3-D Bayesian ultrasound breast image segmentation using the EM/MPM algorithm

L.A. Christopher1, E.J. Delp1, C.R. Meyer2, P.L. Carson2
1Video and Image Processing Laboratory, School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, USA
2Department of Radiology, University of Michigan, Ann Arbor, MI, USA

Tóm tắt

In this paper, ultrasound breast image segmentation is improved by using the volumetric data available in neighboring slices. The new algorithm extends the EM/MPM framework to 3D by including pixels from neighboring frames in the Markov Random Field (MRF) clique. In addition, this paper describes a unique linear cost factor introduced in the optimization loop to compensate for the attenuation common to ultrasound images.

Từ khóa

#Bayesian methods #Ultrasonic imaging #Breast #Image segmentation #Markov random fields #Magnetic resonance imaging #Acoustic noise #Image processing #Cost function #Attenuation

Tài liệu tham khảo

10.1016/S0929-8266(98)00062-7 10.1109/83.869185 10.2307/2289127 10.1109/58.895909 10.1109/42.650884 10.1109/42.650883 moon, 1999, The Expectation-Maximization algorithm, IEEE Signal Processing Magazine, 47 10.1016/0301-5629(95)00018-M 10.1016/S1361-8415(97)85010-4 10.1109/TPAMI.1984.4767596 besag, 1974, Spatial interaction and the statistical analysis of lattice systems, J R Stat Soc B, 36, 192 10.1109/83.913586 10.1109/83.941855 10.1016/S0301-5629(00)00286-6 10.1109/58.981389 10.1109/10.951522