Delayed Enhancement Magnetic Resonance Imaging Research Articles

Correlation between ventricular local impedance and tissue composition by MRI in chronic infarction

Abstract Background Local impedance (LI) mapping is feasible and provides additional tissue characterization of the ventricular tachycardia substrate. Data on tissue composition underlying different LI values are lacking. Purpose To correlate LI with tissue composition evaluated by magnetic resonance imaging (MRI) in a chronic myocardial infarction (MI) swine model. Methods One-month after a non-reperfused anterior MI, eighteen Landrace X Large White pigs underwent delayed-enhancement MRI (deMRI) and endocardial LI mapping. deMRI images were post-processed off-line with commercially available dedicated software; scar subtypes (with thresholds set at 40-60% of maximal pixel signal intensity [PSI]), border-zone corridors and left ventricular wall thickness were measured. LI thresholds were set using the blood-pool LI value to define low, intermediate, and high tissue resistance subtypes. LI maps were co-registered with deMRI for analyses. Results Low LI zones correlated with deMRI dense and transmural scar (exhibiting endo and epicardial PSI >60% in 91% and 80% of the cases, respectively). Intermediate LI tissue exhibited predominantly subendocardial scar with more heterogeneous composition (8%, 47% and 45% of the points had <40%, 40-60% and >60% endocardial PSI, respectively), and with less epicardial involvement (50% of points had epicardial PSI <40%). Border-zone corridors co-localized with intermediate LI tissue in most of the cases (77%). Low LI zones exhibited more pronounced wall thinning compared to intermediate LI tissue (p<0.001) (Figure 1). Conclusions LI correlated with scar density and transmurality on deMRI. Areas of low LI had higher proportion of dense, transmural scar and wall thinning compared to intermediate LI areas. MRI-detected border-zone corridors colocalized with intermediate LI in most cases.LI and deMRI data (endo and epi scar)LI and deMRI data (tissue thickness)

Detection of occult thrombosis in individuals with Fontan circulation by cardiac MRI.

Identifying thrombus formation in Fontan circulation has been highly variable, with reports between 17 and 33%. Initially, thrombus detection was mainly done through echocardiograms. Delayed-enhancement cardiac MRI is emerging as a more effective imaging technique for thrombus identification. This study aims to determine the prevalence of occult cardiac thrombosis in patients undergoing clinically indicated cardiac MRI. A retrospective chart review of children and adults in the Duke University Hospital Fontan registry who underwent delayed-enhancement cardiac MRI. Individuals were excluded if they never received a delayed-enhancement cardiac MRI or had insufficient data. Demographic characteristics, native heart anatomy, cardiac MRI measurements, and thromboembolic events were collected for all patients. In total, 119 unique individuals met inclusion criteria with a total of 171 scans. The median age at Fontan procedure was 3 (interquartile range 1, 4) years. The majority of patients had dominant systemic right ventricle. Cardiac function was relatively unchanged from the first cardiac MRI to the third cardiac MRI. While 36.4% had a thrombotic event by history, only 0.5% (1 patient) had an intracardiac thrombus detected by delayed-enhancement cardiac MRI. Despite previous echocardiographic reports of high prevalence of occult thrombosis in patients with Fontan circulation, we found very low prevalence using delayed-enhancement cardiac MRI. As more individuals are reaching adulthood after requiring early Fontan procedures in childhood, further work is needed to develop thrombus-screening protocols as a part of anticoagulation management.

A novel radiomics nomogram based on T2-sampling perfection with application-optimized contrasts using different flip-angle evolutions (SPACE) images for predicting cochlear and vestibular endolymphatic hydrops in Meniere's disease patients.

To construct and validate a radiomics nomogram based on T2-sampling perfection with application-optimized contrasts using different flip-angle evolutions (SPACE) images for predicting cochlear and vestibular endolymphatic hydrops (EH) in Meniere's disease patients. A total of 156 patients (312 affected ears) with bilateral definite Meniere's disease who underwent delayed enhancement MRI scans were enrolled in this study. All ears of the patients were divided into a training set (n = 218) and an internal validation set (n = 94). A radiomics nomogram was constructed from radiomics features extracted from the T2-SPACE images, and a radiomics score was calculated. Performance of the radiomics nomogram was assessed using receiver operating characteristics analysis. Five features were selected for the construction of the cochlear radiomics nomogram, and seven features for the vestibular radiomics nomogram. The radiomics nomograms exhibited robust performance in differentiating between EH-positive and EH-negative statuses in both training and validation cohorts, with the area under the receiver operating characteristics curve values for cochlear and vestibular radiomic nomograms being 0.703 and 0.728 in the training set, and 0.718 and 0.701 in the validation set, respectively. The novel radiomics nomograms based on T2-SPACE images were successfully constructed to predict cochlear and vestibular EH in Meniere's disease. The models showed a solid and superior performance and may play an important role in the EH prediction. We constructed a novel radiomics nomogram, which can be a very useful tool for predicting cochlear and vestibular endolymphatic hydrops in Meniere's disease patients. • This is the first T2-SPACE-based nomogram to predict cochlear and vestibular endolymphatic hydrops. • The nomogram is of great value to patients who are unable to undergo delayed enhancement MRI scans.

