Automated Myocardial T2 and Extracellular Volume Quantification in Cardiac MRI Using Transfer Learning-based Myocardium Segmentation

Advanced Myocardial MRI Tissue Characterization Combining Contrast Agent-Free T1-Rho Mapping With Fully Automated Analysis

BACKGROUND

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.

PURPOSE

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

STUDY TYPE

Retrospective.

POPULATION

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

FIELD STRENGTH/SEQUENCE

Single-shot bSSFP T1ρ mapping sequence at 1.5 T.

ASSESSMENT

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.

STATISTICAL TESTS

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.

RESULTS

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).

DATA CONCLUSION

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

EVIDENCE LEVEL

3 TECHNICAL EFFICACY: Stage 1.

Role of Artificial Intelligence in Congenital Heart Disease and Interventions

Artificial intelligence has promising impact on patients with congenital heart disease, a vulnerable population with life-long health care needs and, often, a substantially higher risk of death than the general population. This review explores the role artificial intelligence has had on cardiac imaging, electrophysiology, interventional procedures, and intensive care monitoring as it relates to children and adults with congenital heart disease. Machine learning and deep learning algorithms have enhanced not only imaging segmentation and processing but also diagnostic accuracy namely reducing interobserver variability. This has a meaningful impact in complex congenital heart disease improving anatomic diagnosis, assessment of cardiac function, and predicting long-term outcomes. Image processing has benefited procedural planning for interventional cardiology, allowing for a higher quality and density of information to be extracted from the same imaging modalities. In electrophysiology, deep learning models have enhanced the diagnostic potential of electrocardiograms, detecting subtle yet meaningful variation in signals that enable early diagnosis of cardiac dysfunction, risk stratification of mortality, and more accurate diagnosis and prediction of arrhythmias. In the congenital heart disease population, this has the potential for meaningful prolongation of life. Postoperative care in the cardiac intensive care unit is a data-rich environment that is often overwhelming. Detection of subtle data trends in this environment for early detection of morbidity is a ripe avenue for artificial intelligence algorithms to be used. Examples like early detection of catheter-induced thrombosis have already been published. Despite their great promise, artificial intelligence algorithms are still limited by hurdles such as data standardization, algorithm validation, drift, and explainability.

Breaking barriers: noninvasive AI model for BRAFV600E mutation identification

OBJECTIVE

BRAFV600E is the most common mutation found in thyroid cancer and is particularly associated with papillary thyroid carcinoma (PTC). Currently, genetic mutation detection relies on invasive procedures. This study aimed to extract radiomic features and utilize deep transfer learning (DTL) from ultrasound images to develop a noninvasive artificial intelligence model for identifying BRAFV600E mutations.

MATERIALS AND METHODS

Regions of interest (ROI) were manually annotated in the ultrasound images, and radiomic and DTL features were extracted. These were used in a joint DTL-radiomics (DTLR) model. Fourteen DTL models were employed, and feature selection was performed using the LASSO regression. Eight machine learning methods were used to construct predictive models. Model performance was primarily evaluated using area under the curve (AUC), accuracy, sensitivity and specificity. The interpretability of the model was visualized using gradient-weighted class activation maps (Grad-CAM).

RESULTS

Sole reliance on radiomics for identification of BRAFV600E mutations had limited capability, but the optimal DTLR model, combined with ResNet152, effectively identified BRAFV600E mutations. In the validation set, the AUC, accuracy, sensitivity and specificity were 0.833, 80.6%, 76.2% and 81.7%, respectively. The AUC of the DTLR model was higher than that of the DTL and radiomics models. Visualization using the ResNet152-based DTLR model revealed its ability to capture and learn ultrasound image features related to BRAFV600E mutations.

CONCLUSION

The ResNet152-based DTLR model demonstrated significant value in identifying BRAFV600E mutations in patients with PTC using ultrasound images. Grad-CAM has the potential to objectively stratify BRAF mutations visually. The findings of this study require further collaboration among more centers and the inclusion of additional data for validation.

Present and Future Innovations in AI and Cardiac MRI

Cardiac MRI is used to diagnose and treat patients with a multitude of cardiovascular diseases. Despite the growth of clinical cardiac MRI, complicated image prescriptions and long acquisition protocols limit the specialty and restrain its impact on the practice of medicine. Artificial intelligence (AI)-the ability to mimic human intelligence in learning and performing tasks-will impact nearly all aspects of MRI. Deep learning (DL) primarily uses an artificial neural network to learn a specific task from example data sets. Self-driving scanners are increasingly available, where AI automatically controls cardiac image prescriptions. These scanners offer faster image collection with higher spatial and temporal resolution, eliminating the need for cardiac triggering or breath holding. In the future, fully automated inline image analysis will most likely provide all contour drawings and initial measurements to the reader. Advanced analysis using radiomic or DL features may provide new insights and information not typically extracted in the current analysis workflow. AI may further help integrate these features with clinical, genetic, wearable-device, and "omics" data to improve patient outcomes. This article presents an overview of AI and its application in cardiac MRI, including in image acquisition, reconstruction, and processing, and opportunities for more personalized cardiovascular care through extraction of novel imaging markers.

