Cardiac magnetic resonance imaging (MRI) provides detailed and quantitative evaluation of the heart’s structure, function, and tissue characteristics with high-resolution spatial–temporal imaging. However, its slow imaging speed and motion artifacts are notable limitations. Undersampling reconstruction, especially data-driven algorithms, has emerged as a promising solution to accelerate scans and enhance imaging performance using highly under-sampled data. Nevertheless, the scarcity of publicly available cardiac k-space datasets and evaluation platform hinder the development of data-driven reconstruction algorithms. To address this issue, we organized the Cardiac MRI Reconstruction Challenge (CMRxRecon) in 2023, in collaboration with the 26th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI). CMRxRecon presented an extensive k-space dataset comprising cine and mapping raw data, accompanied by detailed annotations of cardiac anatomical structures. With overwhelming participation, the challenge attracted more than 285 teams and over 600 participants. Among them, 22 teams successfully submitted Docker containers for the testing phase, with 7 teams submitted for both cine and mapping tasks. All teams use deep learning based approaches, indicating that deep learning has predominately become a promising solution for the problem. The first-place winner of both tasks utilizes the E2E-VarNet architecture as backbones. In contrast, U-Net is still the most popular backbone for both multi-coil and single-coil reconstructions. This paper provides a comprehensive overview of the challenge design, presents a summary of the submitted results, reviews the employed methods, and offers an in-depth discussion that aims to inspire future advancements in cardiac MRI reconstruction models. The summary emphasizes the effective strategies observed in Cardiac MRI reconstruction, including backbone architecture, loss function, pre-processing techniques, physical modeling, and model complexity, thereby providing valuable insights for further developments in this field.@en

Cardiac T1 mapping by magnetic resonance imaging (MRI) is an important clinical tool for the diagnosis and treatment of cardiovascular diseases. In practice, involuntary cardiac and respiratory motion often results in reduced accuracy and precision in T1 estimation. Motion correction is an essential preprocessing step, however, with intensive contrast changes among baseline images, both optimization-based and deep-learning (DL)-based registration methods still struggle to estimate structural similarity between images, especially when image contrast is poor and displacement is large. In this work, we propose a novel registration metric that is highly insensitive to large contrast changes, based on modified modality independent neighborhood descriptor (mo-MIND). To accommodate severe motions, we further propose pre-deformation as an augmentation strategy at the training stage. We combine the proposed mo-MIND-based metric and the augmentation strategy in a U-Net architecture to tackle the challenges of motion correction for cardiac T1 mapping. Experimental results and ablation studies demonstrated that our method achieved improved registration performance compared to several established baselines, leading to significantly reduced T1 mapping error and improved landmark stability.

@en

Purpose: To develop and evaluate a robust cardiac B+1 mapping sequence at 3 T, using Bloch–Siegert shift (BSS)-based preparations.

Methods: A longitudinal magnetization preparation module was designed to encode |B+1 |. After magnetization tip-down, off-resonant Fermi pulses, placed symmetrically around two refocusing pulses, induced BSS, followed by tipping back of the magnetization. Bloch simulations were used to optimize refocusing pulse parameters and to assess the mapping sensitivity. Relaxation-induced B+1 error was simulated for various T 1 /T 2 times. The effective mapping range was determined in phantom experiments, and |B+1 | maps were compared to the conventional BSS method and subadiabatic hyperbolic-secant 8 (HS8) pulse-sensitized method. Cardiac B+1 maps were acquired in healthy subjects, and evaluated for repeatability and imaging plane intersection consistency. The technique was modified for three-dimensional (3D) acquisition of the whole heart in a single breath-hold, and compared to two-dimensional (2D) acquisition.

Results: Simulations indicate that the proposed preparation can be tailored to achieve high mapping sensitivity across various B+1 ranges, with maximum sensitivity at the upper B+1 range. T 1 /T 2 -induced bias did not exceed 5.2%. Experimentally reproduced B+1 sensitization closely matched simulations for B+1 ≥ 0.3B+1, max (mean difference 0.031±0.022, compared to 0.018±0.025 in the HS8-sensitized method), and showed 20-fold reduction in the standard deviation of repeated scans, compared with conventional BSS B+1 mapping, and an equivalent 2-fold reduction compared with HS8-sensitization. Robust cardiac B+1 map quality was obtained, with an average test-retest variability of 0.027±0.043 relative to normalized B+1 magnitude, and plane intersection bias of 0.052±0.031. 3D acquisitions showed good agreement with2D scans (mean absolute deviation 0.055±0.061).