A 3-hour time interval may not be sufficient for delayed enhancement magnetic resonance imaging with intravenous gadoteridol injection based on 3d-real IR sequence of the inner ear in Meniere’s disease patient

Background Magnetic resonance imaging (MRI) can be applied to visualize endolymphatic hydrops (EH). Aims/Objectives To explore whether a 3-h time interval was feasible for clinical practice. Materials and methods We prospectively enrolled 15 patients with unilateral Meniere’s disease, each of whom underwent delayed enhancement MRI scan of the inner ear after intravenous gadoteridol injection at a 3-h interval. The ears of these patients were divided into two groups (group A: the affected ears; group B: the unaffected ears). Among the two groups, the signal intensity in perilymphatic area of the basal turn of cochlea, the results of visual evaluations in the vestibule, cochlea and semicircular canal and the detection results of EH were compared. Results Regarding the signal intensity, a difference was found between group A and group B (p = .016). Besides, no difference was found between the visual evaluations in the vestibule, cochlea and semicircular canal of the two groups. Regarding the detection results of EH, group A (6 vestibules were undiagnosable; 8 cochleae were undiagnosable); group B (9 vestibules were undiagnosable; 10 cochleae were undiagnosable). Conclusions and significance In the clinical application of gadoteridol for the inner ear, 3-h delayed MR imaging may not be sufficient.

Enhancing Myocardial Disease Prediction with DOC-NET+ Architecture: A Custom Data Analysis Approach for the EMIDEC Challenge

Public health experts are deeply concerned about cardiovascular diseases, including numerous heart-related ailments that can prove fatal. Distinguishing between Myocarditis and myocardial infarction (MI) is difficult due to comparable symptoms and diagnostic complications. While endomyocardial biopsy (EMB), cardiac troponin indicators, electrocardiography (ECG), and echocardiography are useful in making preliminary diagnoses, their effectiveness is limited. The EMIDEC challenge initiative stands out as an important development in this context. It demonstrates how clinical physiological parameters, combined with delayed enhancement magnetic resonance imaging (DE-MRI), can significantly enhance classification reliability.The aim of this research is to provide a thorough assessment of the present state of the art in EMIDEC-related research. Subsequently, our intention is to establish a new database that will serve as the framework for comprehensive testing and evaluation of the architectural models drawn from the literature.We established a constructive collaboration with the Military Hospital of Tunis, which was crucial for patient data collection and access to a diverse group of patients diagnosed with Myocarditis and MI. We proceeded to select the appropriate architectural models for our study. We chose the DOC-NET and DOC-NET+ models based on prior research that demonstrated their effectiveness in similar classification contexts.DOC-NET and DOC-NET+ demonstrated accuracy rates of 95% and 100% in previous studies with the EMIDEC dataset. When these models were applied to our newly produced custom dataset, DOC-NET maintained its impressive performance with an accuracy score of 97%, while DOC-NET+ had an accuracy score of 98%. However, as we increased both the number of patients and the complexity of scenarios in our new dataset, the accuracy of DOC-NET+ dropped significantly to 91%.The CMR with AI combination allows for more customized and efficient evaluation, ultimately increasing patient outcomes and altering the cardiovascular healthcare environment. However, problems connected to the algorithm’s generalizability to complicated data continue to be a crucial concern in further refining these improvements.

