Detection Of Regional Wall Motion Abnormalities Research Articles

Echocardiography-based AI detection of regional wall motion abnormalities and quantification of cardiac function in myocardial infarction.

ObjectiveTo compare the performance of a newly developed deep learning (DL) framework for automatic detection of regional wall motion abnormalities (RWMAs) for patients presenting with the suspicion of myocardial infarction from echocardiograms obtained with portable bedside equipment versus standard equipment.BackgroundBedside echocardiography is increasingly used by emergency department setting for rapid triage of patients presenting with chest pain. However, compared to images obtained with standard equipment, lower image quality from bedside equipment can lead to improper diagnosis. To overcome these limitations, we developed an automatic workflow to process echocardiograms, including view selection, segmentation, detection of RWMAs and quantification of cardiac function that was trained and validated on image obtained from bedside and standard equipment.MethodsWe collected 4,142 examinations from one hospital as training and internal testing dataset and 2,811 examinations from other hospital as the external test dataset. For data pre-processing, we adopted DL model to automatically recognize three apical views and segment the left ventricle. Detection of RWMAs was achieved with 3D convolutional neural networks (CNN). Finally, DL model automatically measured the size of cardiac chambers and left ventricular ejection fraction.ResultsThe view selection model identified the three apical views with an average accuracy of 96%. The segmentation model provided good agreement with manual segmentation, achieving an average Dice of 0.89. In the internal test dataset, the model detected RWMAs with AUC of 0.91 and 0.88 respectively for standard and bedside ultrasound. In the external test dataset, the AUC were 0.90 and 0.85. The automatic cardiac function measurements agreed with echocardiographic report values (e. g., mean bias is 4% for left ventricular ejection fraction).ConclusionWe present a fully automated echocardiography pipeline applicable to both standard and bedside ultrasound with various functions, including view selection, quality control, segmentation, detection of the region of wall motion abnormalities and quantification of cardiac function.

Accuracy of Emergency Physicians for Detection of Regional Wall Motion Abnormalities in Patients With Chest Pain Without ST‐Elevation Myocardial Infarction

Our aim was to evaluate the accuracy of emergency physicians (EPs) in the detection of regional wall motion abnormalities (RWMAs) using focused cardiac ultrasound (FOCUS) in patients suspected of non-having ST-elevation myocardial infarction. We prospectively enrolled patients with chest pain. Three EPs underwent didactics and hands-on-training, of 3 hours each, by an experienced cardiologist, on detecting RWMAs using 2-dimensional echocardiography. They performed a FOCUS examination to evaluate for RWMAs and recorded the echo images. Our reference standard for the detection of RWMAs was accepted as a blinded cardiologist review of the prerecorded video clips. We calculated the corrected sample size and inter-rater agreement between the EPs (82 and 0.83, respectively). The analysis of the study was performed on 89 patients. Eighty-nine patients with chest pain were screened. Emergency physicians demonstrated the detection of RWMAs with good sensitivity and even excellent specificity: 76.9% (95% confidence interval [CI], 56.4%- 91.0%) and 92.1% (95% CI, 82.4%-97.4%), respectively. The accuracy of FOCUS was 87.6% (95% CI, 79.0%-93.7%). The area under the curve from a receiver operating characteristic curve analysis, which evaluated the EPs' rate of detecting the presence or absence of RWMAs, was 0.845 (95% CI, 0.753-0.913). Our study results suggest that EPs with training in bedside echocardiography can accurately rule in patients with RWMAs in suspected non-ST-elevation myocardial infarction cases.

A Deep Learning Approach for Assessment of Regional Wall Motion Abnormality From Echocardiographic Images

ObjectivesThis study investigated whether a deep convolutional neural network (DCNN) could provide improved detection of regional wall motion abnormalities (RWMAs) and differentiate among groups of coronary infarction territories from conventional 2-dimensional echocardiographic images compared with that of cardiologists, sonographers, and resident readers. BackgroundAn effective intervention for reduction of misreading of RWMAs is needed. The hypothesis was that a DCNN trained using echocardiographic images would provide improved detection of RWMAs in the clinical setting. MethodsA total of 300 patients with a history of myocardial infarction were enrolled. From this cohort, 3 groups of 100 patients each had infarctions of the left anterior descending (LAD) artery, the left circumflex (LCX) branch, and the right coronary artery (RCA). A total of 100 age-matched control patients with normal wall motion were selected from a database. Each case contained cardiac ultrasonographs from short-axis views at end-diastolic, mid-systolic, and end-systolic phases. After the DCNN underwent 100 steps of training, diagnostic accuracies were calculated from the test set. Independently, 10 versions of the same model were trained, and ensemble predictions were performed using those versions. ResultsFor detection of the presence of WMAs, the area under the receiver-operating characteristic curve (AUC) produced by the deep learning algorithm was similar to that produced by the cardiologists and sonographer readers (0.99 vs. 0.98, respectively; p = 0.15) and significantly higher than the AUC result of the resident readers (0.99 vs. 0.90, respectively; p = 0.002). For detection of territories of WMAs, the AUC by the deep learning algorithm was similar to the AUC by the cardiologist and sonographer readers (0.97 vs. 0.95, respectively; p = 0.61) and significantly higher than the AUC by resident readers (0.97 vs. 0.83, respectively; p = 0.003). From a validation group at an independent site (n = 40), the AUC by the deep learning algorithm was 0.90. ConclusionsThe present results support the possibility of using DCNN for automated diagnosis of RWMAs in the field of echocardiography.

