Realistic Vendor-Specific Synthetic Ultrasound Data for Quality Assurance of 2-D Speckle Tracking Echocardiography: Simulation Pipeline and Open Access Database

A deep learning phase-based solution in 2D echocardiography motion estimation

In this paper, we propose a new deep learning method based on Quaternion Wavelet Transform (QWT) phases of 2D echocardiographic sequences to estimate the motion and strain of myocardium. The proposed method considers intensity and phases gained from QWT as the inputs of customized PWC-Net structure, a high-performance deep network in motion estimation. We have trained and tested our proposed method performance using two realistic simulated B-mode echocardiographic sequences. We have evaluated our proposed method in terms of both geometrical and clinical indices. Our method achieved an average endpoint error of 0.06 mm per frame and 0.59 mm between End Diastole and End Systole on a simulated dataset. Correlation analysis between ground truth and the computed strain shows a correlation coefficient of 0.89, much better than the most efficient methods in the state-of-the-art 2D echocardiography motion estimation. The results show the superiority of our proposed method in both geometrical and clinical indices.

Left Ventricular Segmentation, Warping, and Myocardial Registration for Automated Strain Measurement

The left ventricular global longitudinal strain (LVGLS) is a crucial prognostic indicator. However, inconsistencies in measurements due to the speckle tracking algorithm and manual adjustments have hindered its standardization and democratization. To solve this issue, we proposed a fully automated strain measurement by artificial intelligence-assisted LV segmentation contours. The LV segmentation model was trained from echocardiograms of 368 adults (11,125 frames). We compared the registration-like effects of dynamic time warping (DTW) with speckle tracking on a synthetic echocardiographic dataset in experiment-1. In experiment-2, we enrolled 80 patients to compare the DTW method with commercially available software. In experiment-3, we combined the segmentation model and DTW method to create the artificial intelligence (AI)-DTW method, which was then tested on 40 patients with general LV morphology, 20 with dilated cardiomyopathy (DCMP), and 20 with transthyretin-associated cardiac amyloidosis (ATTR-CA), 20 with severe aortic stenosis (AS), and 20 with severe mitral regurgitation (MR). Experiments-1 and -2 revealed that the DTW method is consistent with dedicated software. In experiment-3, the AI-DTW strain method showed comparable results for general LV morphology (bias - 0.137 ± 0.398%), DCMP (- 0.397 ± 0.607%), ATTR-CA (0.095 ± 0.581%), AS (0.334 ± 0.358%), and MR (0.237 ± 0.490%). Moreover, the strain curves showed a high correlation in their characteristics, with R-squared values of 0.8879-0.9452 for those LV morphology in experiment-3. Measuring LVGLS through dynamic warping of segmentation contour is a feasible method compared to traditional tracking techniques. This approach has the potential to decrease the need for manual demarcation and make LVGLS measurements more efficient and user-friendly for daily practice.

Deep learning improves test-retest reproducibility of regional strain in echocardiography

Aims

The clinical utility of regional strain measurements in echocardiography is challenged by suboptimal reproducibility. In this study, we aimed to evaluate the test-retest reproducibility of regional longitudinal strain (RLS) per coronary artery perfusion territory (RLSTerritory) and basal-to-apical level of the left ventricle (RLSLevel), measured by a novel fully automated deep learning (DL) method based on point tracking.

Methods and results

We measured strain in a dual-centre test-retest data set that included 40 controls and 40 patients with suspected non-ST elevation acute coronary syndrome. Two consecutive echocardiograms per subject were recorded by different operators. The reproducibility of RLSTerritory and RLSLevel measured by the DL method and by three experienced observers using semi-automatic software (2D Strain, EchoPAC, GE HealthCare) was evaluated as minimal detectable change (MDC). The DL method had MDC for RLSTerritory and RLSLevel ranging from 3.6 to 4.3%, corresponding to a 33-35% improved reproducibility compared with the inter- and intraobserver scenarios (MDC 5.5-6.4% and 4.9-5.4%). Furthermore, the DL method had a lower variance of test-retest differences for both RLSTerritory and RLSLevel compared with inter- and intraobserver scenarios (all P < 0.001). Bland-Altman analyses demonstrated superior reproducibility by the DL method for the whole range of strain values compared with the best observer scenarios. The feasibility of the DL method was 93% and measurement time was only 1 s per echocardiogram.