DeepFittingNet: A deep neural network-based approach for simplifying cardiac T1 and T2 estimation with improved robustness

PURPOSE

To develop and evaluate a deep neural network (DeepFittingNet) for T1 /T2 estimation of the most commonly used cardiovascular MR mapping sequences to simplify data processing and improve robustness.

THEORY AND METHODS

DeepFittingNet is a 1D neural network composed of a recurrent neural network (RNN) and a fully connected (FCNN) neural network, in which RNN adapts to the different number of input signals from various sequences and FCNN subsequently predicts A, B, and Tx of a three-parameter model. DeepFittingNet was trained using Bloch-equation simulations of MOLLI and saturation-recovery single-shot acquisition (SASHA) T1 mapping sequences, and T2 -prepared balanced SSFP (T2 -prep bSSFP) T2 mapping sequence, with reference values from the curve-fitting method. Several imaging confounders were simulated to improve robustness. The trained DeepFittingNet was tested using phantom and in-vivo signals, and compared to the curve-fitting algorithm.

RESULTS

In testing, DeepFittingNet performed T1 /T2 estimation of four sequences with improved robustness in inversion-recovery T1 estimation. The mean bias in phantom T1 and T2 between the curve-fitting and DeepFittingNet was smaller than 30 and 1 ms, respectively. Excellent agreements between both methods was found in the left ventricle and septum T1 /T2 with a mean bias <6 ms. There was no significant difference in the SD of both the left ventricle and septum T1 /T2 between the two methods.

CONCLUSION

DeepFittingNet trained with simulations of MOLLI, SASHA, and T2 -prep bSSFP performed T1 /T2 estimation tasks for all these most used sequences. Compared with the curve-fitting algorithm, DeepFittingNet improved the robustness for inversion-recovery T1 estimation and had comparable performance in terms of accuracy and precision.

Validation of a deep learning-based software for automated analysis of T2 mapping in cardiac magnetic resonance imaging

Background

The reliability and diagnostic performance of deep learning (DL)-based automated T2 measurements on T2 map of 3.0-T cardiac magnetic resonance imaging (MRI) using multi-institutional datasets have not been investigated. We aimed to evaluate the performance of a DL-based software for measuring automated T2 values from 3.0-T cardiac MRI obtained at two centers.

Methods

Eighty-three subjects were retrospectively enrolled from two centers (42 healthy subjects and 41 patients with myocarditis) to validate a commercial DL-based software that was trained to segment the left ventricular myocardium and measure T2 values on T2 mapping sequences. Manual reference T2 values by two experienced radiologists and those calculated by the DL-based software were obtained. The segmentation performance of the DL-based software and the non-inferiority of automated T2 values were assessed compared with the manual reference standard per segment level. The software's performance in detecting elevated T2 values was assessed by calculating the sensitivity, specificity, and accuracy per segment.

Results

The average Dice similarity coefficient for segmentation of myocardium on T2 maps was 0.844. The automated T2 values were non-inferior to the manual reference T2 values on a per-segment analysis (45.35 vs. 44.32 ms). The DL-based software exhibited good performance (sensitivity: 83.6-92.8%; specificity: 82.5-92.0%; accuracy: 82.7-92.2%) in detecting elevated T2 values.

Conclusions

The DL-based software for automated T2 map analysis yields non-inferior measurements at the per-segment level and good performance for detecting myocardial segments with elevated T2 values compared with manual analysis.

Artificial Intelligence to Reduce or Eliminate the Need for Gadolinium-Based Contrast Agents in Brain and Cardiac MRI: A Literature Review

ABSTRACT

Brain and cardiac MRIs are fundamental noninvasive imaging tools, which can provide important clinical information and can be performed without or with gadolinium-based contrast agents (GBCAs), depending on the clinical indication. It is currently a topic of debate whether it would be feasible to extract information such as standard gadolinium-enhanced MRI while injecting either less or no GBCAs. Artificial intelligence (AI) is a great source of innovation in medical imaging and has been explored as a method to synthesize virtual contrast MR images, potentially yielding similar diagnostic performance without the need to administer GBCAs. If possible, there would be significant benefits, including reduction of costs, acquisition time, and environmental impact with respect to conventional contrast-enhanced MRI examinations. Given its promise, we believe additional research is needed to increase the evidence to make these AI solutions feasible, reliable, and robust enough to be integrated into the clinical framework. Here, we review recent AI studies aimed at reducing or replacing gadolinium in brain and cardiac imaging while maintaining diagnostic image quality.