Automatic Myocardium Segmentation in Delayed-Enhancement MRI with Pathology-Specific Data Augmentation and Deep Learning Architectures

The extent of myocardial infarction (MI) can be evaluated thanks to delayed enhancement (DE) cardiac MRI. DE MRI is an imaging technique acquired several minutes after the injection of a contrast agent where MI appears with a bright signal. The automatic myocardium segmentation in DE MRI is quite challenging, especially when MI is present, since these areas usually showcase a heterogeneous aspect in terms of shape and intensity, thus obstructing the myocardium visibility. To overcome this issue, we propose an image processing-based data augmentation algorithm where diverse synthetic cases of MI were created in two different ways: fixed and adaptive. In the first one, the training set is enlarged by a specific factor, whereas in the second, the method receives feedback from the segmentation model during training and performs the augmentation exclusively on complex cases. The method performance was evaluated in single and multi-modality settings. In this latter, information from kinetic images (Cine MRI), which are acquired along DE MRI in the same examination, is also used, and the extracted features from both modalities are fused. The results show that applying the data augmentation in a fixed fashion on a multi-modality setting leads to a more consistent segmentation of the myocardium in DE MRI. The segmentation models, which were all UNet-based architectures, can better relate MI areas with the myocardium, thus increasing its overall robustness to pathology-specific local pattern perturbations.

Segmentation of cardiac infarction in delayed-enhancement MRI using probability map and transformers-based neural networks

Background and Objective: Automatic segmentation of myocardial infarction is of great clinical interest for the quantitative evaluation of myocardial infarction (MI). Late Gadolinium Enhancement cardiac MRI (LGE-MRI) is commonly used in clinical practice to quantify MI, which is crucial for clinical diagnosis and treatment of cardiac diseases. However, the segmentation of infarcted tissue in LGE-MRI is highly challenging due to its high anisotropy and inhomogeneities.Methods: The innovative aspect of our work lies in the utilization of a probability map of the healthy myocardium to guide the localization of infarction, as well as the combination of 2D U-Net and U-Net transformers to achieve the final segmentation. Instead of employing a binary segmentation map, we propose using a probability map of the normal myocardium, obtained through a dedicated 2D U-Net. To leverage spatial information, we employ a U-Net transformers network where we incorporate the probability map into the original image as an additional input. Then, To address the limitations of U-Net in segmenting accurately the contours, we introduce an adapted loss function.Results: Our method has been evaluated on the 2020 MICCAI EMIDEC challenge dataset, yielding competitive results. Specifically, we achieved a Dice score of 92.94% for the myocardium and 92.36% for the infarction. These outcomes highlight the competitiveness of our approach.Conclusion: In the case of the infarction class, our proposed method outperforms state-of-the-art techniques across all metrics evaluated in the challenge, establishing its superior performance in infarction segmentation. This study further reinforces the importance of integrating a contour loss into the segmentation process.

Endomyocardial substrate of ventricular arrhythmias in patients with autoimmune rheumatic diseases.

Delayed enhancement-magnetic resonance imaging (DE-MRI) has demonstrated that nonischemic cardiomyopathy is mainly characterized by intramural or epicardial fibrosis whereas global endomyocardial fibrosis suggests cardiac involvement in autoimmune rheumatic diseases or amyloidosis. Conduction disorders and sudden cardiac death are important manifestations of autoimmune rheumatic diseases with cardiac involvement but the substrates of ventricular arrhythmias in autoimmune rheumatic diseases have not been fully elucidated. 20 patients with autoimmune rheumatic diseases presenting with ventricular tachycardia (VT) (n = 11) or frequent ventricular extrasystoles (n = 9) underwent DE-MRI and/or endocardial electroanatomical mapping of the left ventricle (LV). Ten patients with autoimmune rheumatic diseases underwent VT ablation. Global endomyocardial fibrosis without myocardial thickening and unrelated to coronary territories was detected by DE-MRI or electroanatomical voltage mapping in 9 of 20 patients with autoimmune rheumatic diseases. In the other patients with autoimmune rheumatic diseases, limited regions of predominantly epicardial (n = 4) and intramyocardial (n = 5) fibrosis or only minimal fibrosis (n = 2) were found using DE-MRI. Endocardial low-amplitude diastolic potentials and pre-systolic Purkinje or fascicular potentials, mostly within fibrotic areas, were identified as the targets of successful VT ablation in 7 of 10 patients with autoimmune rheumatic diseases. Global endomyocardial fibrosis can be a tool to diagnose severe cardiac involvement in autoimmune rheumatic diseases and may serve as the substrate of ventricular arrhythmias in a substantial part of patients.

Pixel-wise statistical analysis of myocardial injury in STEMI patients with delayed enhancement MRI.

