Search

Heart failure with reduced ejection fraction (HFrEF) is associated with high mortality rates. The identification of patients with HFrEF in urgent and emergent situations can be delayed in the absen...

Accurate detection of end-systolic (ES) and end-diastolic (ED) frames in an echocardiographic cine series can be difficult but necessary pre-processing step for the development of automatic systems to measure cardiac parameters. The detection task is challenging due to variations in cardiac anatomy and heart rate often associated with pathological conditions. We formulate this problem as a regression problem and propose several deep learning-based architectures that minimize a novel global extrema structured loss function to localize the ED and ES frames. The proposed architectures integrate convolution neural networks (CNNs)-based image feature extraction model and recurrent neural networks (RNNs) to model temporal dependencies between each frame in a sequence. We explore two CNN architectures: DenseNet and ResNet, and four RNN architectures: long short-term memory, bi-directional LSTM, gated recurrent unit (GRU), and Bi-GRU, and compare the performance of these models. The optimal deep learning model consists of a DenseNet and GRU trained with the proposed loss function. On average, we achieved 0.20 and 1.43 frame mismatch for the ED and ES frames, respectively, which are within reported inter-observer variability for the manual detection of these frames.

Left ventricular ejection fraction (LVEF) is one of the key metrics to assess the heart functionality, and cardiac ultrasound (echo) is a standard imaging modality for EF measurement. There is an emerging interest to exploit the point-of-care ultrasound (POCUS) usability due to low cost and ease of access. In this work, we aim to present a computationally efficient mobile application for accurate LVEF estimation. Our proposed mobile application for LVEF estimation runs in real time on Android mobile devices that have either a wired or wireless connection to a cardiac POCUS device. We propose a pipeline for biplane ejection fraction estimation using apical two-chamber (AP2) and apical four-chamber (AP4) echo views. A computationally efficient multi-task deep fully convolutional network is proposed for simultaneous LV segmentation and landmark detection in these views, which is integrated into the LVEF estimation pipeline. An adversarial critic model is used in the training phase to impose a shape prior on the LV segmentation output. The system is evaluated on a dataset of 427 patients. Each patient has a pair of captured AP2 and AP4 echo studies, resulting in a total of more than 40,000 echo frames. The mobile system reaches a noticeably high average Dice score of 92% for LV segmentation, an average Euclidean distance error of 2.85 pixels for the detection of anatomical landmarks used in LVEF calculation, and a median absolute error of 6.2% for LVEF estimation compared to the expert cardiologist’s annotations and measurements. The proposed system runs in real time on mobile devices. The experiments show the effectiveness of the proposed system for automatic LVEF estimation by demonstrating an adequate correlation with the cardiologist’s examination.

Atrial fibrillation (AF) is the most common cardiac arrhythmia and is associated with 4-5 fold increase in stroke risk. An electrocardiogram (ECG) is normally used for its identification. However, ECG is not readily available. Additionally, the rhythm strip at echocardiogram is often misleading. Here, we propose Echo-Rhythm Net, a deep learning-based method to automate AF detection based solely on echocardiogram imagery (echo) without the need for an ECG. The proposed framework consists of three main components: an encoder that is trained using a self-supervised method, a temporal self-similarity matrix layer, and a final supervised detector trained with labels of cardiac rhythm assigned by sonographers. Our Echo-Rhythm Net, trained with 3947 cines of which only 583 are labeled, achieves an accuracy of 79% on the detection of AF in an independent test dataset of 260 cines. This result is superior to that of a trained echocardiographer, who when given the same test data without ECG information, scored an AF detection accuracy of 63%.

The globe is currently in the midst of a COVID-19 pandemic, resulting in significant morbidity and mortality. This pandemic has placed considerable stress on health care resources and providers. This document from the Canadian Association of Interventional Cardiology- Association Canadienne de Cardiologie d'intervention, specifically addresses the implications for the care of patients in the cardiac catheterization laboratory (CCL) in Canada during the COVID-19 pandemic. The key principles of this document are to maintain essential interventional cardiovascular care while minimizing risks of COVID-19 to patients and staff and maintaining the overall health care resources. As the COVID-19 pandemic evolves, procedures will be increased or reduced based on the current level of restriction to health care services. Although some consistency across the country is desirable, provincial and regional considerations will influence how these recommendations are implemented. We believe the framework and recommendations in this document will provide crucial guidance for clinicians and policy makers on the management of coronary and structural procedures in the CCL as the COVID-19 pandemic escalates and eventually abates.