Percutaneous perfusion monitoring for the detection of hemodialysis induced cardiovascular injury.

The safe delivery of hemodialysis (HD) faces dual challenges; the accurate detection of systemic circulatory stress producing cardiovascular (CV) injury, and the ability to enable effective preemptive intervention for such injury. We performed a pilot study to examine the capability of a new noninvasive, real-time monitoring system to detect the deleterious effects of HD on CV stability. Eight patients were evaluated with echocardiography prior to the initiation of HD and again at peak HD stress. Continuous CV physiologic monitoring was performed throughout using oximeter-based pulse waveform analysis (CVInsight® Monitoring System, Intelomed, Inc., Warrendale, PA, USA). Longitudinal strain (LS) values for 12 left ventricular segments were generated using speckle-tracking software (EchoPac, GE), to assess the presence of HD-induced regional wall motion abnormalities (RWMA), indicative of myocardial stunning. A reduction in pulse strength (PS) of ≥40% detected by CVI was associated with the development of RWMA (P = 0.005). This reduction occurred in 6/8 patients, all of whom exhibited myocardial stunning. Two patients had no significant reduction in PS nor evidence of myocardial stunning. In subjects with cardiac stunning, the decrease in PS was evident early during HD, 11.49 ± 10 minutes into HD treatment, prior to the detection of RWMA, which were assessed at peak HD stress, mean 210 ± 16.43 minutes into HD treatment. Percutaneous perfusion monitoring, using pulse wave analysis, appears to be useful in identifying circulatory stress during HD and predicting the development of HD-induced myocardial stunning with a lead time long enough to consider timely intervention.

Pocket-size imaging device: effectiveness for ward-based transthoracic studies

Pocket-size imaging devices (PSID) are now available; their potential role in a hospital environment has been investigated but still remains undefined. We evaluated the effectiveness of PSID in 92 patients referred for bedside transthoracic echocardiogram (TTE). Patients were included where there was a focused clinical question: quantification of left ventricular function (LVF); presence of regional wall motion abnormalities (RWMA); evidence of pericardial effusion, exclusion of significant valve pathology. Each patient underwent an echocardiography evaluation using PSID and TTE. In 83 patients [k = 90%, 95% CI (82.2-95.4)], it was possible to answer the clinical question by PSID examination alone. There was agreement between the findings of PSID and TTE in 86 cases [79%; k = 47%, 95% CI (12.8-82.0)], in three cases, the clinical question was not answered by both modalities. When the clinical question was focused on LVF, the agreement was excellent [k = 96%, 95% CI (95.3-97.9)], as was the agreement in the detection of RWMA [k = 94.57%, 95% CI (82.4-95.1)]. There was also good concordance in the detection of valve pathology and pericardial effusion. Using PSID, the reduction in the scanning and reporting time was 66%. The cost-effectiveness analysis produced very favourable results: with PSE, we obtained an overall cost saving per scan of 76%, compared with TTE. This study demonstrates that PSID can provide a valuable alternative to TTE in the presence of focused clinical questions and can provide an efficient way of delivering a ward-based transthoracic echo service.

Echocardiographic assessment of regional left ventricular wall motion abnormalities in patients with tako-tsubo cardiomyopathy: comparison with anterior myocardial infarction