Myocardial injury assessment from delayed enhancement magnetic resonance images is routinely limited to global descriptors such as size and transmurality. Statistical tools from computational anatomy can drastically improve this characterization, and refine the assessment of therapeutic procedures aiming at infarct size reduction. Based on these techniques, we propose a new characterization of myocardial injury up to the pixel resolution. We demonstrate it on the imaging data from the Minimalist Immediate Mechanical Intervention randomized clinical trial (MIMI: NCT01360242), which aimed at comparing immediate and delayed stenting in acute ST-Elevation Myocardial Infarction (STEMI) patients. We analyzed 123 patients from the MIMI trial (62 ± 12 years, 98 male, 65 immediate 58 delayed stenting). Early and late enhancement images were transported onto a common geometry using techniques inspired by statistical atlases, allowing pixel-wise comparisons across population subgroups. A practical visualization of lesion patterns against specific clinical and therapeutic characteristics was also proposed using state-of-the-art dimensionality reduction. Infarct patterns were roughly comparable between the two treatments across the whole myocardium. Subtle but significant local differences were observed for the LCX and RCA territories with higher transmurality for delayed stenting at lateral and inferior/inferoseptal locations, respectively (15% and 23% of myocardial locations with a p-value <0.05, mainly in these regions). In contrast, global measurements were comparable for all territories (no statistically significant differences for all-except-one measurements before standardization / for all after standardization), although immediate stenting resulted in more subjects without reperfusion injury. Our approach substantially empowers the analysis of lesion patterns with standardized comparisons up to the pixel resolution, and may reveal subtle differences not accessible with global observations. On the MIMI trial data as illustrative case, it confirmed its general conclusions regarding the lack of benefit of delayed stenting, but revealed subgroups differences thanks to the standardized and finer analysis scale.

Cardiac Troponin I in Patients Undergoing Percutaneous and Surgical Myocardial Revascularization: Comparison of Analytical Methods

The myocardial infarction (MI) types 4a and 5 guidelines recommend cardiac troponin (cTn) diagnostic decision limits of 5 and 10 times the 99th percentile, respectively. Different cTn kits elicit different responses, so the MI diagnosis is still challenging. The study aimed to establish the cutoff values and the accuracy of three different cTnI kits in the diagnosis of post-procedural MI. We analyzed 115 patients with multivessel stable chronic coronary artery disease; 26 underwent percutaneous coronary intervention, and 89 underwent coronary artery bypass graft. Delayed-enhancement magnetic resonance imaging was performed before and after each intervention for definitive MI diagnoses. Two contemporary and one high-sensitivity cTnI immunoassays were used. ROC curves determined the accuracy of each assay. Low accuracy was observed after applying the current guidelines recommendations. The three cTnI assays accuracies improved when adjusted by the new ROC cutoffs, reaching 82% for MI type 5 for all assays, and 78%, 88%, and 87% for MI type 4 for Siemens, Beckman, and Abbott, respectively. The ultrasensitive and contemporary tests' accuracy for MI types 4a and 5 diagnoses are equivalent when adjusted for these new cutoffs. The hs-cTnI assays had lower accuracy than contemporary tests for MI types 4a and 5 diagnoses.

Differences in Post Ablation Cardiac MRI Scar Between Radiofrequency and Cryoballoon Ablation: A DECAAF II Sub-analysis.

Pulmonary vein isolation (PVI) using radiofrequency (RF) and cryoballoon (Cryo) ablation are standard approaches for rhythm control in patients with symptomatic atrial fibrillation. Both strategies create scar in the left atrium (LA). There have been few studies investigating the difference in scar formation between patients undergoing RF and Cryo using cardiac magnetic resonance imaging (CMR). The current study is a sub-analysis of the control arm of the Delayed-Enhancement MRI Determinant of Successful Catheter Ablation of Atrial Fibrillation study (DECAAF II). The study was a multicenter, randomized, controlled, single blinded trial that evaluated atrial arrhythmia recurrence (AAR) between PVI alone and PVI plus CMR atrial fibrosis guided ablation. Pre-ablation CMR and 3-6-month post ablation CMR were obtained to assess baseline LA fibrosis and scar formation respectively. Of the 843 patients randomized in the DECAAF II trial, we analyzed the 408 patients in the primary analysis control arm that received standard PVI. Five patients received combined RF and Cryo ablations so were excluded from this sub-analysis. Of the 403 patients analyzed, 345 underwent RF and 58 Cryo. The average procedure duration was 146 minutes for RF and 103 minutes for Cryo (p = 0.001). The rate of AAR at ~15 months occurred in 151 (43.8%) patients in the RF group and 28 (48.3%) patients in the Cryo group (p = 0.62). On 3-month post CMR the RF arm had significantly more scar (8.8% vs. 6.4%, p = 0.001) compared to Cryo. Patients with ≥ 6.5% LA scar (p < 0.001) and ≥ 2.3% LA scar around the PV antra (p = 0.01) on 3-month post CMR had less AAR independent of ablation technique. Cryo caused a greater percentage of right and left pulmonary vein (PV) antral scar (p = 0.04, p = 0.02) and less non-PV antral scar (p = 0.009) compared to RF. On Cox regression Cryo patients free of AAR had a greater percentage of left PV antral scar (p = 0.01) and less non-PV antral scar (p = 0.004) compared to RF free of AAR. In this sub-analysis of the control arm of the DECAAF II trial, we observed that Cryo formed a greater percentage of PV antral scar and less non-PV antral scar compared to RF. Post ablation LA scar ≥ 6.5% predicted freedom from AAR, independent of ablation technique. These findings may have prognostic implications in ablation technique selection and freedom from AAR. This article is protected by copyright. All rights reserved.