Conclusion

The novel DL method provided fully automated measurements of RLS, with improved test-retest reproducibility compared with semi-automatic measurements by experienced observers. RLS measured by the DL method has the potential to advance patient care through a more detailed, more efficient, and less user-dependent clinical assessment of myocardial function.

Cardiac ultrasound simulation for autonomous ultrasound navigation

Introduction

Ultrasound is well-established as an imaging modality for diagnostic and interventional purposes. However, the image quality varies with operator skills as acquiring and interpreting ultrasound images requires extensive training due to the imaging artefacts, the range of acquisition parameters and the variability of patient anatomies. Automating the image acquisition task could improve acquisition reproducibility and quality but training such an algorithm requires large amounts of navigation data, not saved in routine examinations.

Methods

We propose a method to generate large amounts of ultrasound images from other modalities and from arbitrary positions, such that this pipeline can later be used by learning algorithms for navigation. We present a novel simulation pipeline which uses segmentations from other modalities, an optimized volumetric data representation and GPU-accelerated Monte Carlo path tracing to generate view-dependent and patient-specific ultrasound images.

Results

We extensively validate the correctness of our pipeline with a phantom experiment, where structures' sizes, contrast and speckle noise properties are assessed. Furthermore, we demonstrate its usability to train neural networks for navigation in an echocardiography view classification experiment by generating synthetic images from more than 1,000 patients. Networks pre-trained with our simulations achieve significantly superior performance in settings where large real datasets are not available, especially for under-represented classes.

Discussion

The proposed approach allows for fast and accurate patient-specific ultrasound image generation, and its usability for training networks for navigation-related tasks is demonstrated.

Deep Left Ventricular Motion Estimation Methods in Echocardiography: A Comparative Study

Motion estimation in echocardiography is critical when assessing heart function and calculating myocardial deformation indices. Nevertheless, there are limitations in clinical practice, particularly with regard to the accuracy and reliability of measurements retrieved from images. In this study, deep learning-based motion estimation architectures were used to determine the left ventricular longitudinal strain in echocardiography. Three motion estimation approaches, pretrained on popular optical flow datasets, were applied to a simulated echocardiographic dataset. Results show that PWC-Net, RAFT and FlowFormer achieved an average end point error of 0.20, 0.11 and 0.09 mm per frame, respectively. Additionally, global longitudinal strain was calculated from the FlowFormer outputs to assess strain correlation. Notably, there is variability in strain accuracy among different vendors. Thus, optical flow-based motion estimation has the potential to facilitate the use of strain imaging in clinical practice.

Efficient Stratified 3-D Scatterer Sampling for Freehand Ultrasound Simulation

Ultrasound image simulation is a well-explored field with the main objective of generating realistic synthetic images, further used as ground truth for computational imaging algorithms or for radiologists' training. Several ultrasound simulators are already available, most of them consisting in similar steps: 1) generate a collection of tissue mimicking individual scatterers with random spatial positions and random amplitudes; 2) model the ultrasound probe and the emission and reception schemes; and 3) generate the radio frequency (RF) signals resulting from the interaction between the scatterers and the propagating ultrasound waves. This article is focused on the first step. To ensure fully developed speckle, a few tens of scatterers by resolution cell are needed, demanding to handle high amounts of data (especially in 3-D) and resulting into important computational time. The objective of this work is to explore new scatterer spatial distributions, with application to multiple coherent 2-D slice simulations from 3-D volumes. More precisely, lazy evaluation of pseudorandom schemes proves them to be highly computationally efficient compared with uniform random distribution commonly used. We also propose an end-to-end method from the 3-D tissue volume to resulting ultrasound images using coherent and 3-D-aware scatterer generation and usage in a real-time context.

