| Autor: |
Khandait, P. D., Ingole, Kartik, Khandait, S. P. |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 2024, Vol. 2974 Issue 1, p1-5, 5p |
| Abstrakt: |
Echocardiography has an important role to play in the diagnosis of cardiac disease, namely in the detection of intracardiac masses. In order to improve the diagnostic accuracy of echocardiography, a unique fully automated classification approach based on sparse representation is proposed. Before calculating the mass area, a region of interest is first sliced. Then, the speckle is removed while the underlying anatomical structure is preserved with the use of a cutting-edge global denoising technique. Next, the K-singular value decomposition and an updated active contour model are used to depict the mass's contour in relation to the atrial wall. Finally, a sparse representation classifier is used to the motion, boundary, and texture data in order to distinguish between the two masses. Ninety-seven echocardiogram sequences from clinical practice are gathered for analysis. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