Conclusion: BSS-based preparations enable robust and tailorable 2D/3D cardiac B+1 mapping at 3 T in a single breath-hold.@en

Deep learning-based methods have achieved prestigious performance for magnetic resonance imaging (MRI) reconstruction, enabling fast imaging for many clinical applications. Previous methods employ convolutional networks to learn the image prior as the regularization term. In quantitative MRI, the physical model of nuclear magnetic resonance relaxometry is known, providing additional prior knowledge for image reconstruction. However, traditional reconstruction networks are limited to learning the spatial domain prior knowledge, ignoring the relaxometry prior. Therefore, we propose a relaxometry-guided quantitative MRI reconstruction framework to learn the spatial prior from data and the relaxometry prior from MRI physics. Additionally, we also evaluated the performance of two popular reconstruction backbones, namely, recurrent variational networks (RVN) and variational networks (VN) with U-Net. Experiments demonstrate that the proposed method achieves highly promising results in quantitative MRI reconstruction.

@en

Quantitative T1 mapping by MRI is an increasingly important tool for clinical assessment of cardiovascular diseases. The cardiac T1 map is derived by fitting a known signal model to a series of baseline images, while the quality of this map can be deteriorated by involuntary respiratory and cardiac motion. To correct motion, a template image is often needed to register all baseline images, but the choice of template is nontrivial, leading to inconsistent performance sensitive to image contrast. In this work, we propose a novel deep-learning-based groupwise registration framework, which omits the need for a template, and registers all baseline images simultaneously. We design two groupwise losses for this registration framework: the first is a linear principal component analysis (PCA) loss that enforces alignment of baseline images irrespective of the intensity variation, and the second is an auxiliary relaxometry loss that enforces adherence of intensity profile to the signal model. We extensively evaluated our method, termed “PCA-Relax”, and other baseline methods on an in-house cardiac MRI dataset including both pre- and post-contrast T1 sequences. All methods were evaluated under three distinct training-and-evaluation strategies, namely, standard, one-shot, and test-time-adaptation. The proposed PCA-Relax showed further improved performance of registration and mapping over well-established baselines. The proposed groupwise framework is generic and can be adapted to applications involving multiple images.@en

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

Methods: TRAFF2 mapping was performed using a breath-held ECG-gated acquisition of five images: one without preparation, three preceded by RAFF2 trains of varying duration, and one preceded by a saturation prepulse. Pixel-wise TRAFF2 maps were obtained after three-parameter exponential fitting. The repeatability of TRAFF2, T1, and T2 was assessed in phantom via the coefficient of variation (CV) across three repetitions. In seven healthy subjects, TRAFF2 was tested for precision, reproducibility, inter-subject variability, and image quality (IQ) on a Likert scale (1 = Nondiagnostic, 5 = Excellent). Additionally, TRAFF2 mapping was performed in three patients with suspected cardiovascular disease, comparing it to late gadolinium enhancement (LGE), native T1, T2, and ECV mapping.

Results: In phantom, TRAFF2 showed good repeatability (CV < 1.5%) while showing no (R2=0.09) and high (R2=0.99) correlation with T1 and T2, respectively. Myocardial TRAFF2 maps exhibited overall acceptable image quality (IQ = 3.0±1.0) with moderate artifact levels, stemming from off-resonances near the coronary sinus. Average TRAFF2 time across subjects and repetitions was 79.1 ± 7.3 ms. Good precision (7.6 ± 1.4%), reproducibility (1.0 ± 0.6%), and low inter-subject variability (10.0 ± 1.8%) were obtained. In patients, visual agreement of the infarcted area was observed in the TRAFF2 map and LGE.