Fast strain mapping in abdominal aortic aneurysm wall reveals heterogeneous patterns

Abdominal aortic aneurysm patients are regularly monitored to assess aneurysm development and risk of rupture. A preventive surgical procedure is recommended when the maximum aortic antero-posterior diameter, periodically assessed on two-dimensional abdominal ultrasound scans, reaches 5.5 mm. Although the maximum diameter criterion has limited ability to predict aneurysm rupture, no clinically relevant tool that could complement the current guidelines has emerged so far. In vivo cyclic strains in the aneurysm wall are related to the wall response to blood pressure pulse, and therefore, they can be linked to wall mechanical properties, which in turn contribute to determining the risk of rupture. This work aimed to enable biomechanical estimations in the aneurysm wall by providing a fast and semi-automatic method to post-process dynamic clinical ultrasound sequences and by mapping the cross-sectional strains on the B-mode image. Specifically, the Sparse Demons algorithm was employed to track the wall motion throughout multiple cardiac cycles. Then, the cyclic strains were mapped by means of radial basis function interpolation and differentiation. We applied our method to two-dimensional sequences from eight patients. The automatic part of the analysis took under 1.5 min per cardiac cycle. The tracking method was validated against simulated ultrasound sequences, and a maximum root mean square error of 0.22 mm was found. The strain was calculated both with our method and with the established finite-element method, and a very good agreement was found, with mean differences of one order of magnitude smaller than the image spatial resolution. Most patients exhibited a strain pattern that suggests interaction with the spine. To conclude, our method is a promising tool for investigating abdominal aortic aneurysm wall biomechanics as it can provide a fast and accurate measurement of the cyclic wall strains from clinical ultrasound sequences.

Motion Estimation by Deep Learning in 2D Echocardiography: Synthetic Dataset and Validation

Motion estimation in echocardiography plays an important role in the characterization of cardiac function, allowing the computation of myocardial deformation indices. However, there exist limitations in clinical practice, particularly with regard to the accuracy and robustness of measurements extracted from images. We therefore propose a novel deep learning solution for motion estimation in echocardiography. Our network corresponds to a modified version of PWC-Net which achieves high performance on ultrasound sequences. In parallel, we designed a novel simulation pipeline allowing the generation of a large amount of realistic B-mode sequences. These synthetic data, together with strategies during training and inference, were used to improve the performance of our deep learning solution, which achieved an average endpoint error of 0.07 ± 0.06 mm per frame and 1.20 ± 0.67 mm between ED and ES on our simulated dataset. The performance of our method was further investigated on 30 patients from a publicly available clinical dataset acquired from a GE system. The method showed promise by achieving a mean absolute error of the global longitudinal strain of 2.5 ± 2.1% and a correlation of 0.77 compared to GLS derived from manual segmentation, much better than one of the most efficient methods in the state-of-the-art (namely the FFT-Xcorr block-matching method). We finally evaluated our method on an auxiliary dataset including 30 patients from another center and acquired with a different system. Comparable results were achieved, illustrating the ability of our method to maintain high performance regardless of the echocardiographic data processed.

Deep learning in ultrasound elastography imaging: A review

It is known that changes in the mechanical properties of tissues are associated with the onset and progression of certain diseases. Ultrasound elastography is a technique to characterize tissue stiffness using ultrasound imaging either by measuring tissue strain using quasi-static elastography or natural organ pulsation elastography, or by tracing a propagated shear wave induced by a source or a natural vibration using dynamic elastography. In recent years, deep learning has begun to emerge in ultrasound elastography research. In this review, several common deep learning frameworks in the computer vision community, such as multilayer perceptron, convolutional neural network, and recurrent neural network are described. Then, recent advances in ultrasound elastography using such deep learning techniques are revisited in terms of algorithm development and clinical diagnosis. Finally, the current challenges and future developments of deep learning in ultrasound elastography are prospected. This article is protected by copyright. All rights reserved.

Estimating Echocardiographic Myocardial Strain of Left Ventricle with Deep Learning

The global longitudinal strain of the myocardial tissue has been shown to be a better indicator of cardiac pathologies in the subclinical stage than other indices, such as the ejection fraction. This article presents a new deep learning approach for strain estimation in 2D echocardiograms. The proposed method improves the performance of the state of the art without losing stability with noisy echocardiograms and achieved an average end point error of 0.14 ± 0.17 pixels in the estimation of the optical flow in the myocardium and an error of 1.34 ± 2.34 % in the estimation of the global longitudinal strain indicator when evaluated in a synthetic echocardiographic dataset. Further research will validate the proposed method by a clinical in-vivo dataset. Clinical relevance- This paper presents a method to estimate the global longitudinal strain index in noisy echocardiograms, which promises to be a better indicator of cardiac pathologies in the subclinical stage than other indices such as the ejection fraction.