Magnetic resonance myocardial T1ρ mapping : Technical overview, challenges, emerging developments, and clinical applications

The potential of cardiac magnetic resonance to improve cardiovascular care and patient management is considerable. Myocardial T1-rho (T1ρ) mapping, in particular, has emerged as a promising biomarker for quantifying myocardial injuries without exogenous contrast agents. Its potential as a contrast-agent-free ("needle-free") and cost-effective diagnostic marker promises high impact both in terms of clinical outcomes and patient comfort. However, myocardial T1ρ mapping is still at a nascent stage of development and the evidence supporting its diagnostic performance and clinical effectiveness is scant, though likely to change with technological improvements. The present review aims at providing a primer on the essentials of myocardial T1ρ mapping, and to describe the current range of clinical applications of the technique to detect and quantify myocardial injuries. We also delineate the important limitations and challenges for clinical deployment, including the urgent need for standardization, the evaluation of bias, and the critical importance of clinical testing. We conclude by outlining technical developments to be expected in the future. If needle-free myocardial T1ρ mapping is shown to improve patient diagnosis and prognosis, and can be effectively integrated in cardiovascular practice, it will fulfill its potential as an essential component of a cardiac magnetic resonance examination.

Artificial Intelligence as a Diagnostic Tool in Non-Invasive Imaging in the Assessment of Coronary Artery Disease

Coronary artery disease (CAD) remains a leading cause of mortality and morbidity worldwide, and it is associated with considerable economic burden. In an ageing, multimorbid population, it has become increasingly important to develop reliable, consistent, low-risk, non-invasive means of diagnosing CAD. The evolution of multiple cardiac modalities in this field has addressed this dilemma to a large extent, not only in providing information regarding anatomical disease, as is the case with coronary computed tomography angiography (CCTA), but also in contributing critical details about functional assessment, for instance, using stress cardiac magnetic resonance (S-CMR). The field of artificial intelligence (AI) is developing at an astounding pace, especially in healthcare. In healthcare, key milestones have been achieved using AI and machine learning (ML) in various clinical settings, from smartwatches detecting arrhythmias to retinal image analysis and skin cancer prediction. In recent times, we have seen an emerging interest in developing AI-based technology in the field of cardiovascular imaging, as it is felt that ML methods have potential to overcome some limitations of current risk models by applying computer algorithms to large databases with multidimensional variables, thus enabling the inclusion of complex relationships to predict outcomes. In this paper, we review the current literature on the various applications of AI in the assessment of CAD, with a focus on multimodality imaging, followed by a discussion on future perspectives and critical challenges that this field is likely to encounter as it continues to evolve in cardiology.

Automated Measurement of Native T1 and Extracellular Volume Fraction in Cardiac Magnetic Resonance Imaging Using a Commercially Available Deep Learning Algorithm

OBJECTIVE

T1 mapping provides valuable information regarding cardiomyopathies. Manual drawing is time consuming and prone to subjective errors. Therefore, this study aimed to test a DL algorithm for the automated measurement of native T1 and extracellular volume (ECV) fractions in cardiac magnetic resonance (CMR) imaging with a temporally separated dataset.

MATERIALS AND METHODS

CMR images obtained for 95 participants (mean age ± standard deviation, 54.5 ± 15.2 years), including 36 left ventricular hypertrophy (12 hypertrophic cardiomyopathy, 12 Fabry disease, and 12 amyloidosis), 32 dilated cardiomyopathy, and 27 healthy volunteers, were included. A commercial deep learning (DL) algorithm based on 2D U-net (Myomics-T1 software, version 1.0.0) was used for the automated analysis of T1 maps. Four radiologists, as study readers, performed manual analysis. The reference standard was the consensus result of the manual analysis by two additional expert readers. The segmentation performance of the DL algorithm and the correlation and agreement between the automated measurement and the reference standard were assessed. Interobserver agreement among the four radiologists was analyzed.

RESULTS

DL successfully segmented the myocardium in 99.3% of slices in the native T1 map and 89.8% of slices in the post-T1 map with Dice similarity coefficients of 0.86 ± 0.05 and 0.74 ± 0.17, respectively. Native T1 and ECV showed strong correlation and agreement between DL and the reference: for T1, r = 0.967 (95% confidence interval [CI], 0.951-0.978) and bias of 9.5 msec (95% limits of agreement [LOA], -23.6-42.6 msec); for ECV, r = 0.987 (95% CI, 0.980-0.991) and bias of 0.7% (95% LOA, -2.8%-4.2%) on per-subject basis. Agreements between DL and each of the four radiologists were excellent (intraclass correlation coefficient [ICC] of 0.98-0.99 for both native T1 and ECV), comparable to the pairwise agreement between the radiologists (ICC of 0.97-1.00 and 0.99-1.00 for native T1 and ECV, respectively).

CONCLUSION

The DL algorithm allowed automated T1 and ECV measurements comparable to those of radiologists.

A Deep Learning Segmentation Pipeline for Cardiac T1 Mapping Using MRI Relaxation-based Synthetic Contrast Augmentation

PURPOSE

To design and evaluate an automated deep learning method for segmentation and analysis of cardiac MRI T1 maps with use of synthetic T1-weighted images for MRI relaxation-based contrast augmentation.

