Published in:
Japanese Journal of Applied Physics, 55 (2016) 7S1, Seite 07KF09
Language:
Not determined
DOI:
10.7567/jjap.55.07kf09
ISSN:
0021-4922;
1347-4065
Origination:
Footnote:
Description:
Abstract A diagnosis of liver fibrosis using an ultrasound B-mode image has the advantages of real-time observation and noninvasive properties. In our previous study, a multi-Rayleigh model was proposed to express a probability density function (PDF) of echo amplitudes from a fibrotic liver. From the multi-Rayleigh model, fibrosis parameters, such as the amount of fibrotic tissue and its progressive ratio, can be extracted. To quantitatively evaluate liver fibrosis using the multi-Rayleigh model, it is important to evaluate the stability of the estimation method of multi-Rayleigh model parameters. In this study, a numerical simulation using random variables following the multi-Rayleigh model was performed and the estimation stability of the parameters of the multi-Rayleigh model with two components was examined. From the simulation results, it was found that estimation becomes unstable under a certain condition owing to statistical variations of moments, which are inputs in the estimation algorithm. The instability of estimated parameters could be evaluated by focusing on changes in moments upon changes in multi-Rayleigh model parameters. It was indicated that we can evaluate the reliability of the estimated parameters of the multi-Rayleigh model only from the estimated values.