A method for direct estimation of left ventricular global longitudinal strain rate from echocardiograms

We present a new method for measuring global longitudinal strain and global longitudinal strain rate from 2D echocardiograms using a logarithmic-transform correlation (LTC) method. Traditional echocardiography strain analysis depends on user inputs and chamber segmentation, which yield increased measurement variability. In contrast, our approach is automated and does not require cardiac chamber segmentation and regularization, thus eliminating these issues. The algorithm was benchmarked against two conventional strain analysis methods using synthetic left ventricle ultrasound images. Measurement error was assessed as a function of contrast-to-noise ratio (CNR) using mean absolute error and root-mean-square error. LTC showed better agreement to the ground truth strain \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${({\varvec{R}}}^{2}=0.91)$$\end{document}(R2=0.91) and ground truth strain rate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${({\varvec{R}}}^{2}=0.85)$$\end{document}(R2=0.85) compared with agreement to ground truth for two block-matching speckle tracking algorithms (one based on sum of square difference and the other on Fourier transform correlation; strain \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${({\varvec{R}}}^{2}=0.70)$$\end{document}(R2=0.70), strain rate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${({\varvec{R}}}^{2}=0.70)$$\end{document}(R2=0.70)). A 200% increase in strain measurement accuracy was observed compared to the conventional algorithms. Subsequently, we tested the method using a 53-subject clinical cohort (20 subjects diseased with cardiomyopathy, 33 healthy controls). Our method distinguished between normal and abnormal left ventricular function with an AUC = 0.89, a 5% improvement over the conventional GLS algorithms.

A Pipeline for the Generation of Synthetic Cardiac Color Doppler

Color Doppler imaging (CDI) is the modality of choice for simultaneous visualization of myocardium and intracavitary flow over a wide scan area. This visualization modality is subject to several sources of error, the main ones being aliasing and clutter. Mitigation of these artifacts is a major concern for better analysis of intracardiac flow. One option to address these issues is through simulations. In this article, we present a numerical framework for generating clinical-like CDI. Synthetic blood vector fields were obtained from a patient-specific computational fluid dynamics CFD model. Realistic texture and clutter artifacts were simulated from real clinical ultrasound cineloops. We simulated several scenarios highlighting the effects of 1) flow acceleration; 2) wall clutter; and 3) transmit wavefronts, on Doppler velocities. As a comparison, an "ideal" color Doppler was also simulated, without these harmful effects. This synthetic dataset is made publicly available and can be used to evaluate the quality of Doppler estimation techniques. Besides, this approach can be seen as a first step toward the generation of comprehensive datasets for training neural networks to improve the quality of Doppler imaging.

Influence of Heart Rate on Left and Right Ventricular Longitudinal Strain in Patients with Chronic Heart Failure

Over the past years, a number of studies have demonstrated the relevance of strain assessed by two-dimensional speckle tracking echocardiography (STE) in evaluating ventricular function. The aim of this study was to analyze changes in left (LV) and right ventricular (RV) longitudinal strain associated with variations of heart rate (HR) in participants with and without chronic heart failure (CHF). We enrolled 45 patients, 38 of these diagnosed with CHF and carrying an implantable cardioverter defibrillator, and seven patients with pacemakers and without CHF. The frequency of atrial stimulation was increased to 90 beats/min and an echocardiogram was performed at each increase of 10 beats/min. Global LV and RV longitudinal strain (LVGLS and RVGLS, respectively) and RV free wall longitudinal strain (RVfwLS) were calculated at each HR. When analyzed as continuous variables, significant reductions in LVGLS were detected at higher HRs, whereas improvements in both RVGLS and RVfwLS were observed. Patients with a worsening of LVGLS (76% overall) were more likely to present lower baseline LV function. Only a few patients (18% for RVGLS and 16% for RVfwLS) exhibited HR-related deteriorations of RV strain measures, which was associated with lower levels of baseline RV function and higher pulmonary systolic pressures. Finally, 21 (47%) and 25 (56%) participants responded with improvements in RVGLS and RVfwLS, respectively. Our findings revealed heterogeneous RV and LV responses to increases in HR. These findings might ultimately be used to optimize cardiac functionality in patients diagnosed with CHF.

