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

Abstract

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.

Designing Chinese hospital emergency departments to leverage artificial intelligence—a systematic literature review on the challenges and opportunities

“…Digital medical tools combined with AI technologies can rapidly and effortlessly gather and analyse vast medical data. This enables emergency physicians to receive timely decision insights, improving patient treatment outcomes ( 29 , 31 , 40 ). AI can address domain adaptation issues.…”

“…AI can accurately predict clinical deteriorations, thus shortening the prediction time for septic shock and diagnosing myocardial infarctions. Early predictions promote timely interventions, improving patient outcomes and decreasing care costs and complexity ( 30 , 31 , 33 , 40 , 41 , 47 ).…”

“…Support clinical decisions (28,30,31,(33)(34)(35)(36)(40)(41)(42)(43)(44) Enhance healthcare professional performance (28, 30, 33, 36, 38-41, 43, 45-47) Reduce workload of clinicians (28,31,32,35,37,42,45,48) Improve data acquisition (29-32, 35, 40, 42) Enhance efficiency (including cost) (29-33, 37-41, 43, 45, 47, 49, 50) Help in resource management (47)(48)(49) Tan and Mills 10.3389/fmedt.2024.1307625…”

Predicting Cardiac Issues From Echocardiograms: A Literature Review Using Deep Learning and Machine Learning Techniques

“…In order to automatically recognize regional wall motion abnormalities (RWMAs) in echocardiography data, Lin et al [45] built a DL model for myocardial infarction patients. Initially, the view selection of myocardium was processed using Xception model.…”

Ensemble learning of myocardial displacements for myocardial infarction detection in echocardiography

“…Neural networks such as U-Net ( 24 ) and U-Net++ ( 25 ) have been widely used for semantic segmentation. Zhang et al ( 26 ), Leclerc et al ( 27 ), Lin et al ( 28 ) presented a large dataset of 2D echocardiography images and proposed a U-Net-based model for accurate segmentation of the LV wall. Degerli et al ( 11 ) utilized the accurate segmentation of the LV wall to detect MI.…”