Conclusion: Myocardial TRAFF2 quantification at 3 T was successfully achieved in a single breath-hold with acceptable image quality, albeit with residual off-resonance artifacts. Nonetheless, preliminary clinical data indicate potential sensitivity of TRAFF2 mapping to myocardial infarction detection without the need for contrast agents, but off-resonance artifacts mitigation warrants further investigation.@en

Background
Cardiovascular magnetic resonance (CMR) is increasingly utilized to evaluate expanding cardiovascular conditions. The Society for Cardiovascular Magnetic Resonance (SCMR) Registry is a central repository for real-world clinical data to support cardiovascular research, including those relating to outcomes, quality improvement, and machine learning. The SCMR Registry is built on a regulatory-compliant, cloud-based infrastructure that houses searchable content and Digital Imaging and Communications in Medicine images. The goal of this study is to summarize the status of the SCMR Registry at 150,000 exams.

Methods
The processes for data security, data submission, and research access are outlined. We interrogated the Registry and presented a summary of its contents.

Results
Data were compiled from 154,458 CMR scans across 20 United States sites, containing 299,622,066 total images (∼100 terabytes of storage). Across reported values, the human subjects had an average age of 58 years (range 1 month to >90 years old), were 44% (63,070/145,275) female, 72% (69,766/98,008) Caucasian, and had a mortality rate of 8% (9,962/132,979). The most common indication was cardiomyopathy (35,369/131,581, 27%), and most frequently used current procedural terminology code was 75561 (57,195/162,901, 35%). Macrocyclic gadolinium-based contrast agents represented 89% (83,089/93,884) of contrast utilization after 2015. Short-axis cines were performed in 99% (76,859/77,871) of tagged scans, short-axis late gadolinium enhancement (LGE) in 66% (51,591/77,871), and stress perfusion sequences in 30% (23,241/77,871). Mortality data demonstrated increased mortality in patients with left ventricular ejection fraction <35%, the presence of wall motion abnormalities, stress perfusion defects, and infarct LGE, compared to those without these markers. There were 456,678 patient-years of all-cause mortality follow-up, with a median follow-up time of 3.6 years.

Conclusion
The vision of the SCMR Registry is to promote evidence-based utilization of CMR through a collaborative effort by providing a web mechanism for centers to securely upload de-identified data and images for research, education, and quality control. The Registry quantifies changing practice over time and supports large-scale real-world multicenter observational studies of prognostic utility.@en

Background: Voltage mapping to detect ventricular scar is important for guiding catheter ablation, but the field-of-view of unipolar, bipolar, conventional, and microelectrodes as it relates to the extent of viable myocardium (VM) is not well defined. Objectives: The purpose of this study was to evaluate electroanatomic voltage-mapping (EAVM) with different-size electrodes for identifying VM, validated against high-resolution ex-vivo cardiac magnetic resonance (HR-LGE-CMR). Methods: A total of 9 swine with early-reperfusion myocardial infarction were mapped with the QDOT microcatheter. HR-LGE-CMR (0.3-mm slices) were merged with EAVM. At each EAVM point, the underlying VM in multisize transmural cylinders and spheres was quantified from ex vivo CMR and related to unipolar and bipolar voltages recorded from conventional and microelectrodes. Results: In each swine, 220 mapping points (Q1, Q3: 216, 260 mapping points) were collected. Infarcts were heterogeneous and nontransmural. Unipolar and bipolar voltage increased with VM volumes from >175 mm³ up to >525 mm³ (equivalent to a 5-mm radius cylinder with height >6.69 mm). VM volumes in subendocardial cylinders with 1- or 3-mm depth correlated poorly with all voltages. Unipolar voltages recorded with conventional and microelectrodes were similar (difference 0.17 ± 2.66 mV) and correlated best to VM within a sphere of radius 10 and 8 mm, respectively. Distance-weighting did not improve the correlation. Conclusions: Voltage increases with transmural volume of VM but correlates poorly with small amounts of VM, which limits EAVM in defining heterogeneous scar. Microelectrodes cannot distinguish thin from thick areas of subendocardial VM. The field-of-view for unipolar recordings for microelectrodes and conventional electrodes appears to be 8 to 10 mm, respectively, and unexpectedly similar.