Uncertainty of labels in clinical data resulting from intra-observer variability can have direct impact on the reliability of assessments made by deep neural networks. In this paper, we propose a method for modelling such uncertainty in the context of 2D echocardiography (echo), which is a routine procedure for detecting cardiovascular disease at point-of-care. Echo imaging quality and acquisition time is highly dependent on the operator’s experience level. Recent developments have shown the possibility of automating echo image quality quantification by mapping an expert’s assessment of quality to the echo image via deep learning techniques. Nevertheless, the observer variability in the expert’s assessment can impact the quality quantification accuracy. Here, we aim to model the intra-observer variability in echo quality assessment as an aleatoric uncertainty modelling regression problem with the introduction of a novel method that handles the regression problem with categorical labels. A key feature of our design is that only a single forward pass is sufficient to estimate the level of uncertainty for the network output. Compared to the 0.11 ± 0.09 absolute error (in a scale from 0 to 1) archived by the conventional regression method, the proposed method brings the error down to 0.09 ± 0.08, where the improvement is statistically significant and equivalents to 5.7% test accuracy improvement. The simplicity of the proposed approach means that it could be generalized to other applications of deep learning in medical imaging, where there is often uncertainty in clinical labels.

Atrial volumes indexed to body surface area (AVI) are robust predictors of nonvalvular atrial fibrillation (AF) recurrence after direct current cardioversion (DCCV). The incremental value of atrial emptying fraction (EmF) compared with atrial volumes as a predictor for recurrent AF after DCCV has not been evaluated. We sought to compare the predictive ability of baseline left atrial (LA) EmF, right atrial (RA) EmF, LAVI, and RAVI for post-DCCV AF recurrence at 6 months. The first 95 patients enrolled in the AF Clinic Registry with adequate echocardiogram imaging constituted the study cohort. Each patient underwent echocardiogram within 6 months before cardioversion. Maximal LAVI and RAVI, LA EmF, and RA EmF were performed offline using 4-chamber single-plane Simpson's method, averaged over 5 cycles. The mean age of the study cohort was 64 ± 12 years, and 67% were men. Only 28 patients (29%) who underwent DCCV remained in sinus rhythm at 6 months of follow-up. The remaining, 67 (71%) had reverted to AF or underwent ablation during the 6 months of follow-up. The overall performance for prediction of AF recurrence was greatest for RA EmF, area under the receiver operator characteristic curve (AUC): RA EmF 0.92, LA EmF 0.89, RAVI 0.76, and LAVI 0.63. RA and LA EmF AUCs were significantly higher than for LAVI or RAVI (max p = 0.02). In conclusion, although RAVI and LAVI are strong predictors of AF recurrence after DCCV, RA and LA EmF outperformed in this cohort.

Accurate segmentation of left ventricle (LV) in apical four chamber echocardiography cine is a key step in cardiac functionality assessment. Cardiologists roughly annotate two frames in the cardiac cycle, namely, the end-diastolic and end-systolic frames, as part of their clinical workflow, limiting the annotated data to less than 5% of the frames in the cardiac cycle. In this paper, we propose a semi-supervised learning algorithm to leverage the unlabeled data to improve the performance of LV segmentation algorithms. This approach is based on a generative model which learns an inverse mapping from segmentation masks to their corresponding echo frames. This generator is then used as a critic to assess and improve the LV segmentation mask generated by a given segmentation algorithm such as U-Net. This semi-supervised approach enforces a prior on the segmentation model based on the perceptual similarity of the generated frame with the original frame. This approach promotes utilization of the unlabeled samples, which, in turn, improves the segmentation accuracy.