Myocardial strain analysis of echocardiography based on deep learning

BACKGROUND

Strain analysis provides more thorough spatiotemporal signatures for myocardial contraction, which is helpful for early detection of cardiac insufficiency. The use of deep learning (DL) to automatically measure myocardial strain from echocardiogram videos has garnered recent attention. However, the development of key techniques including segmentation and motion estimation remains a challenge. In this work, we developed a novel DL-based framework for myocardial segmentation and motion estimation to generate strain measures from echocardiogram videos.

METHODS

Three-dimensional (3D) Convolutional Neural Network (CNN) was developed for myocardial segmentation and optical flow network for motion estimation. The segmentation network was used to define the region of interest (ROI), and the optical flow network was used to estimate the pixel motion in the ROI. We performed a model architecture search to identify the optimal base architecture for motion estimation. The final workflow design and associated hyperparameters are the result of a careful implementation. In addition, we compared the DL model with a traditional speck tracking algorithm on an independent, external clinical data. Each video was double-blind measured by an ultrasound expert and a DL expert using speck tracking echocardiography (STE) and DL method, respectively.

RESULTS

The DL method successfully performed automatic segmentation, motion estimation, and global longitudinal strain (GLS) measurements in all examinations. The 3D segmentation has better spatio-temporal smoothness, average dice correlation reaches 0.82, and the effect of target frame is better than that of previous 2D networks. The best motion estimation network achieved an average end-point error of 0.05 ± 0.03 mm per frame, better than previously reported state-of-the-art. The DL method showed no significant difference relative to the traditional method in GLS measurement, Spearman correlation of 0.90 (p < 0.001) and mean bias -1.2 ± 1.5%.

CONCLUSION

In conclusion, our method exhibits better segmentation and motion estimation performance and demonstrates the feasibility of DL method for automatic strain analysis. The DL approach helps reduce time consumption and human effort, which holds great promise for translational research and precision medicine efforts.

Segmentation Enhanced Elastic Image Registration for 2D Speckle Tracking Echocardiography-Performance Study In Silico

Although the two dimensional Speckle Tracking Echocardiography has gained a strong position among medical diagnostic techniques in cardiology, it still requires further developments to improve its repeatability and reliability. Few works have attempted to incorporate the left ventricle segmentation results in the process of displacements and strain estimation to improve its performance. We proposed the use of mask information as an additional penalty in the elastic image registration based displacements estimation. This approach was studied using a short axis view synthetic echocardiographic data, segmented using an active contour method. The obtained masks were distorted to a different degree, using different methods to assess the influence of the segmentation quality on the displacements and strain estimation process. The results of displacements and circumferential strain estimations show, that even though the method is dependent on the mask quality, the potential loss in accuracy due to the poor segmentation quality is much lower than the potential accuracy gain in cases where the segmentation performs well.

Quality Assurance of Segmental Strain Values Provided by Commercial 2-D Speckle Tracking Echocardiography Using in Silico Models: A Report from the EACVI-ASE Strain Standardization Task Force

The aim of this study was to determine the accuracy and reproducibility of vendor-specific regional strain values by echocardiography using in silico data. Synthetic 2-D ultrasound gray-scale images of the left ventricle (LV) were generated with knowledge of the longitudinal segmental strain values from the underlying electromechanical LV model. Four of five models mimicked transmural infarctions with systolic segmental stretching in different vascular areas. Cine loops in the three apical views were synthetically generated at four noise levels. All in silico images were repeatedly analyzed by a single investigator and some by another investigator. The absolute errors varied significantly between vendors from 3.3 ± 3.1% to 11.2 ± 5.9%. The area under the curve for the identification of segmental stretching ranged from 0.80 (confidence interval: 0.77-0.83) to 0.96 (0.95-0.98). The levels of agreement for intra-investigator variability varied between -3.0% to 2.9% and -5.2% to 4.8%, and for inter-investigator variability, between -3.6% to 3.5% and -14.5% to 8.5%. Segmental strain analysis allows the identification of areas with segmental stretching with good accuracy. However, single segmental peak-strain values are not accurate and should be interpreted with caution. Nevertheless, our results indicate the usefulness of semiquantitative strain assessment for the detection of regional dysfunction.

