Ejection Fraction Estimation from Echocardiograms Using Optimal Left Ventricle Feature Extraction Based on Clinical Methods
TL;DR: In this paper , a convolutional neural network (CNN) was used along with Long Short-Term Memory Networks (LSTM) for the regression of the left ventricle ejection fraction (LVEF).
read more
Abstract: Echocardiography is one of the imaging systems most often utilized for assessing heart anatomy and function. Left ventricle ejection fraction (LVEF) is an important clinical variable assessed from echocardiography via the measurement of left ventricle (LV) parameters. Significant inter-observer and intra-observer variability is seen when LVEF is quantified by cardiologists using huge echocardiography data. Machine learning algorithms have the capability to analyze such extensive datasets and identify intricate patterns of structure and function of the heart that highly skilled observers might overlook, hence paving the way for computer-assisted diagnostics in this field. In this study, LV segmentation is performed on echocardiogram data followed by feature extraction from the left ventricle based on clinical methods. The extracted features are then subjected to analysis using both neural networks and traditional machine learning algorithms to estimate the LVEF. The results indicate that employing machine learning techniques on the extracted features from the left ventricle leads to higher accuracy than the utilization of Simpson’s method for estimating the LVEF. The evaluations are performed on a publicly available echocardiogram dataset, EchoNet-Dynamic. The best results are obtained when DeepLab, a convolutional neural network architecture, is used for LV segmentation along with Long Short-Term Memory Networks (LSTM) for the regression of LVEF, obtaining a dice similarity coefficient of 0.92 and a mean absolute error of 5.736%.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
EFNet: A multitask deep learning network for simultaneous quantification of left ventricle structure and function
Samana Batool,Imtiaz Ahmad Taj,Mubeen Ghafoor +2 more
TL;DR: EFNet is a multitask deep learning network that automates left ventricle structure and function quantification from echocardiogram videos, eliminating manual frame identification and variability, improving cardiovascular diagnosis and management.
1
ViViEchoformer: Deep Video Regressor Predicting Ejection Fraction
Taymaz Akan,Sait Alp,Md. Shenuarin Bhuiyan,Tarek Helmy,A. Wayne Orr,Mohammad Musharraf Uddin Bhuiyan,Steven A. Conrad,John A. Vanchiere,Christopher G. Kevil,Mohammad Alfrad Nobel Bhuiyan +9 more
- 22 Jun 2024
TL;DR: ViViEchoformer accurately predicts ejection fraction from echocardiogram videos, offering a precise and reliable alternative to human evaluation.
1
The Promise of Artificial Intelligence-Assisted Point-of-Care Ultrasonography in Perioperative Care.
Ricardo A. Serrano,Alan M. Smeltz +1 more
TL;DR: The role of point-of-care ultrasonography in the perioperative setting has expanded rapidly over recent years and is integrating artificial intelligence to assist clinicians in optimizing images, identifying anomalies, performing automated measurements and calculations, and facilitating diagnoses.
1
HQMC-CPC: A Hybrid Quantum Multiclass Cardiac Pathologies Classification Integrating a Modified Hardware Efficient Ansatz
D. Shoieb,Ahmed Younes,Sherin M. Youssef,Karma M. Fathalla +3 more
TL;DR: The proposed HQMC-CPC model is able to classify different cardiac pathologies with an average minimum performance gap of 3.19%, and speeds up the testing process by around 60% and 40% compared to the classical classifiers and well-established HEA respectively.
Segmentation-Free Estimation of Left Ventricular Ejection Fraction Using 3D CNN Is Reliable and Improves as Multiple Cardiac MRI Cine Orientations Are Combined
Philippe Germain,A. Labani,Armine Vardazaryan,Nicolas Padoy,Catherine Roy,Soraya El Ghannudi +5 more
TL;DR: This study evaluates the reliability of 3D CNNs in estimating left ventricular ejection fraction from cardiac MRI cine orientations, demonstrating improved accuracy with multiple orientations combined, without the need for contour tracing or segmentation.
References
U-Net: Convolutional Networks for Biomedical Image Segmentation
Olaf Ronneberger,Philipp Fischer,Thomas Brox +2 more
- 05 Oct 2015
TL;DR: Neber et al. as discussed by the authors proposed a network and training strategy that relies on the strong use of data augmentation to use the available annotated samples more efficiently, which can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks.
DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs
TL;DR: This work addresses the task of semantic image segmentation with Deep Learning and proposes atrous spatial pyramid pooling (ASPP), which is proposed to robustly segment objects at multiple scales, and improves the localization of object boundaries by combining methods from DCNNs and probabilistic graphical models.
18.8K
Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
Roberto M. Lang,Luigi P. Badano,Victor Mor-Avi,Jonathan Afilalo,Anderson C. Armstrong,Laura Ernande,Frank A. Flachskampf,Elyse Foster,Steven A. Goldstein,Tatiana Kuznetsova,Patrizio Lancellotti,Denisa Muraru,Michael H. Picard,Ernst Rietzschel,Lawrence G. Rudski,Kirk T. Spencer,Wendy Tsang,Jens-Uwe Voigt +17 more
TL;DR: This document provides updated normal values for all four cardiac chambers, including three-dimensional echocardiography and myocardial deformation, when possible, on the basis of considerably larger numbers of normal subjects, compiled from multiple databases.
15.2K
Heart Disease and Stroke Statistics—2017 Update: A Report From the American Heart Association
Emelia J. Benjamin,Michael J. Blaha,Stephanie E. Chiuve,Mary Cushman,Sandeep R Das,Rajat Deo,Sarah D. de Ferranti,James S. Floyd,Myriam Fornage,Cathleen Gillespie,Carmen R. Isasi,Monik C. Jim'nez,Lori Chaffn Jordan,Suzanne E. Judd,Daniel T. Lackland,Judith H. Lichtman,Lynda D. Lisabeth,Simin Liu,Chris T. Longenecker,Rachel H. Mackey,Kunihiro Matsushita,Dariush Mozaffarian,Michael E. Mussolino,Khurram Nasir,Robert W. Neumar,Latha Palaniappan,Dilip K. Pandey,Ravi R. Thiagarajan,Mathew J. Reeves,Matthew D. Ritchey,Carlos J. Rodriguez,Gregory A. Roth,Wayne D. Rosamond,Comilla Sasson,Amytis Towfghi,Connie W. Tsao,Melanie B. Turner,Salim S. Virani,Jenifer H. Voeks,Joshua Z. Willey,John T. Wilkins,Jason H Y Wu,Heather M. Alger,Sally S. Wong,Paul Muntner +44 more
TL;DR: WRITING GROUP MEMBERS Emelia J. Benjamin, MD, SCM, FAHA Michael J. Reeves, PhD Matthew Ritchey, PT, DPT, OCS, MPH Carlos J. Jiménez, ScD, SM Lori Chaffin Jordan,MD, PhD Suzanne E. Judd, PhD
10.9K
Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentations
Carole H. Sudre,Carole H. Sudre,Wenqi Li,Tom Vercauteren,Sebastien Ourselin,Sebastien Ourselin,M. Jorge Cardoso,M. Jorge Cardoso +7 more
- 14 Sep 2017
TL;DR: In this paper, the authors investigate the behavior of these loss functions and their sensitivity to learning rate tuning in the presence of different rates of label imbalance across 2D and 3D segmentation tasks.
2K