Transthoracic echocardiography (echo) is the most common imaging modality for diagnosis of cardiac conditions. Echo is acquired from a multitude of views, each of which distinctly highlights specific regions of the heart anatomy. In this paper, we present an approach based on knowledge distillation to obtain a highly accurate lightweight deep learning model for classification of 12 standard echocardiography views. The knowledge of several deep learning architectures based on the three common state-of-the-art architectures, VGG-16, DenseNet, and Resnet, are distilled to train a set of lightweight models. Networks were developed and evaluated using a dataset of 16,612 echo cines obtained from 3,151 unique patients across several ultrasound imaging machines. The best accuracy of 89.0% is achieved by an ensemble of the three very deep models while we show an ensemble of lightweight models has a comparable accuracy of 88.1%. The lightweight models have approximately 1% of the very deep model parameters and are six times faster in run-time. Such lightweight view classification models could be used to build fast mobile applications for real-time point-of-care ultrasound diagnosis.

Accurate detection of heart-related diseases in echocardiography (echo) often requires determining the performance of cardiac valves or contractile events such as strain at a high temporal resolution. In high-end cart-based imaging systems, this is achieved by increasing the frame rate using specialized beamforming and imaging hardware, or by limiting the imaging field of view (FOV). In point-of-care imaging, such a high frame rate imaging technology is currently unavailable. In this paper, we propose a new frame rate up-conversion technique, as a post-processing step during or after the echo acquisition. The proposed technique takes advantage of both variational autoencoders (VAE) and generative adversarial networks (GAN), and produces realistic frames at a high frame rate that can be used to augment conventional imaging. The proposed technique is robust to variations in heart rate since its latent space not only uses immediate previous frames, but it also takes into account the appearance of end-diastolic and end-systolic frames in its estimation. Our results show that the proposed technique can increase the frame rate by at least 5 times without any requirement for limiting the imaging FOV.

Accurate segmentation of left ventricle (LV) from echocardiograms is a key step toward diagnosis of cardiovascular diseases. Manual segmentation of the LV done by sonographers or cardiologists can be time-consuming, and its accuracy is subjective to the operator’s experience and skill level. Automation of LV segmentation is a challenging task due to a number of factors such as the presence of speckle and a high operator-dependent variability in acquiring echocardiography data. In this paper, we present a method that integrates deep recurrent fully-convolutional networks and optical flow estimation to accurately segment the LV in the apical four-chamber (A4C) view. Our method analyzes the temporal information in echocardiogram cines with the use of convolutional bi-directional long short-term memory units. Furthermore, it uses optical flow motion estimation between consecutive frames to improve the segmentation accuracy. The proposed method is evaluated over an echo cine dataset of 566 patients. Experiments show that the proposed system can reach a noticeably high mean accuracy of 97.9%, and mean Dice score of 92.7% for LV segmentation in A4C view.

Accurate diagnosis in cardiac ultrasound requires high quality images, containing different specific features and structures depending on which of the 14 standard cardiac views the operator is attempting to acquire. Inexperienced operators can have a great deal of difficulty recognizing these features and thus can fail to capture diagnostically relevant heart cines. This project aims to mitigate this challenge by providing operators with real-time feedback in the form of view classification and quality estimation. Our system uses a frame grabber to capture the raw video output of the ultrasound machine, which is then fed into an Android mobile device, running a customized mobile implementation of the TensorFlow inference engine. By multi-threading four TensorFlow instances together, we are able to run the system at 30 Hz with a latency of under 0.4 s.

Aims Left atrial appendage (LAA) device imaging after endovascular closure is important to assess for device thrombus, residual leak, positioning, surrounding structures, and pericardial effusion. Cardiac CT angiography (CCTA) is well suited to assess these non-invasively. Methods and results We report our consecutive series of non-valvular atrial fibrillation patients who underwent CCTA post-LAA closure with Amplatzer Cardiac Plug (ACP), Amulet (second generation ACP), or WATCHMAN devices. Patients underwent CCTA typically 1–6 months post-implantation. Prospective cardiac-gated CCTA was performed with Toshiba 320-detector or Siemens 2nd generation 128-slice dual-source scanners, and images interpreted with VitreaWorkstation™. GFR

Fabry disease is an X-linked lysosomal storage disorder that results from a deficiency of α-galactosidase A. Increased left ventricular wall thickness has been the most commonly described cardiovascular manifestation of the disease. However, a variety of other structural and functional abnormalities have also been reported. Echocardiography is an effective noninvasive method of assessing the cardiac involvement of Fabry disease. A more precise and comprehensive characterization of Fabry cardiomyopathy using conventional and novel echocardiographic techniques may lead to earlier diagnosis, more accurate prognostication, and timely treatment. The aim of this review is to provide a comprehensive overview of the structural and functional abnormalities on echocardiography that have thus far been described in patients with Fabry disease and to highlight potential areas that would benefit from further research.