Abstract 15228: Manganese-Enhanced MRI (MEMRI) Determines the Mechanistic Efficacy of Exosome Therapy in a Porcine Myocardial Infarction Model

Introduction: Transplantation of exosomes from mesenchymal stem cells (MSCs) and induced pluripotent stem cell derived cardiomyocytes (iCMs) improves heart function after myocardial infarction (MI). However, imaging methods have not been established to assess the mechanism of these treatments. Contrast imaging with gadolinium delayed-enhancement MRI (DEMRI) allows non-specific quantification of myocardial fibrosis. In contrast, MEMRI demonstrates active uptake of manganese into intracellular space by the viable cardiomyocytes via L-type calcium channel. MEMRI delineates the level of cellular metabolism and identifies the recovery of the injured but viable cardiomyocytes in the peri-infarct region (PIR). Hypothesis: MEMRI reflects the mechanistic efficacy of cell therapy. Methods: Yorkshire swine were subject to percutaneous proximal left anterior descending artery occlusion for 1 hour. Ten 500ul injections containing 5 x 10 11 exosomes isolated from MSCs and iCMs were delivered transendocardially to the PIR. Four weeks after injury, left ventricular ejection fraction (LVEF), and myocardial scar/viability were measured. DEMRI and MEMRI delineated the PIR to quantify the effects of exosomes. Results: The LVEF of PBS (control), MSC-exosome, and iCM-exosome groups was 18.66% ± 4.23, 27.46% ± 6.16, and 29.5% ± 3.96*, respectively (*p&lt;0.05). The % contrast enhancement by DEMRI over total LV volume were 29.38% ± 4.86, 28.76% ± 5.02, and 29.58% ± 5.55, respectively. Meanwhile, the % contrast nulling (CN, non-viable) by MEMRI were 20.51% ± 3.86, 9.53% ± 1.99, and 7.31% ± 2.62*, respectively (*p&lt;0.01). mRNA seq of the porcine myocardium and subsequent gene ontology and KEGG analyses demonstrated significant differential gene expression of iCM-exosome group related to calcium regulation (adenylate cyclase 7, 3.30-fold change) and sarcoplasmic reticulum calcium ion transport (chemokine receptor 5, 3.15-fold change). Conclusions: The % CN reduction of MEMRI reflects improved cardiomyocyte metabolism and calcium handing by exosome therapy. The underlying mechanism and functional improvement are identified by MEMRI.

Abstract 15179: Histology-Validated Delayed Enhancement Magnetic Resonance Imaging of Atrial Fibrillation Driver Substrate