@en

Accurate biventricular segmentation of cardiac magnetic resonance (CMR) cine images is essential for the clinical evaluation of heart function. However, compared to left ventricle (LV), right ventricle (RV) segmentation is still more challenging and less reproducible. Degenerate performance frequently occurs at the RV base, where the in-plane anatomical structures are complex (with atria, valve, and aorta) and vary due to the strong interplanar motion. In this work, we propose to address the currently unsolved issues in CMR segmentation, specifically at the RV base, with two strategies: first, we complemented the public resource by reannotating the RV base in the ACDC dataset, with refined delineation of the right ventricle outflow tract (RVOT), under the guidance of an expert cardiologist. Second, we proposed a novel dual encoder U-Net architecture that leverages temporal incoherence to inform the segmentation when interplanar motions occur. The inter-planar motion is characterized by loss-of-tracking, via Bayesian uncertainty of a motion-tracking model. Our experiments showed that our method significantly improved RV base segmentation taking into account temporal incoherence. Furthermore, we investigated the reproducibility of deep learning-based segmentation and showed that the combination of consistent annotation and loss of tracking could enhance the reproducibility of RV segmentation, potentially facilitating a large number of clinical studies focusing on RV.@en

Quantitative cardiac magnetic resonance imaging (MRI) is an increasingly important diagnostic tool for cardiovascular diseases. Yet, co-registration of all baseline images within the quantitative MRI sequence is essential for the accuracy and precision of quantitative maps. However, co-registering all baseline images from a quantitative cardiac MRI sequence remains a nontrivial task because of the simultaneous changes in intensity and contrast, in combination with cardiac and respiratory motion. To address the challenge, we propose a novel motion correction framework based on robust principle component analysis (rPCA) that decomposes quantitative cardiac MRI into low-rank and sparse components, and we integrate the groupwise CNN-based registration backbone within the rPCA framework. The low-rank component of rPCA corresponds to the quantitative mapping (i.e. limited degree of freedom in variation), while the sparse component corresponds to the residual motion, making it easier to formulate and solve the groupwise registration problem. We evaluated our proposed method on cardiac T1 mapping by the modified Look-Locker inversion recovery (MOLLI) sequence, both before and after the Gadolinium contrast agent administration. Our experiments showed that our method effectively improved registration performance over baseline methods without introducing rPCA, and reduced quantitative mapping error in both in-domain (pre-contrast MOLLI) and out-of-domain (post-contrast MOLLI) inference. The proposed rPCA framework is generic and can be integrated with other registration backbones.

@en

In radiological practice, multi-sequence MRI is routinely acquired to characterize anatomy and tissue. However, due to the heterogeneity of imaging protocols and contraindications to contrast agents, some MRI sequences, e.g. contrast-enhanced T1-weighted image (T1ce), may not be acquired. This creates difficulties for large-scale clinical studies for which heterogeneous datasets are aggregated. Modern deep learning techniques have demonstrated the capability of synthesizing missing sequences from existing sequences, through learning from an extensive multi-sequence MRI dataset. In this paper, we propose a novel MR image translation solution based on local implicit neural representations. We split the available MRI sequences into local patches and assign to each patch a local multi-layer perceptron (MLP) that represents a patch in the T1ce. The parameters of these local MLPs are generated by a hypernetwork based on image features. Experimental results and ablation studies on the BraTS challenge dataset showed that the local MLPs are critical for recovering fine image and tumor details, as they allow for local specialization that is highly important for accurate image translation. Compared to a classical pix2pix model, the proposed method demonstrated visual improvement and significantly improved quantitative scores (MSE 0.86 × 10^-3 vs. 1.02 × 10^-3 and SSIM 94.9 vs 94.3).

@en

Coronavirus disease 2019 (COVID-19) is an ongoing global pandemic that has affected nearly 600 million people to date across the world. While COVID-19 is primarily a respiratory illness, cardiac injury is also known to occur. Cardiovascular magnetic resonance (CMR) imaging is uniquely capable of characterizing myocardial tissue properties in-vivo, enabling insights into the pattern and degree of cardiac injury. The reported prevalence of myocardial involvement identified by CMR in the context of COVID-19 infection among previously hospitalized patients ranges from 26 to 60%. Variations in the reported prevalence of myocardial involvement may result from differing patient populations (e.g. differences in severity of illness) and the varying intervals between acute infection and CMR evaluation. Standardized methodologies in image acquisition, analysis, interpretation, and reporting of CMR abnormalities across would likely improve concordance between studies. This consensus document by the Society for Cardiovascular Magnetic Resonance (SCMR) provides recommendations on CMR imaging and reporting metrics towards the goal of improved standardization and uniform data acquisition and analytic approaches when performing CMR in patients with COVID-19 infection.