MATERIALS AND METHODS

This retrospective study included MRI scans acquired between 2016 and 2019 from 100 patients (mean age ± SD, 55 years ± 13; 72 men) across various clinical abnormalities with use of a modified Look-Locker inversion recovery, or MOLLI, sequence to quantify native T1 (T1_native), postcontrast T1 (T1_post), and extracellular volume (ECV). Data were divided into training (n = 60) and internal (n = 40) test subsets. "Synthetic" T1-weighted images were generated from the T1 exponential inversion-recovery signal model at a range of optimal inversion times, yielding high blood-myocardium contrast, and were used for contrast-based image augmentation during training and testing of a convolutional neural network for myocardial segmentation. Automated segmentation, T1, and ECV were compared with experts with use of Dice similarity coefficients (DSCs), correlation coefficients, and Bland-Altman analysis. An external test dataset (n = 147) was used to assess model generalization.

RESULTS

Internal testing showed high myocardial DSC relative to experts (0.81 ± 0.08), which was similar to interobserver DSC (0.81 ± 0.08). Automated segmental measurements strongly correlated with experts (T1_native, R = 0.87; T1_post, R = 0.91; ECV, R = 0.92), which were similar to interobserver correlation (T1_native, R = 0.86; T1_post, R = 0.94; ECV, R = 0.95). External testing showed strong DSC (0.80 ± 0.09) and T1_native correlation (R = 0.88) between automatic and expert analysis.

CONCLUSION

This deep learning method leveraging synthetic contrast augmentation may provide accurate automated T1 and ECV analysis for cardiac MRI data acquired across different abnormalities, centers, scanners, and T1 sequences.Keywords: MRI, Cardiac, Tissue Characterization, Segmentation, Convolutional Neural Network, Deep Learning Algorithms, Machine Learning Algorithms, Supervised Learning Supplemental material is available for this article. © RSNA, 2022.

Impact of deep learning architectures on accelerated cardiac T1 mapping using MyoMapNet

The objective of the current study was to investigate the performance of various deep learning (DL) architectures for MyoMapNet, a DL model for T1 estimation using accelerated cardiac T1 mapping from four T1 -weighted images collected after a single inversion pulse (Look-Locker 4 [LL4]). We implemented and tested three DL architectures for MyoMapNet: (a) a fully connected neural network (FC), (b) convolutional neural networks (VGG19, ResNet50), and (c) encoder-decoder networks with skip connections (ResUNet, U-Net). Modified Look-Locker inversion recovery (MOLLI) images from 749 patients at 3 T were used for training, validation, and testing. The first four T1 -weighted images from MOLLI5(3)3 and/or MOLLI4(1)3(1)2 protocols were extracted to create accelerated cardiac T1 mapping data. We also prospectively collected data from 28 subjects using MOLLI and LL4 to further evaluate model performance. Despite rigorous training, conventional VGG19 and ResNet50 models failed to produce anatomically correct T1 maps, and T1 values had significant errors. While ResUNet yielded good quality maps, it significantly underestimated T1 . Both FC and U-Net, however, yielded excellent image quality with good T1 accuracy for both native (FC/U-Net/MOLLI = 1217 ± 64/1208 ± 61/1199 ± 61 ms, all p < 0.05) and postcontrast myocardial T1 (FC/U-Net/MOLLI = 578 ± 57/567 ± 54/574 ± 55 ms, all p < 0.05). In terms of precision, the U-Net model yielded better T1 precision compared with the FC architecture (standard deviation of 61 vs. 67 ms for the myocardium for native [p < 0.05], and 31 vs. 38 ms [p < 0.05], for postcontrast). Similar findings were observed in prospectively collected LL4 data. It was concluded that U-Net and FC DL models in MyoMapNet enable fast myocardial T1 mapping using only four T1 -weighted images collected from a single LL sequence with comparable accuracy. U-Net also provides a slight improvement in precision.

Optimized automated cardiac MR scar quantification with GAN-based data augmentation

BACKGROUND

The clinical utility of late gadolinium enhancement (LGE) cardiac MRI is limited by the lack of standardization, and time-consuming postprocessing. In this work, we tested the hypothesis that a cascaded deep learning pipeline trained with augmentation by synthetically generated data would improve model accuracy and robustness for automated scar quantification.

METHODS

A cascaded pipeline consisting of three consecutive neural networks is proposed, starting with a bounding box regression network to identify a region of interest around the left ventricular (LV) myocardium. Two further nnU-Net models are then used to segment the myocardium and, if present, scar. The models were trained on the data from the EMIDEC challenge, supplemented with an extensive synthetic dataset generated with a conditional GAN.

RESULTS

The cascaded pipeline significantly outperformed a single nnU-Net directly segmenting both the myocardium (mean Dice similarity coefficient (DSC) (standard deviation (SD)): 0.84 (0.09) vs 0.63 (0.20), p < 0.01) and scar (DSC: 0.72 (0.34) vs 0.46 (0.39), p < 0.01) on a per-slice level. The inclusion of the synthetic data as data augmentation during training improved the scar segmentation DSC by 0.06 (p < 0.01). The mean DSC per-subject on the challenge test set, for the cascaded pipeline augmented by synthetic generated data, was 0.86 (0.03) and 0.67 (0.29) for myocardium and scar, respectively.

