Ejection Fraction Estimation from Echocardiograms Using Optimal Left Ventricle Feature Extraction Based on Clinical Methods
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...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-06-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/13/13/2155 |
| _version_ | 1797591877927567360 |
|---|---|
| author | Samana Batool Imtiaz Ahmad Taj Mubeen Ghafoor |
| author_facet | Samana Batool Imtiaz Ahmad Taj Mubeen Ghafoor |
| author_sort | Samana Batool |
| collection | DOAJ |
| description | 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%. |
| first_indexed | 2024-03-11T01:43:39Z |
| format | Article |
| id | doaj.art-0bb81aeab66345c0b481a3d5049a8a31 |
| institution | Directory Open Access Journal |
| issn | 2075-4418 |
| language | English |
| last_indexed | 2024-03-11T01:43:39Z |
| publishDate | 2023-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj.art-0bb81aeab66345c0b481a3d5049a8a312023-11-18T16:20:46ZengMDPI AGDiagnostics2075-44182023-06-011313215510.3390/diagnostics13132155Ejection Fraction Estimation from Echocardiograms Using Optimal Left Ventricle Feature Extraction Based on Clinical MethodsSamana Batool0Imtiaz Ahmad Taj1Mubeen Ghafoor2Electrical Engineering, Capital University of Science and Technology, Islamabad Expressway, Kahuta Road, Islamabad 44000, PakistanElectrical Engineering, Capital University of Science and Technology, Islamabad Expressway, Kahuta Road, Islamabad 44000, PakistanSchool of Computer Science, University of Lincoln, Brayford Way, Brayford, Pool, Lincoln LN6 7TS, UKEchocardiography 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%.https://www.mdpi.com/2075-4418/13/13/2155medical imagingtransthoracic echocardiographyleft ventricle ejection fractionregressionmachine learningSimpson’s biplane method |
| spellingShingle | Samana Batool Imtiaz Ahmad Taj Mubeen Ghafoor Ejection Fraction Estimation from Echocardiograms Using Optimal Left Ventricle Feature Extraction Based on Clinical Methods Diagnostics medical imaging transthoracic echocardiography left ventricle ejection fraction regression machine learning Simpson’s biplane method |
| title | Ejection Fraction Estimation from Echocardiograms Using Optimal Left Ventricle Feature Extraction Based on Clinical Methods |
| title_full | Ejection Fraction Estimation from Echocardiograms Using Optimal Left Ventricle Feature Extraction Based on Clinical Methods |
| title_fullStr | Ejection Fraction Estimation from Echocardiograms Using Optimal Left Ventricle Feature Extraction Based on Clinical Methods |
| title_full_unstemmed | Ejection Fraction Estimation from Echocardiograms Using Optimal Left Ventricle Feature Extraction Based on Clinical Methods |
| title_short | Ejection Fraction Estimation from Echocardiograms Using Optimal Left Ventricle Feature Extraction Based on Clinical Methods |
| title_sort | ejection fraction estimation from echocardiograms using optimal left ventricle feature extraction based on clinical methods |
| topic | medical imaging transthoracic echocardiography left ventricle ejection fraction regression machine learning Simpson’s biplane method |
| url | https://www.mdpi.com/2075-4418/13/13/2155 |
| work_keys_str_mv | AT samanabatool ejectionfractionestimationfromechocardiogramsusingoptimalleftventriclefeatureextractionbasedonclinicalmethods AT imtiazahmadtaj ejectionfractionestimationfromechocardiogramsusingoptimalleftventriclefeatureextractionbasedonclinicalmethods AT mubeenghafoor ejectionfractionestimationfromechocardiogramsusingoptimalleftventriclefeatureextractionbasedonclinicalmethods |