The aim of this study was to assess the echocardiographic distribution of regional wall motion abnormalities (RWMA) in patients with tako-tsubo cardiomyopathy (TTC) compared with anterior ST-elevation myocardial infarction (ant-STEMI). Thirty-seven TTC and 37 ant-STEMI patients underwent standard echocardiographic examination at the time of hospital admission. RWMA and the involvement of the left ventricular territories supplied by each coronary artery according to the American Society of Echocardiography classification were reported. TTC patients showed a lower left ventricular ejection fraction (37.6 ± 5.1 vs. 40.9 ± 3.7%; P = 0.002) and a higher wall motion score index (WMSI; 1.98 ± 0.2 vs. 1.51 ± 0.14; P < 0.001) compared with ant-STEMI patients. No significant differences were observed between groups with regard to detection of RWMA in the territory supplied by the left anterior descending coronary artery (LAD) (37 vs. 37; P = 1). Conversely, in TTC patients, the territories supplied by the LAD/left circumflex coronary artery (LCX) (37 vs. 31; P = 0.011), LAD/right coronary artery (RCA) (34 vs. 13; P < 0.001), RCA (33 vs. 5; P < 0.001), and RCA/LCX (31 vs. 2; P < 0.001) were more frequently involved. A cut-off value of WMSI ≥1.75 (area under the curve 0.956) and for the number of territories with RWMA ≥4 (AUC = 0.928) predicted TTC with a sensitivity of 83 and 84% and a specificity of 100 and 97%, respectively. Echocardiography revealed a distinctive pattern of contractility in TTC patients, characterized by symmetrical RWMA extending equally into the territory of distribution of all coronary arteries.

Left ventricular modelling: a quantitative functional assessment tool based on cardiac magnetic resonance imaging

We present the development and testing of a semi-automated tool to support the diagnosis of left ventricle (LV) dysfunctions from cardiac magnetic resonance (CMR). CMR short-axis images of the LVs were obtained in 15 patients and processed to detect endocardial and epicardial contours and compute volume, mass and regional wall motion (WM). Results were compared with those obtained from manual tracing by an expert cardiologist. Nearest neighbour tracking and finite-element theory were merged to calculate local myocardial strains and torsion. The method was tested on a virtual phantom, on a healthy LV and on two ischaemic LVs with different severity of the pathology. Automated analysis of CMR data was feasible in 13/15 patients: computed LV volumes and wall mass correlated well with manually extracted data. The detection of regional WM abnormalities showed good sensitivity (77.8%), specificity (85.1%) and accuracy (82%). On the virtual phantom, computed local strains differed by less than 14 per cent from the results of commercial finite-element solver. Strain calculation on the healthy LV showed uniform and synchronized circumferential strains, with peak shortening of about 20 per cent at end systole, progressively higher systolic wall thickening going from base to apex, and a 10° torsion. In the two pathological LVs, synchronicity and homogeneity were partially lost, anomalies being more evident for the more severely injured LV. Moreover, LV torsion was dramatically reduced. Preliminary testing confirmed the validity of our approach, which allowed for the fast analysis of LV function, even though future improvements are possible.

611 Surface detection and color-encoding applied to real-time three-dimensional echocardiographic images as a basis for automated assessment of left ventricular wall motion

Purpose: Color encoding of left ventricular (LV) endocardial motion in 2D echo (2DE) has been shown to improve visual detection of regional wall motion abnormalities (RWMA). Our goals were: 1) to extend this technique to color encode detected LV endocardial surfaces from real-time 3D echo (RT3DE) images; 2) to automatically detect systolic RWMA and test its accuracy against visual interpretation of 2DE images. Methods: 19 normal subjects (N), and 14 patients with RWMA (including 7 with global LV dysfunction) underwent RT3DE and 2DE (Philips). 2DE images (apical 2-, 3and 4-chamber) were reviewed by an expert cardiologist, who graded wall motion (18 segments model) as normal or abnormal (AB). RT3DE datasets were analyzed using custom software: frame-by-frame semiautomated LV surface detection was followed by logical operations applied to each pair of consecutive frames, to track and color encode endocardial motion in 3D. Then, regional fractional volume change (RFVC) in % of regional end-diastolic volume was calculated automatically by colored voxels count, and displayed as stacked color-histograms. In RWMA, RFVCs were compared with regional thresholds, derived from RFVC computed in N and optimized using ROC analysis, for automated classification of wall motion. Results: The generation of 3D color-encoded wall-motion and RFVC quantification took <10 sec. In RWMA, 182/252 segments were graded as AB (Figure 1, arrows). The automated technique agreed with the expert reader in 225/252 (89%) segments, with only 8/252 (3%) false positive and 19/252 (8%) false negative detections (sensitivity: 0.91; specificity: 0.80; accuracy: 0.84). Conclusions: The proposed technique could improve the visual assessment of wall motion from RT3DE images, and represent the basis for automated detection of RWMA.