CONCLUSION

A cascaded deep learning-based pipeline trained with augmentation by synthetically generated data leads to myocardium and scar segmentations that are similar to the manual operator, and outperforms direct segmentation without the synthetic images.

T2 and T2⁎ mapping and weighted imaging in cardiac MRI

Cardiac imaging is progressing from simple imaging of heart structure and function to techniques visualizing and measuring underlying tissue biological changes that can potentially define disease and therapeutic options. These techniques exploit underlying tissue magnetic relaxation times: T₁, T₂ and T₂*. Initial weighting methods showed myocardial heterogeneity, detecting regional disease. Current methods are now fully quantitative generating intuitive color maps that do not only expose regionality, but also diffuse changes - meaning that between-scan comparisons can be made to define disease (compared to normal) and to monitor interval change (compared to old scans). T₁ is now familiar and used clinically in multiple scenarios, yet some technical challenges remain. T₂ is elevated with increased tissue water - oedema. Should there also be blood troponin elevation, this oedema likely reflects inflammation, a key biological process. T₂* falls in the presence of magnetic/paramagnetic materials - practically, this means it measures tissue iron, either after myocardial hemorrhage or in myocardial iron overload. This review discusses how T₂ and T₂^⁎ imaging work (underlying physics, innovations, dependencies, performance), current and emerging use cases, quality assurance processes for global delivery and future research directions.

Radiomics and deep learning for myocardial scar screening in hypertrophic cardiomyopathy

BACKGROUND

Myocardial scar burden quantified using late gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR), has important prognostic value in hypertrophic cardiomyopathy (HCM). However, nearly 50% of HCM patients have no scar but undergo repeated gadolinium-based CMR over their life span. We sought to develop an artificial intelligence (AI)-based screening model using radiomics and deep learning (DL) features extracted from balanced steady state free precession (bSSFP) cine sequences to identify HCM patients without scar.

METHODS

We evaluated three AI-based screening models using bSSFP cine image features extracted by radiomics, DL, or combined DL-Radiomics. Images for 759 HCM patients (50 ± 16 years, 66% men) in a multi-center/vendor study were used to develop and test model performance. An external dataset of 100 HCM patients (53 ± 14 years, 70% men) was used to assess model generalizability. Model performance was evaluated using area-under-receiver-operating curve (AUC).

RESULTS

The DL-Radiomics model demonstrated higher AUC compared to DL and Radiomics in the internal (0.83 vs 0.77, p = 0.006 and 0.78, p = 0.05; n = 159) and external (0.74 vs 0.64, p = 0.006 and 0.71, p = 0.27; n = 100) datasets. The DL-Radiomics model correctly identified 43% and 28% of patients without scar in the internal and external datasets compared to 42% and 16% by Radiomics model and 42% and 23% by DL model, respectively.

CONCLUSIONS

A DL-Radiomics AI model using bSSFP cine images outperforms DL or Radiomics models alone as a scar screening tool prior to gadolinium administration. Despite its potential, the clinical utility of the model remains limited and further investigation is needed to improve the accuracy and generalizability.

The Road Toward Reproducibility of Parametric Mapping of the Heart: A Technical Review

Parametric mapping of the heart has become an essential part of many cardiovascular magnetic resonance imaging exams, and is used for tissue characterization and diagnosis in a broad range of cardiovascular diseases. These pulse sequences are used to quantify the myocardial T₁, T₂, T2*, and T_1ρ relaxation times, which are unique surrogate indices of fibrosis, edema and iron deposition that can be used to monitor a disease over time or to compare patients to one another. Parametric mapping is now well-accepted in the clinical setting, but its wider dissemination is hindered by limited inter-center reproducibility and relatively long acquisition times. Recently, several new parametric mapping techniques have appeared that address both of these problems, but substantial hurdles remain for widespread clinical adoption. This review serves both as a primer for newcomers to the field of parametric mapping and as a technical update for those already well at home in it. It aims to establish what is currently needed to improve the reproducibility of parametric mapping of the heart. To this end, we first give an overview of the metrics by which a mapping technique can be assessed, such as bias and variability, as well as the basic physics behind the relaxation times themselves and what their relevance is in the prospect of myocardial tissue characterization. This is followed by a summary of routine mapping techniques and their variations. The problems in reproducibility and the sources of bias and variability of these techniques are reviewed. Subsequently, novel fast, whole-heart, and multi-parametric techniques and their merits are treated in the light of their reproducibility. This includes state of the art segmentation techniques applied to parametric maps, and how artificial intelligence is being harnessed to solve this long-standing conundrum. We finish up by sketching an outlook on the road toward inter-center reproducibility, and what to expect in the future.