Echocardiography (echo) is a skilled technical procedure that depends on the experience of the operator. The aim of this paper is to reduce user variability in data acquisition by automatically computing a score of echo quality for operator feedback. To do this, a deep convolutional neural network model, trained on a large set of samples, was developed for scoring apical four-chamber (A4C) echo. In this paper, 6,916 end-systolic echo images were manually studied by an expert cardiologist and were assigned a score between 0 (not acceptable) and 5 (excellent). The images were divided into two independent training-validation and test sets. The network architecture and its parameters were based on the stochastic approach of the particle swarm optimization on the training-validation data. The mean absolute error between the scores from the ultimately trained model and the expert’s manual scores was 0.71 ± 0.58. The reported error was comparable to the measured intra-rater reliability. The learned features of the network were visually interpretable and could be mapped to the anatomy of the heart in the A4C echo, giving confidence in the training result. The computation time for the proposed network architecture, running on a graphics processing unit, was less than 10 ms per frame, sufficient for real-time deployment. The proposed approach has the potential to facilitate the widespread use of echo at the point-of-care and enable early and timely diagnosis and treatment. Finally, the approach did not use any specific assumptions about the A4C echo, so it could be generalizable to other standard echo views.

Echocardiography (echo) is the most common test for diagnosis and management of patients with cardiac condi- tions. While most medical imaging modalities benefit from a relatively automated procedure, this is not the case for echo and the quality of the final echo view depends on the competency and experience of the sonographer. It is not uncommon that the sonographer does not have adequate experience to adjust the transducer and acquire a high quality echo, which may further affect the clinical diagnosis. In this work, we aim to aid the operator during image acquisition by automatically assessing the quality of the echo and generating the Automatic Echo Score (AES). This quality assessment method is based on a deep convolutional neural network, trained in an end-to-end fashion on a large dataset of apical four-chamber (A4C) echo images. For this project, an expert car- diologist went through 2,904 A4C images obtained from independent studies and assessed their condition based on a 6-scale grading system. The scores assigned by the expert ranged from 0 to 5. The distribution of scores among the 6 levels were almost uniform. The network was then trained on 80% of the data (2,345 samples). The average absolute error of the trained model in calculating the AES was 0.8 ± 0:72. The computation time of the GPU implementation of the neural network was estimated at 5 ms per frame, which is sufficient for real-time deployment.

Characterisation of cardiac cycle phase in echocardiography data is a necessary preprocessing step for developing automated systems that measure various cardiac parameters. Accurate characterisation is challenging, due to differences in appearance of the cardiac anatomy and the variability of heart rate in individuals. Here, we present a method for automatic recognition of cardiac cycle phase from echocardiograms by using a new deep neural networks architecture. Specifically, we propose to combine deep residual neural networks (ResNets), which extract the hierarchical features from the individual echocardiogram frames, with recurrent neural networks (RNNs), which model the temporal dependencies between sequential frames. We demonstrate that such new architecture produces results that outperform baseline architecture for the automatic characterisation of cardiac cycle phase in large datasets of echocardiograms containing different levels of pathological conditions.

Echocardiography (echo) is a clinical imaging technique which is highly dependent on operator experience. We aim to reduce operator variability in data acquisition by automatically computing an echo quality score for real-time feedback. We achieve this with a deep neural network model, with convolutional layers to extract hierarchical features from the input echo cine and recurrent layers to leverage the sequential information in the echo cine loop. Using data from 509 separate patient studies, containing 2,450 echo cines across five standard echo imaging planes, we achieved a mean quality score accuracy of 85\(\%\) compared to the gold-standard score assigned by experienced echosonographers. The proposed approach calculates the quality of a given 20 frame echo sequence within 10 ms, sufficient for real-time deployment.