Automated peak prominence-based iterative Dijkstras algorithm for segmentation of B-mode echocardiograms

We present a user-initialized, automated segmentation method for use with echocardiograms (echo). The method uses an iterative Dijkstra's algorithm, a strategic node selection, and a novel cost matrix formulation based on intensity peak prominence, termed the Prominence Iterative Dijkstras algorithm, or ProID. ProID is initialized with three user-input clicks per time-series scan. ProID was tested using artificial echo images representing five different systems. Results showed accurate LV contours and volume estimations as compared to the ground-truth for all systems. Using the CAMUS dataset, we demonstrate ProID maintained similar Dice similarity scores to other automated methods. ProID was then used to analyze a clinical cohort of 66 pediatric patients, including normal and diseased hearts. Output segmentations, end-diastolic, end-systolic volumes, and ejection fraction were compared against manual segmentations from two expert readers. ProID maintained an average Dice score of 0.93 when comparing against manual segmentation. Comparing the two expert readers, the manual segmentations maintained a score of 0.93 which increased to 0.95 when they used ProID. Thus, ProID reduced the inter-operator variability across the expert readers. Overall, this work demonstrates ProID yields accurate boundaries across age groups, disease states, and echo platforms with low computational cost and no need for training data.

Evaluation of strain averaging area and strain estimation errors in a spheroidal left ventricular model using synthetic image data and speckle tracking

BACKGROUND

In majority of studies on speckle tracking echocardiography (STE) the strain estimates are averaged over large areas of the left ventricle. This may impair the diagnostic capability of the STE in the case of e.g. local changes of the cardiac contractility. This work attempts to evaluate, how far one can reduce the averaging area, without sacrificing the estimation accuracy that could be important from the clinical point of view.

METHODS

Synthetic radio frequency (RF) data of a spheroidal left ventricular (LV) model were generated using FIELD II package and meshes obtained from finite element method (FEM) simulation. The apical two chamber (A2C) view and the mid parasternal short axis view (pSAXM) were simulated. The sector encompassed the entire cross-section (full view) of the LV model or its part (partial view). The wall segments obtained according to the American Heart Association (AHA17) were divided into subsegments of area decreasing down to 3 mm2. Longitudinal, circumferential and radial strain estimates, obtained using a hierarchical block-matching method, were averaged over these subsegments. Estimation accuracy was assessed using several error measures, making most use of the prediction of the maximal relative error of the strain estimate obtained using the FEM derived reference. Three limits of this predicted maximal error were studied, namely 16.7%, 33% and 66%. The smallest averaging area resulting in the strain estimation error below one of these limits was considered the smallest allowable averaging area (SAAA) of the strain estimation.

RESULTS

In all AHA17 segments, using the A2C projection, the SAAA ensuring maximal longitudinal strain estimates error below 33% was below 3 mm2, except for the segment no 17 where it was above 278 mm2. The SAAA ensuring maximal circumferential strain estimates error below 33% depended on the AHA17 segment position within the imaging sector and view type and ranged from below 3-287 mm2. The SAAA ensuring maximal radial strain estimates error below 33% obtained in the pSAXM projection was not less than 287 mm2. The SAAA values obtained using other maximal error limits differ from SAAA values observed for the 33% error limit only in limited number of cases. SAAA decreased when using maximal error limit equal to 66% in these cases. The use of the partial view (narrow sector) resulted in a decrease of the SAAA.

CONCLUSIONS

The SAAA varies strongly between strain components. In a vast part of the LV model wall in the A2C view the longitudinal strain could be estimated using SAAA below 3 mm2, which is smaller than the averaging area currently used in clinic, thus with a higher resolution. The SAAA of the circumferential strain estimation strongly depends on the position of the region of interest and the parameters of the acquisition. The SAAA of the radial strain estimation takes the highest values. The use of a narrow sector could increase diagnostic capabilities of 2D STE.

Artificial Intelligence for Automatic Measurement of Left Ventricular Strain in Echocardiography

OBJECTIVES

This study sought to examine if fully automated measurements of global longitudinal strain (GLS) using a novel motion estimation technology based on deep learning and artificial intelligence (AI) are feasible and comparable with a conventional speckle-tracking application.