Cardiac Magnetic Resonance Left Ventricle Segmentation and Function Evaluation Using a Trained Deep-Learning Model

Cardiac MRI is the gold standard for evaluating left ventricular myocardial mass (LVMM), end-systolic volume (LVESV), end-diastolic volume (LVEDV), stroke volume (LVSV), and ejection fraction (LVEF). Deep convolutional neural networks (CNNs) can provide automatic segmentation of LV myocardium (LVF) and blood cavity (LVC) and quantification of LV function; however, the performance is typically degraded when applied to new datasets. A 2D U-net with Monte-Carlo dropout was trained on 45 cine MR images and the model was used to segment 10 subjects from the ACDC dataset. The initial segmentations were post-processed using a continuous kernel-cut method. The refined segmentations were employed to update the trained model. This procedure was iterated several times and the final updated U-net model was used to segment the remaining 90 ACDC subjects. Algorithm and manual segmentations were compared using Dice coefficient (DSC) and average surface distance in a symmetric manner (ASSD). The relationships between algorithm and manual LV indices were evaluated using Pearson correlation coefficient (r), Bland-Altman analyses, and paired t-tests. Direct application of the pre-trained model yielded DSC of 0.74 ± 0.12 for LVM and 0.87 ± 0.12 for LVC. After fine-tuning, DSC was 0.81 ± 0.09 for LVM and 0.90 ± 0.09 for LVC. Algorithm LV function measurements were strongly correlated with manual analyses (r = 0.86–0.99, p < 0.0001) with minimal biases of −8.8 g for LVMM, −0.9 mL for LVEDV, −0.2 mL for LVESV, −0.7 mL for LVSV, and −0.6% for LVEF. The procedure required ∼12 min for fine-tuning and approximately 1 s to contour a new image on a Linux (Ubuntu 14.02) desktop (Inter(R) CPU i7-7770, 4.2 GHz, 16 GB RAM) with a GPU (GeForce, GTX TITAN X, 12 GB Memory). This approach provides a way to incorporate a trained CNN to segment and quantify previously unseen cardiac MR datasets without needing manual annotation of the unseen datasets.

Accelerated cardiac T1 mapping in four heartbeats with inline MyoMapNet: a deep learning-based T1 estimation approach

PURPOSE

To develop and evaluate MyoMapNet, a rapid myocardial T1 mapping approach that uses fully connected neural networks (FCNN) to estimate T1 values from four T1-weighted images collected after a single inversion pulse in four heartbeats (Look-Locker, LL4).

METHOD

We implemented an FCNN for MyoMapNet to estimate T1 values from a reduced number of T1-weighted images and corresponding inversion-recovery times. We studied MyoMapNet performance when trained using native, post-contrast T1, or a combination of both. We also explored the effects of number of T1-weighted images (four and five) for native T1. After rigorous training using in-vivo modified Look-Locker inversion recovery (MOLLI) T1 mapping data of 607 patients, MyoMapNet performance was evaluated using MOLLI T1 data from 61 patients by discarding the additional T1-weighted images. Subsequently, we implemented a prototype MyoMapNet and LL4 on a 3 T scanner. LL4 was used to collect T1 mapping data in 27 subjects with inline T1 map reconstruction by MyoMapNet. The resulting T1 values were compared to MOLLI.

RESULTS

MyoMapNet trained using a combination of native and post-contrast T1-weighted images had excellent native and post-contrast T1 accuracy compared to MOLLI. The FCNN model using four T1-weighted images yields similar performance compared to five T1-weighted images, suggesting that four T1 weighted images may be sufficient. The inline implementation of LL4 and MyoMapNet enables successful acquisition and reconstruction of T1 maps on the scanner. Native and post-contrast myocardium T1 by MOLLI and MyoMapNet was 1170 ± 55 ms vs. 1183 ± 57 ms (P = 0.03), and 645 ± 26 ms vs. 630 ± 30 ms (P = 0.60), and native and post-contrast blood T1 was 1820 ± 29 ms vs. 1854 ± 34 ms (P = 0.14), and 508 ± 9 ms vs. 514 ± 15 ms (P = 0.02), respectively.

CONCLUSION

A FCNN, trained using MOLLI data, can estimate T1 values from only four T1-weighted images. MyoMapNet enables myocardial T1 mapping in four heartbeats with similar accuracy as MOLLI with inline map reconstruction.

