Inverse imaging of the breast using the conjugate gradient-bivariate material classification technique

Author

Zhang, Xiaodong ; Broschat, Shira L. ; Flynn, Patrick J.

Author_Institution

Dept. of Electr. Eng. & Comput. Sci., Washington State Univ., Pullman, WA, USA

Volume

2

fYear

1999

fDate

1999

Firstpage

1609

Abstract

Presents the bivariate material classification (BMC) technique for ultrasound imaging of the breast. The BMC technique is used in conjunction with the conjugate gradient (CG) method to solve the inverse problem. Results are shown for a one-quarter scale, 2D model of the breast. It is found that the CG-BMC technique increases convergence and, in addition, gives more accurate results. Approximately 30% fewer iterations are required, significantly reducing the computational cost. The CG-BMC algorithm appears to be robust to noise, achieving good results for a signal-to-noise ratio of 50 dB

Keywords

biomedical ultrasonics; image classification; inverse problems; iterative methods; mammography; 2D model; computational cost; conjugate gradient method; conjugate gradient-bivariate material classification technique; inverse breast imaging; medical US imaging; medical diagnostic imaging; signal-to-noise ratio; Acoustic noise; Acoustic scattering; Breast; Character generation; Convergence; Integral equations; Newton method; Noise robustness; Signal to noise ratio; Ultrasonic imaging;

fLanguage

English

Publisher

ieee

Conference_Titel

Ultrasonics Symposium, 1999. Proceedings. 1999 IEEE

Conference_Location

Caesars Tahoe, NV

ISSN

1051-0117

Print_ISBN

0-7803-5722-1

Type

conf

DOI

10.1109/ULTSYM.1999.849303

Filename

849303