplane LAmax. Average atrial volumes measured: monoplane 4ch: LAmin 13.1± 4.8mL/m2 , LAmax 33.4± 8.8mL/m2 , RAmin 18.5± 6.8mL/m2 , RAmax 33.2± 9.6mL/m2 ; monoplane 2ch: LAmin 12.7± 4.9mL/m2 , LAmax 30.5± 9.5mL/m2 ; biplane: LAmin 12.3± 4.5mL/m2 , LAmax 30.9± 8.7mL/m2 . Pediatric MRI reference values for atrial volumes have been provided. 2 EVIDENCE LEVEL: 4.

DeU-Net 2.0: Enhanced deformable U-Net for 3D cardiac cine MRI segmentation.

Automatic segmentation of cardiac magnetic resonance imaging (MRI) facilitates efficient and accurate volume measurement in clinical applications. However, due to anisotropic resolution, ambiguous borders and complicated shapes, existing methods suffer from the degradation of accuracy and robustness in cardiac MRI segmentation. In this paper, we propose an enhanced Deformable U-Net (DeU-Net) for 3D cardiac cine MRI segmentation, composed of three modules, namely Temporal Deformable Aggregation Module (TDAM), Enhanced Deformable Attention Network (EDAN), and Probabilistic Noise Correction Module (PNCM). TDAM first takes consecutive cardiac MR slices (including a target slice and its neighboring reference slices) as input, and extracts spatio-temporal information by an offset prediction network to generate fused features of the target slice. Then the fused features are also fed into EDAN that exploits several flexible deformable convolutional layers and generates clear borders of every segmentation map. A Multi-Scale Attention Module (MSAM) in EDAN is proposed to capture long range dependencies between features of different scales. Meanwhile, PNCM treats the fused features as a distribution to quantify uncertainty. Experimental results show that our DeU-Net achieves the state-of-the-art performance in terms of the commonly used evaluation metrics on the Extended ACDC dataset and competitive performance on other two datasets, validating the robustness and generalization of DeU-Net.

Detection of Fibrosis in Cine Magnetic Resonance Images Using Artificial Intelligence Techniques

Background: Artificial intelligence techniques have demonstrated great potential in cardiology, especially to detect imperceptible patterns for the human eye. In this sense, these techniques seem to be adequate to identify patterns in the myocardial texture which could lead to characterize and quantify fibrosis. Purpose: The aim of this study was to postulate a new artificial intelligence method to identify fibrosis in cine cardiac magnetic resonance (CMR) imaging. Methods: A retrospective observational study was carried out in a population of 75 subjects from a clinical center of San Carlos de Bariloche. The proposed method analyzes the myocardial texture in cine CMR images using a convolutional neural network to determine local myocardial tissue damage. Results: An accuracy of 89% for quantifying local tissue damage was observed for the validation data set and 70% for the test set. In addition, the qualitative analysis showed a high spatial correlation in lesion location. Conclusions: The postulated method enables to spatially identify fibrosis using only the information from cine nuclear magnetic resonance studies, demonstrating the potential of this technique to quantify myocardial viability in the future or to study the lesions etiology

DeepStrain Evidence of Asymptomatic Left Ventricular Diastolic and Systolic Dysfunction in Young Adults With Cardiac Risk Factors.

PurposeTo evaluate if a fully-automatic deep learning method for myocardial strain analysis based on magnetic resonance imaging (MRI) cine images can detect asymptomatic dysfunction in young adults with cardiac risk factors.MethodsAn automated workflow termed DeepStrain was implemented using two U-Net models for segmentation and motion tracking. DeepStrain was trained and tested using short-axis cine-MRI images from healthy subjects and patients with cardiac disease. Subsequently, subjects aged 18–45 years were prospectively recruited and classified among age- and gender-matched groups: risk factor group (RFG) 1 including overweight without hypertension or type 2 diabetes; RFG2 including hypertension without type 2 diabetes, regardless of overweight; RFG3 including type 2 diabetes, regardless of overweight or hypertension. Subjects underwent cardiac short-axis cine-MRI image acquisition. Differences in DeepStrain-based left ventricular global circumferential and radial strain and strain rate among groups were evaluated.ResultsThe cohort consisted of 119 participants: 30 controls, 39 in RFG1, 30 in RFG2, and 20 in RFG3. Despite comparable (>0.05) left-ventricular mass, volumes, and ejection fraction, all groups (RFG1, RFG2, RFG3) showed signs of asymptomatic left ventricular diastolic and systolic dysfunction, evidenced by lower circumferential early-diastolic strain rate (<0.05, <0.001, <0.01), and lower septal circumferential end-systolic strain (<0.001, <0.05, <0.001) compared with controls. Multivariate linear regression showed that body surface area correlated negatively with all strain measures (<0.01), and mean arterial pressure correlated negatively with early-diastolic strain rate (<0.01).ConclusionDeepStrain fully-automatically provided evidence of asymptomatic left ventricular diastolic and systolic dysfunction in asymptomatic young adults with overweight, hypertension, and type 2 diabetes risk factors.

Structure-aware Unsupervised Tagged-to-Cine MRI Synthesis with Self Disentanglement.

Cycle reconstruction regularized adversarial training-e.g., CycleGAN, DiscoGAN, and DualGAN-has been widely used for image style transfer with unpaired training data. Several recent works, however, have shown that local distortions are frequent, and structural consistency cannot be guaranteed. Targeting this issue, prior works usually relied on additional segmentation or consistent feature extraction steps that are task-specific. To counter this, this work aims to learn a general add-on structural feature extractor, by explicitly enforcing the structural alignment between an input and its synthesized image. Specifically, we propose a novel input-output image patches self-training scheme to achieve a disentanglement of underlying anatomical structures and imaging modalities. The translator and structure encoder are updated, following an alternating training protocol. In addition, the information w.r.t. imaging modality can be eliminated with an asymmetric adversarial game. We train, validate, and test our network on 1,768, 416, and 1,560 unpaired subject-independent slices of tagged and cine magnetic resonance imaging from a total of twenty healthy subjects, respectively, demonstrating superior performance over competing methods.