@en

Purpose: The aim of this study is to develop and optimize an adiabatic (Formula presented.) ((Formula presented.)) mapping method for robust quantification of spin-lock (SL) relaxation in the myocardium at 3T. Methods: Adiabatic SL (aSL) preparations were optimized for resilience against (Formula presented.) and (Formula presented.) inhomogeneities using Bloch simulations. Optimized (Formula presented.) -aSL, Bal-aSL and (Formula presented.) -aSL modules, each compensating for different inhomogeneities, were first validated in phantom and human calf. Myocardial (Formula presented.) mapping was performed using a single breath-hold cardiac-triggered bSSFP-based sequence. Then, optimized (Formula presented.) preparations were compared to each other and to conventional SL-prepared (Formula presented.) maps (RefSL) in phantoms to assess repeatability, and in 13 healthy subjects to investigate image quality, precision, reproducibility and intersubject variability. Finally, aSL and RefSL sequences were tested on six patients with known or suspected cardiovascular disease and compared with LGE, (Formula presented.), and ECV mapping. Results: The highest (Formula presented.) preparation efficiency was obtained in simulations for modules comprising 2 HS pulses of 30 ms each. In vivo (Formula presented.) maps yielded significantly higher quality than RefSL maps. Average myocardial (Formula presented.) values were 183.28 (Formula presented.) 25.53 ms, compared with 38.21 (Formula presented.) 14.37 ms RefSL-prepared (Formula presented.). (Formula presented.) maps showed a significant improvement in precision (avg. 14.47 (Formula presented.) 3.71% aSL, 37.61 (Formula presented.) 19.42% RefSL, p < 0.01) and reproducibility (avg. 4.64 (Formula presented.) 2.18% aSL, 47.39 (Formula presented.) 12.06% RefSL, p < 0.0001), with decreased inter-subject variability (avg. 8.76 (Formula presented.) 3.65% aSL, 51.90 (Formula presented.) 15.27% RefSL, p < 0.0001). Among aSL preparations, (Formula presented.) -aSL achieved the better inter-subject variability. In patients, (Formula presented.) -aSL preparations showed the best artifact resilience among the adiabatic preparations. (Formula presented.) times show focal alteration colocalized with areas of hyper-enhancement in the LGE images. Conclusion: Adiabatic preparations enable robust in vivo quantification of myocardial SL relaxation times at 3T.

@en

Background: Electroanatomical voltage mapping (EAVM) has been compared with late gadolinium enhancement cardiovascular magnetic resonance (LGE-CMR), which cannot delineate diffuse fibrosis. T1-mapping CMR overcomes the limitations of LGE-CMR, but it has not been directly compared against EAVM. Objectives: This study aims to assess the relationship between left ventricular (LV) endocardial voltage obtained by EAVM and extracellular volume (ECV) obtained by T1 mapping. Methods: The study investigated patients who underwent endocardial EAVM for ventricular arrhythmias (CARTO 3, Biosense Webster) together with preprocedural contrast-enhanced T1 mapping (Ingenia 3T, Philips Healthcare). After image integration, EAVM datapoints were projected onto LGE-CMR and ECV-encoded images. Average values of unipolar voltage (UV), bipolar voltage (BV), LGE transmurality, and ECV were merged from corresponding cardiac segments (6 per slice) and pooled for analysis. Results: The analysis included data from 628 segments from 18 patients (57 ± 13 years of age, 17% females, LV ejection fraction 48% ± 14%, nonischemic/ischemic cardiomyopathy/controls: 8/6/4 patients). Based on the 95th and 5th percentile values obtained from the controls, ECV >33%, BV <2.9 mV, and UV <6.7 mV were considered abnormal. There was a significant inverse association between voltage and ECV, but only in segments with abnormal ECV. Increased ECV could predict abnormal BV and UV with acceptable accuracy (area under the curve of 0.78 [95% CI: 0.74-0.83] and 0.84 [95% CI: 0.79-0.88]). Conclusions: This study found a significant inverse relationship between LV endocardial voltage and ECV. Real-time integration of T1 mapping may guide catheter mapping and may allow identification of areas of diffuse fibrosis potentially related to ventricular arrhythmias.