Introduction: Persistent (per) AF may be maintained by localized extra-PV drivers (DRs). Regional fibrosis can serve as a substrate for AF DRs, but its visualization by delayed enhancement (DE)-MRI lacks proper validation for driver substrates. The objective of this study is to evaluate if histology-validated normalized (HVN) DE-MRI thresholds can define arrhythmogenic fibrotic content in multielectrode mapping (MEM)-defined DRs from perAF patients’ (Pts) DE-MRI. Methods: PerAF Pts (n=14, 71% male; 63±10 y/o) had DE-MRI at 3T (0.63x0.63x0.63mm 3 , 0.2mmol/kg Gd) before their perAF ablation procedure. Chamber-specific DE-MRI intensity was normalized to the healthy atria (mean lowest 5% intensity, 0 AU) and mitral (LA) or tricuspid (RA) valve (100 AU). 3 control and 5 perAF canines had DE-MRI and regional histology to validate DE-MRI fibrotic thresholds for the LA and RA, which were defined as standard deviations (SD) above 40% of the mean healthy atrial tissue intensity (LA: 3.7 SD, RA: 2.7 SD). Fibrosis-enhancement of segmented MEM-defined DRs (LA: n=25, RA: n=11) were analyzed using conventional, non-normalized 3 SD and HVN thresholds and compared to 5 random regions (non-DRs) within the same anatomical region(s). Results: Only our HVN approach could reveal arrhythmogenic fibrotic substrates in LA DR regions, but not the conventional approach. Both approaches could not provide significant differences for RA DR vs non-DR fibrosis regions. The cut-off of 33.6% of LA DE-MRI fibrotic enhancement using the HVN approach allowed to distinguish LA DR vs. non-DR regions with 56% sensitivity and 70% specificity. At 12 months follow-up, 10 Pts remained free of AF. Conclusion: MEM-defined DRs within fibrotic regions, identified by HVN DE-MRI, may represent efficient targets for ablation of LA arrhythmogenic perAF substrates. Integration of MEM and MRI may lead to improved targeted ablation and long-term AF freedom.

Abstract 14457: Cardiac MRI Results Following RFA and Cryoballoon Ablation: DECAAF II Subanalysis

Introduction: Radiofrequency (RFA) and cryoballon ablation are standard approaches for rhythm control of atrial fibrillation. Both strategies involve scar formation of the left atrial tissue. To date there have been no studies using post ablation cardiac magnetic resonance imaging (MRI) to assess residual fibrosis and scar created by standard radiofrequency and cryoballoon ablation. Methods: The current study is a sub analysis of the control group of the Delayed-Enhancement MRI Determinant of Successful Catheter Ablation of Atrial Fibrillation trial (DECAAF II). This was a multicenter, randomized, controlled, single blindedtrial that evaluated atrial arrhythmia recurrence (AAR) between PVI alone and PVI plus fibrosis guided ablation. Baseline MRI and 3-month post ablation MRI were obtained to assess lesion and scar formation. Results: Of the 843 patients randomized in the trial, 428 patients were in the control arm that received standard PVI alone. Of those 428 patients, 367 patients (85.7%) underwent RFA and 61 (14.3%) underwent cryoballoon ablation. Average procedure duration was 147 minutes for RFA and 107 minutes for cryoballoon (p = 0.001). The rate of AAR at 451 days was 42.0% in the RFA group and 45.9% in the cryoballoon group (p = 0.66). At the 3-month post MRI the RFA arm had significantly more left atrial scar formation compared to cryoballoon arm (8.8% vs. 6.4%, p = 0.001). Also, the RFA arm had more covered fibrosis (3.6% vs. 2.9%, p = 0.02) and less non-covered fibrosis (14.9% vs. 17.2%, p = 0.05) compared to the cryoballoon arm. The amount of scar formation did not predict AAR (p = 0.07). Conclusions: In this subanalysis of the DECAAF II trial, there was no significant difference in the rate of AAR between RFA and cryoballoon ablation. However, RFA caused significantly more left atrial scar compared to cryoballoon ablation on 3-month post MRI.

Auto-MyIn: Automatic diagnosis of myocardial infarction via multiple GLCMs, CNNs, and SVMs

This paper proposes an automated diagnostic tool namely, Auto-MyIn, for diagnosing myocardial infarction (MI) using multiple convolutional neural networks (CNN). Rather than utilizing the spatial information of the original delayed-enhancement magnetic resonance (DE-MRI) images, Auto-MyIn uses the textural information obtained by applying grey level co-occurrence matrix (GLCM) of four different grey levels to train the three CNNs (ResNet-18, DarkNet-19, and SqueezeNet). First, the images generated from each GLCM grey level are used to train each CNN individually. Next, for each GLCM grey level, the textural-based deep features extracted from the three CNNs are concatenated and used to train several support vector machine (SVM) classifiers. Finally, Auto-MyIn fuses textural-based deep features of the four GLCM grey levels using principal component analysis (PCA). The results of Auto-MyIn indicated that fusing the textural-based deep features of each level of GLCM is better than the end-to-end deep learning classification of the three CNNs trained with each grey level of GLCM images. Furthermore, it showed that fusing textural-based deep features of the four grey levels of the GLCM using PCA has further improvement on diagnostic accuracy. Moreover, the results prove that using textural information is superior to using spatial information of the original DE-MRI images. In addition, the results of Auto-MyIn when compared with other related studies demonstrated its competitive ability. Moreover, the performance of Auto-MyIn shows an accuracy of 0.984, a sensitivity of 0.992, specificity of 0.968, and precision of 0.967, which indicate that it is a reliable tool, therefore it could be employed to help in the clinical decision making and facilitate the diagnostic process of MI thus avoiding the limitations of manual diagnosis.