Emerging Techniques in Cardiac Magnetic Resonance Imaging

Cardiovascular disease is the leading cause of death and a significant contributor of health care costs. Noninvasive imaging plays an essential role in the management of patients with cardiovascular disease. Cardiac magnetic resonance (MR) can noninvasively assess heart and vascular abnormalities, including biventricular structure/function, blood hemodynamics, myocardial tissue composition, microstructure, perfusion, metabolism, coronary microvascular function, and aortic distensibility/stiffness. Its ability to characterize myocardial tissue composition is unique among alternative imaging modalities in cardiovascular disease. Significant growth in cardiac MR utilization, particularly in Europe in the last decade, has laid the necessary clinical groundwork to position cardiac MR as an important imaging modality in the workup of patients with cardiovascular disease. Although lack of availability, limited training, physician hesitation, and reimbursement issues have hampered widespread clinical adoption of cardiac MR in the United States, growing clinical evidence will ultimately overcome these challenges. Advances in cardiac MR techniques, particularly faster image acquisition, quantitative myocardial tissue characterization, and image analysis have been critical to its growth. In this review article, we discuss recent advances in established and emerging cardiac MR techniques that are expected to strengthen its capability in managing patients with cardiovascular disease. LEVEL OF EVIDENCE: 5 TECHNICAL EFFICACY: Stage 1.

Self-configuring nnU-net pipeline enables fully automatic infarct segmentation in late enhancement MRI after myocardial infarction

PURPOSE

To fully automatically derive quantitative parameters from late gadolinium enhancement (LGE) cardiac MR (CMR) in patients with myocardial infarction and to investigate if phase sensitive or magnitude reconstructions or a combination of both results in best segmentation accuracy.

METHODS

In this retrospective single center study, a convolutional neural network with a U-Net architecture with a self-configuring framework ("nnU-net") was trained for segmentation of left ventricular myocardium and infarct zone in LGE-CMR. A database of 170 examinations from 78 patients with history of myocardial infarction was assembled. Separate fitting of the model was performed, using phase sensitive inversion recovery, the magnitude reconstruction or both contrasts as input channels. Manual labelling served as ground truth. In a subset of 10 patients, the performance of the trained models was evaluated and quantitatively compared by determination of the Sørensen-Dice similarity coefficient (DSC) and volumes of the infarct zone compared with the manual ground truth using Pearson's r correlation and Bland-Altman analysis.

RESULTS

The model achieved high similarity coefficients for myocardium and scar tissue. No significant difference was observed between using PSIR, magnitude reconstruction or both contrasts as input (PSIR and MAG; mean DSC: 0.83 ± 0.03 for myocardium and 0.72 ± 0.08 for scars). A strong correlation for volumes of infarct zone was observed between manual and model-based approach (r = 0.96), with a significant underestimation of the volumes obtained from the neural network.

CONCLUSION

The self-configuring nnU-net achieves predictions with strong agreement compared to manual segmentation, proving the potential as a promising tool to provide fully automatic quantitative evaluation of LGE-CMR.

Detection of liver cirrhosis in standard T2-weighted MRI using deep transfer learning

Objectives

To investigate the diagnostic performance of deep transfer learning (DTL) to detect liver cirrhosis from clinical MRI.

Methods

The dataset for this retrospective analysis consisted of 713 (343 female) patients who underwent liver MRI between 2017 and 2019. In total, 553 of these subjects had a confirmed diagnosis of liver cirrhosis, while the remainder had no history of liver disease. T2-weighted MRI slices at the level of the caudate lobe were manually exported for DTL analysis. Data were randomly split into training, validation, and test sets (70%/15%/15%). A ResNet50 convolutional neural network (CNN) pre-trained on the ImageNet archive was used for cirrhosis detection with and without upstream liver segmentation. Classification performance for detection of liver cirrhosis was compared to two radiologists with different levels of experience (4^th-year resident, board-certified radiologist). Segmentation was performed using a U-Net architecture built on a pre-trained ResNet34 encoder. Differences in classification accuracy were assessed by the χ²-test.

Results

Dice coefficients for automatic segmentation were above 0.98 for both validation and test data. The classification accuracy of liver cirrhosis on validation (vACC) and test (tACC) data for the DTL pipeline with upstream liver segmentation (vACC = 0.99, tACC = 0.96) was significantly higher compared to the resident (vACC = 0.88, p < 0.01; tACC = 0.91, p = 0.01) and to the board-certified radiologist (vACC = 0.96, p < 0.01; tACC = 0.90, p < 0.01).

Conclusion

This proof-of-principle study demonstrates the potential of DTL for detecting cirrhosis based on standard T2-weighted MRI. The presented method for image-based diagnosis of liver cirrhosis demonstrated expert-level classification accuracy.

Key Points

• A pipeline consisting of two convolutional neural networks (CNNs) pre-trained on an extensive natural image database (ImageNet archive) enables detection of liver cirrhosis on standard T2-weighted MRI.

• High classification accuracy can be achieved even without altering the pre-trained parameters of the convolutional neural networks.

• Other abdominal structures apart from the liver were relevant for detection when the network was trained on unsegmented images.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00330-021-07858-1.

Myocardial T1-mapping and extracellular volume in pulmonary arterial hypertension: A systematic review and meta-analysis

INTRODUCTION

Elevated myocardial T₁-mapping and extracellular volume (ECV) measured on cardiac MR (CMR) imaging is associated with myocardial abnormalities such as oedema or fibrosis. This meta-analysis aims to provide a summary of T₁-mapping and ECV values in pulmonary arterial hypertension (PAH) and compare their values with controls.