Background Detecting bioprosthetic mitral valve dysfunction on transthoracic echocardiography can be challenging because of acoustic shadowing of regurgitant jets and a wide normal range of transvalvular gradients. Several studies in mechanical mitral valves have demonstrated the utility of the transthoracically derived parameters E (peak early mitral inflow velocity), pressure half-time, and the ratio of mitral inflow velocity-time integral (VTI MV ) to left ventricular outflow tract velocity-time integral (VTI LVOT ) in detecting significant prosthetic dysfunction. Uncertainty exists as to their applicability and appropriate cutoff levels in bioprosthetic valves. This study was designed to establish the accuracy and appropriate normal limits of routinely collected transthoracic Doppler parameters when used to assess bioprosthetic mitral valve function. Methods A total of 135 clinically stable patients with bioprosthetic mitral valves who had undergone both transthoracic and transesophageal echocardiography within a 6-month period were retrospectively identified from the past 11 years of the echocardiography database. Transthoracic findings were labeled as normal ( n = 81), regurgitant ( n = 44), or stenotic ( n = 10) according to the patient's transesophageal echocardiographic findings. Univariate and multivariate analyses were performed to identify Doppler parameters that detected dysfunction; then receiver operating characteristic curves were created to establish appropriate normal cutoff levels. Results The VTI MV /VTI LVOT ratio was the most accurate Doppler parameter at detecting valvular dysfunction, with a ratio of >2.5 providing sensitivity of 100% and specificity of 95%. E > 1.9 m/sec was slightly less accurate (93% sensitivity, 72% specificity), while a pressure half-time of >170 msec had both 100% specificity and sensitivity for detecting significant bioprosthetic mitral valve stenosis, (although it did not differentiate between regurgitant and normal). Conclusions This study demonstrates that Doppler parameters derived from transthoracic echocardiography can accurately detect bioprosthetic mitral valve dysfunction. These parameters, particularly a VTI MV /VTI LVOT ratio of >2.5, are a sensitive way of selecting patients to undergo more invasive examination with transesophageal echocardiography.

In the above-title paper [ibid., vol. 36, no. 6, pp. 1221-1230, Jun. 2017], the first footnote should have indicated the following information: A. H. Abdi and C. Luong are joint first authors.

Systemic hypertension is a major risk factor for heart disease and stroke. Data regarding temporal relationship of left atrial (LA) remodeling to onset of hypertension are sparse. We aimed to quantitate LA structural and functional remodeling in newly diagnosed hypertensive patients.We prospectively identified 380 patients with newly diagnosed systemic hypertension naive to drug therapy, and 380 age-matched control subjects without any history or evidence of hypertension. History or evidence of prior cardiovascular events, congenital or valvular heart disease, and renal dysfunction were exclusion criteria. Prevalence of LA enlargement, LA mechanical dysfunction expressed in total emptying fraction, left ventricular (LV) diastolic dysfunction, LV hypertrophy, and their interrelationships were assessed.Of the 380 newly diagnosed hypertensive patients, 285 (75%) had LA enlargement, 308 (81%) had LA mechanical dysfunction, and 19 (5%) had LVH. Diastolic dysfunction was present in 334 (88%) of the patients. Compared to the controls, the hypertensive group had larger maximal, minimal, and pre-A LA volumes (all P 0.001). Total and active LA emptying fraction were significantly reduced (both P 0.001). Total LA emptying fraction was strongly associated with systolic blood pressure [per 10 mmHg, HR 0.94 (0.89-0.98); P 0.001], with stepwise decrease in LA emptying fraction of 6%, 10%, and 16% from the lowest (141-150 mmHg) to the top tertile of systolic blood pressure (160 mmHg).In this drug-naive cohort with newly diagnosed hypertension, LA structural and functional remodeling, and LV diastolic dysfunction were common findings prior to initiation of drug treatment. LVH was uncommon. Impairment of LA mechanical function was evident even in the mildly hypertensive subgroup.