BACKGROUND

GLS is an important parameter when evaluating left ventricular function. However, analyses of GLS are time consuming and demand expertise, and thus are underused in clinical practice.

METHODS

In this study, 200 patients with a wide range of left ventricle (LV) function were included. Three standard apical cine-loops were analyzed using the AI pipeline. The AI method measured GLS and was compared with a commercially available semiautomatic speckle-tracking software (EchoPAC v202, GE Healthcare, Chicago, Illinois).

RESULTS

The AI method succeeded to both correctly classify all 3 standard apical views and perform timing of cardiac events in 89% of patients. Furthermore, the method successfully performed automatic segmentation, motion estimates, and measurements of GLS in all examinations, across different cardiac pathologies and throughout the spectrum of LV function. GLS was -12.0 ± 4.1% for the AI method and -13.5 ± 5.3% for the reference method. Bias was -1.4 ± 0.3% (95% limits of agreement: 2.3 to -5.1), which is comparable with intervendor studies. The AI method eliminated measurement variability and a complete GLS analysis was processed within 15 s.

CONCLUSIONS

Through the range of LV function this novel AI method succeeds, without any operator input, to automatically identify the 3 standard apical views, perform timing of cardiac events, trace the myocardium, perform motion estimation, and measure GLS. Fully automated measurements based on AI could facilitate the clinical implementation of GLS.

Myocardial Function Imaging in Echocardiography Using Deep Learning

Deformation imaging in echocardiography has been shown to have better diagnostic and prognostic value than conventional anatomical measures such as ejection fraction. However, despite clinical availability and demonstrated efficacy, everyday clinical use remains limited at many hospitals. The reasons are complex, but practical robustness has been questioned, and a large inter-vendor variability has been demonstrated. In this work, we propose a novel deep learning based framework for motion estimation in echocardiography, and use this to fully automate myocardial function imaging. A motion estimator was developed based on a PWC-Net architecture, which achieved an average end point error of (0.06±0.04) mm per frame using simulated data from an open access database, on par or better compared to previously reported state of the art. We further demonstrate unique adaptability to image artifacts such as signal dropouts, made possible using trained models that incorporate relevant image augmentations. Further, a fully automatic pipeline consisting of cardiac view classification, event detection, myocardial segmentation and motion estimation was developed and used to estimate left ventricular longitudinal strain in vivo. The method showed promise by achieving a mean deviation of (-0.7±1.6)% compared to a semi-automatic commercial solution for N=30 patients with relevant disease, within the expected limits of agreement. We thus believe that learning-based motion estimation can facilitate extended use of strain imaging in clinical practice.

Deep Network for Scatterer Distribution Estimation for Ultrasound Image Simulation

Simulation-based ultrasound (US) training can be an essential educational tool. Realistic US image appearance with typical speckle texture can be modeled as convolution of a point spread function with point scatterers representing tissue microstructure. Such scatterer distribution, however, is in general not known and its estimation for a given tissue type is fundamentally an ill-posed inverse problem. In this article, we demonstrate a convolutional neural network approach for probabilistic scatterer estimation from observed US data. We herein propose to impose a known statistical distribution on scatterers and learn the mapping between US image and distribution parameter map by training a convolutional neural network on synthetic images. In comparison with several existing approaches, we demonstrate in numerical simulations and with in vivo images that the synthesized images from scatterer representations estimated with our approach closely match the observations with varying acquisition parameters such as compression and rotation of the imaged domain.

A Pilot Study on Convolutional Neural Networks for Motion Estimation From Ultrasound Images

In recent years, deep learning (DL) has been successfully applied to the analysis and processing of ultrasound images. To date, most of this research has focused on segmentation and view recognition. This article benchmarks different convolutional neural network algorithms for motion estimation in ultrasound imaging. We evaluated and compared several networks derived from FlowNet2, one of the most efficient architectures in computer vision. The networks were tested with and without transfer learning, and the best configuration was compared against the particle imaging velocimetry method, a popular state-of-the-art block-matching algorithm. Rotations are known to be difficult to track from ultrasound images due to a significant speckle decorrelation. We thus focused on the images of rotating disks, which could be tracked through speckle features only. Our database consisted of synthetic and in vitro B-mode images after log compression and covered a large range of rotational speeds. One of the FlowNet2 subnetworks, FlowNet2SD, produced competitive results with a motion field error smaller than 1 pixel on real data after transfer learning based on the simulated data. These errors remain small for a large velocity range without the need for hyperparameter tuning, which indicates the high potential and adaptability of DL solutions to motion estimation in ultrasound imaging.

