An Improved Expectation-Maximization Algorithm to Detect Liver Image Boundary in CT Scan Images.

Autor:	Siri, Sangeeta K., Kumar S., Pramod, V. Latte, Mrityunjaya
Předmět:	COMPUTED tomography COMPUTER-aided diagnosis LIVER EXPECTATION-maximization algorithms DIAGNOSIS
Zdroj:	IETE Journal of Research; Oct2023, Vol. 69 Issue 10, p6846-6854, 9p
Abstrakt:	Liver segmentation is a prolific and important area of research that has been deeply studied for the last three decades. Its prominence is increasing in modern Computer-Aided disease Diagnosis (CAD) to deal with a huge amount of images. In this paper, the CT scan image is resized to 256 X 256, and noise is reduced by a median filter, and then local peaks are acquired. The optimal clusters (k) to be formed by Expectation-Maximization (EM) algorithm are obtained by setting the distance between local peaks and height greater than 5. Formulate k number of clusters using the EM algorithm. Crop random section of liver and obtain all the local peaks greater than average of local peaks. This provides the minimum and maximum threshold values using which a threshold-based segmentation is performed. The anticipated algorithm that is verified on 55 CT scan images offers promising results. The experimental outcomes are compared with the existing cluster-based liver segmentation algorithms. [ABSTRACT FROM AUTHOR]
Databáze:	Complementary Index
Externí odkaz:	Zobrazit plný text záznamu Plný text ve formátu HTML
Nepřihlášeným uživatelům se plný text nezobrazuje	K zobrazení výsledku je třeba se přihlásit.