Classification of myocardial fibrosis in DE-MRI based on semi-supervised semantic segmentation and dual attention mechanism

ObjectiveIt is essential to utilize cardiac delayed-enhanced magnetic resonance imaging (DE-MRI) to diagnose cardiovascular disease. By segmenting myocardium DE-MRI images, it provides critical information for the evaluation and treatment of myocardial infarction. As a consequence, it is vital to investigate the segmentation and classification technique of myocardial DE-MRI. MethodsFirstly, an end-to-end minimally supervised and semi-supervised semantic DE-MRI myocardial fibrosis segmentation framework is proposed, which combines image classification and semantic segmentation branches based on the self-attention mechanism. Following that, a residual hole network fused with the dual attention mechanism was built, and a double attention metabolic pathway classification method for cardiac fibrosis in DE-MRI images was developed. ResultsBy adding pixel-level labels to an extra 40 training images, the segmentation model may enhance semantic segmentation performance by 2.6 percent (from 61.2 percent to 63.8 percent). When the number of pixel-level labels is increased to 80, semi-supervised feature extraction increases by 4.7 percent when compared to weakly guided semantic segmentation. Adding an attention mechanism to the critical network DRN (Deep Residual Network) can increase the classifier's performance by a small amount. Experiments revealed that the models worked effectively. ConclusionThis paper investigates the segmentation and classification of cardiac fibrosis in DE-MRI data using a semi-supervised semantic segmentation and dual attention mechanism, dealing with the issue that existing segmentation algorithms have difficulty segmenting myocardial fibrosis tissue. In the future, we can consider optimizing the design of the attention module to reduce the module computation.

Effect of MRI-Guided Fibrosis Ablation vs Conventional Catheter Ablation on Atrial Arrhythmia Recurrence in Patients With Persistent Atrial Fibrillation

Ablation of persistent atrial fibrillation (AF) remains a challenge. Left atrial fibrosis plays an important role in the pathophysiology of AF and has been associated with poor procedural outcomes. To investigate the efficacy and adverse events of targeting atrial fibrosis detected on magnetic resonance imaging (MRI) in reducing atrial arrhythmia recurrence in persistent AF. The Efficacy of Delayed Enhancement-MRI-Guided Fibrosis Ablation vs Conventional Catheter Ablation of Atrial Fibrillation trial was an investigator-initiated, multicenter, randomized clinical trial involving 44 academic and nonacademic centers in 10 countries. A total of 843 patients with symptomatic or asymptomatic persistent AF and undergoing AF ablation were enrolled from July 2016 to January 2020, with follow-up through February 19, 2021. Patients with persistent AF were randomly assigned to pulmonary vein isolation (PVI) plus MRI-guided atrial fibrosis ablation (421 patients) or PVI alone (422 patients). Delayed-enhancement MRI was performed in both groups before the ablation procedure to assess baseline atrial fibrosis and at 3 months postablation to assess for ablation scar. The primary end point was time to first atrial arrhythmia recurrence after a 90-day blanking period postablation. The primary safety composite outcome was defined by the occurrence of 1 or more of the following events within 30 days postablation: stroke, PV stenosis, bleeding, heart failure, or death. Among 843 patients who were randomized (mean age 62.7 years; 178 [21.1%] women), 815 (96.9%) completed the 90-day blanking period and contributed to the efficacy analyses. There was no significant difference in atrial arrhythmia recurrence between groups (fibrosis-guided ablation plus PVI patients, 175 [43.0%] vs PVI-only patients, 188 [46.1%]; hazard ratio [HR], 0.95 [95% CI, 0.77-1.17]; P = .63). Patients in the fibrosis-guided ablation plus PVI group experienced a higher rate of safety outcomes (9 [2.2%] vs 0 in PVI group; P = .001). Six patients (1.5%) in the fibrosis-guided ablation plus PVI group had an ischemic stroke compared with none in PVI-only group. Two deaths occurred in the fibrosis-guided ablation plus PVI group, and the first one was possibly related to the procedure. Among patients with persistent AF, MRI-guided fibrosis ablation plus PVI, compared with PVI catheter ablation only, resulted in no significant difference in atrial arrhythmia recurrence. Findings do not support the use of MRI-guided fibrosis ablation for the treatment of persistent AF. ClinicalTrials.gov Identifier: NCT02529319.