METHODS

We searched CENTRAL, MEDLINE, Embase, and Web of Science in August 2020. We included CMR studies reporting T₁-mapping or ECV values in adults with any type of PAH. We calculated the mean difference of T₁-values and ECV between PAH and controls.

RESULTS

We included 12 studies with 674 participants. T₁-values were significantly higher in PAH with the highest mean difference (MD) recorded at the RV insertion points (RVIP) (108 milliseconds (ms), 95% confidence intervals (CI) 89 to 128), followed by the RV free wall (MD 91 ms, 95% CI 56 to 126). The pooled mean T₁-value in PAH at the RVIP was 1084, 95% CI (1071 to 1097) measured using 1.5 Tesla Siemens systems. ECV was also higher in PAH with an MD of 7.5%, 95% CI (5.9 to 9.1) at the RV free wall.

CONCLUSION

T₁ mapping values in PAH patients are on average 9% higher than healthy controls when assessed under the same conditions including the same MRI system, magnetic field strength or sequence used for acquisition. The highest T₁ and ECV values are at the RVIP. T₁ mapping and ECV values in PH are higher than the values reported in cardiomyopathies and were associated with poor RV function and RV dilatation.

Technical Note: Fully automatic segmental relaxometry (FASTR) for cardiac magnetic resonance T1 mapping

PURPOSE

Cardiac relaxometry techniques, particularly T1 mapping, have recently gained clinical importance in various cardiac pathologies. Myocardial T1 and extracellular volume are usually calculated from manual identification of left ventricular epicardial and endocardial regions. This is a laborious process, particularly for large volume studies. Here we present a fully automated relaxometry framework (FASTR) for segmental analysis of T1 maps (both native and postcontrast) and partition coefficient (λ).

METHODS

Patients (N = 11) were imaged postacute myocardial infarction on a 1.5T clinical scanner. The scan protocol involved CINE-SSFP imaging, native, and post-contrast T1 mapping using the Modified Look-Locker Inversion (MOLLI) recovery sequence. FASTR consisted of automatic myocardial segmentation of spatio-temporally coregistered CINE images as an initial guess, followed by refinement of the contours on the T1 maps to derive segmental T1 and λ. T1 and λ were then compared to those obtained from two trained expert observers.

RESULTS

Robust endocardial and epicardial contours were achieved on T1 maps despite the presence of infarcted tissue. Relative to experts, FASTR resulted in myocardial Dice coefficients (native T1: 0.752 ± 0.041; postcontrast T1: 0.751 ± 0.057) that were comparable to interobserver Dice (native T1: 0.803 ± 0.045; postcontrast T1: 0.799 ± 0.054). There were strong correlations observed for T1 and λ derived from experts and FASTR (native T1: r = 0.83; postcontrast T1: r = 0.87; λ: r = 0.78; P < 0.0001), which were comparable to inter-expert correlation coefficients (native T1: r = 0.90; postcontrast T1: r = 0.93; λ: r = 0.80; P < 0.0001).

CONCLUSIONS

Our fully automated framework, FASTR, can generate accurate myocardial segmentations for native and postcontrast MOLLI T1 analysis without the need for manual intervention. Such a design is appealing for high volume clinical protocols.

Retraining Convolutional Neural Networks for Specialized Cardiovascular Imaging Tasks: Lessons from Tetralogy of Fallot

Ventricular contouring of cardiac magnetic resonance imaging is the gold standard for volumetric analysis for repaired tetralogy of Fallot (rTOF), but can be time-consuming and subject to variability. A convolutional neural network (CNN) ventricular contouring algorithm was developed to generate contours for mostly structural normal hearts. We aimed to improve this algorithm for use in rTOF and propose a more comprehensive method of evaluating algorithm performance. We evaluated the performance of a ventricular contouring CNN, that was trained on mostly structurally normal hearts, on rTOF patients. We then created an updated CNN by adding rTOF training cases and evaluated the new algorithm's performance generating contours for both the left and right ventricles (LV and RV) on new testing data. Algorithm performance was evaluated with spatial metrics (Dice Similarity Coefficient (DSC), Hausdorff distance, and average Hausdorff distance) and volumetric comparisons (e.g., differences in RV volumes). The original Mostly Structurally Normal (MSN) algorithm was better at contouring the LV than the RV in patients with rTOF. After retraining the algorithm, the new MSN + rTOF algorithm showed improvements for LV epicardial and RV endocardial contours on testing data to which it was naïve (N = 30; e.g., DSC 0.883 vs. 0.905 for LV epicardium at end diastole, p < 0.0001) and improvements in RV end-diastolic volumetrics (median %error 8.1 vs 11.4, p = 0.0022). Even with a small number of cases, CNN-based contouring for rTOF can be improved. This work should be extended to other forms of congenital heart disease with more extreme structural abnormalities. Aspects of this work have already been implemented in clinical practice, representing rapid clinical translation. The combined use of both spatial and volumetric comparisons yielded insights into algorithm errors.