@en

Purpose: To investigate and mitigate the influence of physiological and acquisition-related parameters on myocardial blood flow (MBF) measurements obtained with myocardial Arterial Spin Labeling (myoASL). Methods: A Flow-sensitive Alternating Inversion Recovery (FAIR) myoASL sequence with bSSFP and spoiled GRE (spGRE) readout is investigated for MBF quantification. Bloch-equation simulations and phantom experiments were performed to evaluate how variations in acquisition flip angle (FA), acquisition matrix size (AMS), heart rate (HR) and blood (Formula presented.) relaxation time ((Formula presented.)) affect quantification of myoASL-MBF. In vivo myoASL-images were acquired in nine healthy subjects. A corrected MBF quantification approach was proposed based on subject-specific (Formula presented.) values and, for spGRE imaging, subtracting an additional saturation-prepared baseline from the original baseline signal. Results: Simulated and phantom experiments showed a strong dependence on AMS and FA ((Formula presented.) >0.73), which was eliminated in simulations and alleviated in phantom experiments using the proposed saturation-baseline correction in spGRE. Only a very mild HR dependence ((Formula presented.) >0.59) was observed which was reduced when calculating MBF with individual (Formula presented.). For corrected spGRE, in vivo mean global spGRE-MBF ranged from 0.54 to 2.59 mL/g/min and was in agreement with previously reported values. Compared to uncorrected spGRE, the intra-subject variability within a measurement (0.60 mL/g/min), between measurements (0.45 mL/g/min), as well as the inter-subject variability (1.29 mL/g/min) were improved by up to 40% and were comparable with conventional bSSFP. Conclusion: Our results show that physiological and acquisition-related factors can lead to spurious changes in myoASL-MBF if not accounted for. Using individual (Formula presented.) and a saturation-baseline can reduce these variations in spGRE and improve reproducibility of FAIR-myoASL against acquisition parameters.

@en

Quantitative MRI (qMRI) of the heart has become an important clinical tool for examining myocardial tissue properties. Because heart is a moving object, it is usually imaged with electrocardiogram and respiratory gating during acquisition, to “freeze” its motion. In reality, gating is more-often-than-not imperfect given the heart rate variability and nonideal breath-hold. qMRI of the heart, consequently, is characteristic of varying image contrast as well as residual motion, the latter compromising the quality of quantitative mapping. Motion correction is an important step prior to parametric mapping, however, a long-standing difficulty for registering the dynamic sequence is that the contrast across frames varies wildly: depending on the acquisition scheme some frames can have extremely poor contrast, which fails both traditional optimization-based and modern learning-based registration methods. In this work, we propose a novel framework named DisQ, which Disentangles Quantitative mapping sequences into the latent space of contrast and anatomy, fully unsupervised. The disentangled latent spaces serve for the purpose of generating a series of images with identical contrast, which enables easy and accurate registration of all frames. We applied our DisQ method to the modified Look-Locker inversion recovery (MOLLI) sequence, and demonstrated improved performance of T₁ mapping. In addition, we showed the possibility of generating a dynamic series of baseline images with exactly the same shape, strictly registered and perfectly “frozen". Our proposed DisQ methodology readily extends to other types of cardiac qMRI such as T₂ mapping and perfusion.