Prediction of Myocardial Infarction From Patient Features With Machine Learning.

This study proposes machine learning-based models to automatically evaluate the severity of myocardial infarction (MI) from physiological, clinical, and paraclinical features. Two types of machine learning models are investigated for the MI assessment: the classification models classify the presence of the infarct and the persistent microvascular obstruction (PMO), and the regression models quantify the Percentage of Infarcted Myocardium (PIM) of patients suspected of having an acute MI during their reception in the emergency department. The ground truth labels for these supervised models are derived from the corresponding Delayed Enhancement MRI (DE-MRI) exams and manual annotations of the myocardium and scar tissues. Experiments were conducted on 150 cases and evaluated with cross-validation. Results showed that for the MI (PMO inclusive) and the PMO (infarct exclusive), the best models obtained respectively a mean error of 0.056 and 0.012 for the quantification, and 88.67 and 77.33% for the classification accuracy of the state of the myocardium. The study of the features' importance also revealed that the troponin value had the strongest correlation to the severity of the MI among the 12 selected features. For the proposal's translational perspective, in cardiac emergencies, qualitative and quantitative analysis can be obtained prior to the achievement of MRI by relying only on conventional tests and patient features, thus, providing an objective reference for further treatment by physicians.

An Improved 3D Deep Learning-Based Segmentation of Left Ventricular Myocardial Diseases from Delayed-Enhancement MRI with Inclusion and Classification Prior Information U-Net (ICPIU-Net).

Accurate segmentation of the myocardial scar may supply relevant advancements in predicting and controlling deadly ventricular arrhythmias in subjects with cardiovascular disease. In this paper, we propose the architecture of inclusion and classification of prior information U-Net (ICPIU-Net) to efficiently segment the left ventricle (LV) myocardium, myocardial infarction (MI), and microvascular-obstructed (MVO) tissues from late gadolinium enhancement magnetic resonance (LGE-MR) images. Our approach was developed using two subnets cascaded to first segment the LV cavity and myocardium. Then, we used inclusion and classification constraint networks to improve the resulting segmentation of the diseased regions within the pre-segmented LV myocardium. This network incorporates the inclusion and classification information of the LGE-MRI to maintain topological constraints of pathological areas. In the testing stage, the outputs of each segmentation network obtained with specific estimated parameters from training were fused using the majority voting technique for the final label prediction of each voxel in the LGE-MR image. The proposed method was validated by comparing its results to manual drawings by experts from 50 LGE-MR images. Importantly, compared to various deep learning-based methods participating in the EMIDEC challenge, the results of our approach have a more significant agreement with manual contouring in segmenting myocardial diseases.

Automatic Left Ventricle Segmentation from Short-Axis Cardiac MRI Images Based on Fully Convolutional Neural Network.

Background: Left ventricle (LV) segmentation using a cardiac magnetic resonance imaging (MRI) dataset is critical for evaluating global and regional cardiac functions and diagnosing cardiovascular diseases. LV clinical metrics such as LV volume, LV mass and ejection fraction (EF) are frequently extracted based on the LV segmentation from short-axis MRI images. Manual segmentation to assess such functions is tedious and time-consuming for medical experts to diagnose cardiac pathologies. Therefore, a fully automated LV segmentation technique is required to assist medical experts in working more efficiently. Method: This paper proposes a fully convolutional network (FCN) architecture for automatic LV segmentation from short-axis MRI images. Several experiments were conducted in the training phase to compare the performance of the network and the U-Net model with various hyper-parameters, including optimization algorithms, epochs, learning rate, and mini-batch size. In addition, a class weighting method was introduced to avoid having a high imbalance of pixels in the classes of image’s labels since the number of background pixels was significantly higher than the number of LV and myocardium pixels. Furthermore, effective image conversion with pixel normalization was applied to obtain exact features representing target organs (LV and myocardium). The segmentation models were trained and tested on a public dataset, namely the evaluation of myocardial infarction from the delayed-enhancement cardiac MRI (EMIDEC) dataset. Results: The dice metric, Jaccard index, sensitivity, and specificity were used to evaluate the network’s performance, with values of 0.93, 0.87, 0.98, and 0.94, respectively. Based on the experimental results, the proposed network outperforms the standard U-Net model and is an advanced fully automated method in terms of segmentation performance. Conclusion: This proposed method is applicable in clinical practice for doctors to diagnose cardiac diseases from short-axis MRI images.