Myocardial strain imaging: review of general principles, validation, and sources of discrepancies

Myocardial tissue tracking imaging techniques have been developed for a more accurate evaluation of myocardial deformation (i.e. strain), with the potential to overcome the limitations of ejection fraction (EF) and to contribute, incremental to EF, to the diagnosis and prognosis in cardiac diseases. While most of the deformation imaging techniques are based on the similar principles of detecting and tracking specific patterns within an image, there are intra- and inter-imaging modality inconsistencies limiting the wide clinical applicability of strain. In this review, we aimed to describe the particularities of the echocardiographic and cardiac magnetic resonance deformation techniques, in order to understand the discrepancies in strain measurement, focusing on the potential sources of variation: related to the software used to analyse the data, to the different physics of image acquisition and the different principles of 2D vs. 3D approaches. As strain measurements are not interchangeable, it is highly desirable to work with validated strain assessment tools, in order to derive information from evidence-based data. There is, however, a lack of solid validation of the current tissue tracking techniques, as only a few of the commercial deformation imaging softwares have been properly investigated. We have, therefore, addressed in this review the neglected issue of suboptimal validation of tissue tracking techniques, in order to advocate for this matter.

An optimisation-based iterative approach for speckle tracking echocardiography

Speckle tracking is the most prominent technique used to estimate the regional movement of the heart based on echocardiograms. In this study, we propose an optimised-based block matching algorithm to perform speckle tracking iteratively. The proposed technique was evaluated using a publicly available synthetic echocardiographic dataset with known ground-truth from several major vendors and for healthy/ischaemic cases. The results were compared with the results from the classic (standard) two-dimensional block matching. The proposed method presented an average displacement error of 0.57 pixels, while classic block matching provided an average error of 1.15 pixels. When estimating the segmental/regional longitudinal strain in healthy cases, the proposed method, with an average of 0.32 ± 0.53, outperformed the classic counterpart, with an average of 3.43 ± 2.84. A similar superior performance was observed in ischaemic cases. This method does not require any additional ad hoc filtering process. Therefore, it can potentially help to reduce the variability in the strain measurements caused by various post-processing techniques applied by different implementations of the speckle tracking. Graphical Abstract Standard block matching versus proposed iterative block matching approach.

Quantitative Assessment of the Effect of the Out-of-Plane Movement of the Homogenous Ellipsoidal Model of the Left Ventricle on the Deformation Measures Estimated Using 2-D Speckle Tracking-An In-Silico Study

Effect of the out-of-plane (OOP) movement amplitude on estimates of global displacements (radial, circumferential) and strains (radial , circumferential ) was studied in an ellipsoidal model of the left ventricle using finite-element modeling (FEM), synthetic ultrasonic data, and short-axis view. This effect was assessed using median of the absolute relative error (RE) of the global parameters. FEM provided node displacements for synthetic ultrasonic data and reference data generation. Displacements were estimated using block-matching (BM) and B-spline (BS) methods. FEM-derived data analysis, free from errors resulting from speckle tracking, indicated that the tissue motion introduced REs of global strain estimates below 4.5%. The effect of the OOP motion amplitude on strain estimates was strain specific and depended on the displacement estimation method. In the case of , the increase of the OOP amplitude resulted in quasi-linear increase of the RE from approximately 10% to 15%. The modulus of the end-systolic (ES) errors of the estimates almost linearly increased with increasing OOP amplitude approximately from 10% to 16%. REs of the estimate were close to 80% and 40%, respectively, in the case of the BM and BS methods, and increased with increasing OOP amplitude. The modulus of the ES errors of the estimates in the case of the BS method was about -40% and showed low sensitivity to the OOP amplitude; in the BM case, these errors varied approximately from -70% to -58% for OOP amplitude from 0 to 15 mm.