@en

Purpose: To develop automated vestibular schwannoma measurements on contrast-enhanced T1-and T2-weighted MRI scans. Materials and Methods: MRI data from 214 patients in 37 different centers were retrospectively analyzed between 2020 and 2021. Patients with hearing loss (134 positive for vestibular schwannoma [mean age 6 SD, 54 years 6 12; 64 men] and 80 negative for vestibular schwannoma) were randomly assigned to a training and validation set and to an independent test set. A convolutional neural network (CNN) was trained using fivefold cross-validation for two models (T1 and T2). Quantitative analysis, including Dice index, Hausdorff distance, surface-to-surface distance (S2S), and relative volume error, was used to compare the computer and the human delineations. An observer study was performed in which two experienced physicians evaluated both delineations. Results: The T1-weighted model showed state-of-the-art performance, with a mean S2S distance of less than 0.6 mm for the whole tumor and the intrameatal and extrameatal tumor parts. The whole tumor Dice index and Hausdorff distance were 0.92 and 2.1 mm in the independent test set, respectively. T2-weighted images had a mean S2S distance less than 0.6 mm for the whole tumor and the intrameatal and extrameatal tumor parts. The whole tumor Dice index and Hausdorff distance were 0.87 and 1.5 mm in the independent test set. The observer study indicated that the tool was similar to human delineations in 85%–92% of cases. Conclusion: The CNN model detected and delineated vestibular schwannomas accurately on contrast-enhanced T1-and T2-weighted MRI scans and distinguished the clinically relevant difference between intrameatal and extrameatal tumor parts.

@en

The self-configuring nnU-Net has achieved leading performance in a large range of medical image segmentation challenges. It is widely considered as the model of choice and a strong baseline for medical image segmentation. However, despite its extraordinary performance, nnU-Net does not supply a measure of uncertainty to indicate its possible failure. This can be problematic for large-scale image segmentation applications, where data are heterogeneous and nnU-Net may fail without notice. In this work, we introduce a novel method to estimate nnU-Net uncertainty for medical image segmentation. We propose a highly effective scheme for posterior sampling of weight space for Bayesian uncertainty estimation. Different from previous baseline methods such as Monte Carlo Dropout and mean-field Bayesian Neural Networks, our proposed method does not require a variational architecture and keeps the original nnU-Net architecture intact, thereby preserving its excellent performance and ease of use. Additionally, we boost the segmentation performance over the original nnU-Net via marginalizing multi-modal posterior models. We applied our method on the public ACDC and M &M datasets of cardiac MRI and demonstrated improved uncertainty estimation over a range of baseline methods. The proposed method further strengthens nnU-Net for medical image segmentation in terms of both segmentation accuracy and quality control.

@en

Left ventricular thrombus (LVT) after acute ST-segment elevation myocardial infarction (STEMI) are generally associated with poorer outcomes for patients at long-term follow-up. We hypothesis that tissue characteristics and strain parameters by cardiac magnetic resonance (CMR) imaging may indicate the interactions of LVT with ventricular myocardium remodeling at both acute stage and chronic stages in STEMI patients. This retrospective study included 111 consecutive STEMI patients (38 with LVT and 73 without LVT). All patients underwent CMR during acute stage (within 7 days) and chronic stage (after at least 2 months) periods after percutaneous coronary intervention (PCI). Left ventricular native T1, extracellular volume (ECV), radial, circumferential, and longitudinal strain were analyzed in both phases. Major adverse cardiac events (MACE, including cardiovascular death, myocardial reinfarction, and hospitalization for heart failure), thromboembolic and bleeding events, were the clinical endpoints of the study. During the acute stage, left ventricular ejection fraction (LVEF) (OR 0.77, P value = 0.01) and longitudinal strain (OR 1.90, P value < 0.001) were correlated with LVT formation. Strain parameters were reduced, while the native T1 and ECV values of both the infarcted area and remote myocardium were elevated in LVT patients. During the chronic stage, LVT resolved in 29 of 38 patients (76%). LVT remaining patients had lower LVEF, a larger LV, and higher ECV in the acute stage than those of the LVT-resolved patients. In the long-term follow up of 678 days, LVT (HR 2.45, P value = 0.02), aneurysm (HR 1.81, P value = 0.04), and native T1 (HR 2.44, P value = 0.01) were identified as three independent predictors of MACE, the incidence of thromboembolic events and bleeding events by a multivariable stepwise Cox proportional hazards regression. STEMI patients developing LVT had worse LV function, myocardial infarction extent, strain, and higher T1 and ECV values than STEMI patients without LVT. The LVT-remaining patients in the chronic stage had poorer functional and mapping parameters beginning in the first week. During the acute stage, LVEF and global longitudinal strain were independent correlated with LVT formation. During the long-term follow up, LVT, aneurysm and elevated myocardial T1 were associated with adverse outcomes in acute STEMI patients.@en