Analysis of Regional Left Ventricular Function by Cineventriculography, Cardiac Magnetic Resonance Imaging, and Unenhanced and Contrast-Enhanced Echocardiography: A Multicenter Comparison of Methods

To define the use of cineventriculography, cardiac magnetic resonance imaging (cMRI), and unenhanced and contrast-enhanced echocardiography for detection of left ventricular (LV) regional wall motion abnormalities (RWMA). Detection of RWMA is integral to the evaluation of LV function. In 100 patients, cineventriculography and unenhanced and contrast-enhanced echocardiography were performed. Fifty-six of the patients underwent additional cMRI. RWMA were assessed referring to a 16-segment model for cMRI, unenhanced and contrast echocardiography. Cineventriculography was evaluated on a 7-segment model. Hypokinesia in one or more segments defined presence of RWMA. Interobserver agreement among three readers was determined within each imaging modality. Intermethod agreement between imaging modalities was analyzed. A standard of truth for the presence of RWMA was obtained by an independent expert panel decision (EPD) based on clinical data, electrocardiogram, coronary angiography, and blinded information from the imaging modalities. Sixty-seven patients were found to have an RWMA by EPD. Interobserver agreement expressed as kappa coefficient was 0.41 (range 0.37 to 0.44) for unenhanced echocardiography, 0.43 (range 0.29 to 0.79) for cMRT, 0.56 (range 0.44 to 0.70) for cineventriculography, and 0.77 (range 0.71 to 0.88) for contrast echocardiography. Contrast enhancement compared to unenhanced echocardiography improved agreement of echocardiography related to cMRI (kappa 0.46 vs. 0.29) and related to cineventriculography (kappa 0.59 vs. 0.28). Accuracy to detect EPD-defined RWMA was highest for contrast echocardiography, followed by cMRI, unenhanced echocardiography, and cineventriculography. Analysis of RWMA is characterized by considerable interobserver variability even using high-quality imaging modalities. Interobserver agreement on RWMA and accuracy to detect panel-defined RWMA is good using contrast echocardiography.

Volumetric Quantification of Global and Regional Left Ventricular Function From Real-Time Three-Dimensional Echocardiographic Images

Real-time 3D echocardiographic (RT3DE) data sets contain dynamic volumetric information on cardiac function. However, quantification of left ventricular (LV) function from 3D echocardiographic data is performed on cut-planes extracted from the 3D data sets and thus does not fully exploit the volumetric information. Accordingly, we developed a volumetric analysis technique aimed at quantification of global and regional LV function. RT3DE images obtained in 30 patients (Philips 7500) were analyzed by use of custom software based on the level-set approach for semiautomated detection of LV endocardial surface throughout the cardiac cycle, from which global and regional LV volume (LVV)-time and wall motion (WM)-time curves were obtained. The study design included 3 protocols. In protocol 1, time curves obtained in 16 patients were compared point-by-point with MRI data (linear regression and Bland-Altman analyses). Global LVV correlated highly with MRI (r=0.98; y=0.99x+2.3) with minimal bias (1.4 mL) and narrow limits of agreement (+/-20 mL). WM correlated highly only in basal and midventricular segments (r=0.88; y=0.85x+0.7). In protocol 2, we tested the ability of this technique to differentiate populations with known differences in LV function by studying 9 patients with dilated cardiomyopathy and 9 normal subjects. All calculated indices of global and regional systolic and diastolic LV function were significantly different between the groups. In protocol 3, we tested the feasibility of automated detection of regional WM abnormalities in 11 patients. In each segment, abnormality was detected when regional shortening fraction was below a threshold obtained in normal subjects. The automated detection agreed with expert interpretation of 2D WM in 86% of segments. Volumetric analysis of RT3DE data is clinically feasible and allows fast, semiautomated, dynamic measurement of LVV and automated detection of regional WM abnormalities.

The value of echocardiographic regional wall motion abnormalities in detecting coronary artery disease in patients with or without a dilated left ventricle

To evaluate the usefulness of echocardiographic regional wall motion abnormalities (RWMA) in detecting coronary artery disease (CAD) in patients with left ventricular (LV) sysfunction and a normal-sized or dilated left ventricle, 103 patients were studied by two-dimensional echocardiography (2DE) and cardiac catheterization. In 60 patients (group I) who had LV dysfunction and a dilated left ventricle by echo (patients with dilated cardiomyopathy), RWMA were detected in 44 patients and 38 (86%) of them had significant CAD, usually two- or three-vessel obstruction; of the 16 patients with dilated cardiomyopathy (DCM) and diffuse LV hypokinesis, eight (50%) had evidence of CAD. Thus the presence of RWMA by 2DE had an 83% sensitivity, a 57% specificity, and a 77% predictive accuracy in detecting CAD in patients with DCM and thus in distinguishing ischemic from idiopathic DCM. In 43 patients with LV dysfunction but normal LV size (group II), the sensitivity, specificity, and predictive accuracy of RWMA in detecting significant CAD was 95%, 100%, and 95%, respectively. We conclude that the detection of RWMA by 2DE is highly suggestive of significant CAD in patients with LV dysfunction and normal-sized or dilated left ventricle; the finding, however, of diffuse LV hypokinesis does not exclude CAD in these patients, especially when the left ventricle is dilated.