Abstract
This paper focuses on characteristics of RF signal respectively on time domain and frequency domain, and applies the pattern recognition technology to grade fatty liver. Firstly, selected the regions of interest (ROI) of RF signal. Secondly, get some feasures, such as the mathematical expectation (PM), the multi-fractal spectrum width (MSW), the low-frequency wavelet coefficients mean (LWCM), and the wavelet modulus maximum mean (WMMM) are extracted from ROI area. Then, C-means clustering algorithm and backpropagation (BP) neural network are employed respectively to classify. In the experiment, 200 patient's liver RF signals are proceeded. These RF signals are four different kinds of liver, like normal liver, mild fatty liver, moderate liver, and severe fatty liver, consisting of 50 signals of each kind. Liver RF signals are be classified by some feature vectors that are some weighted combination of PM, MSW, LWCM, WMMM parameters. There, C-means clustering algorithm was used to find the best feature vector that just includes PM, LWCM, WMMM. Then, the best feature vector is served in the BP neural network. As a result of research, the accuracy rate of normal liver is 92.0%, mild fatty liver is 88.0%, moderate fatty liver is 90.0%, and 94.0% is for severe fatty liver. This study shows that the ultrasonic RF signal is effective in diagnosing liver fatty degree, and this paper provides a new method for computer-aided diagnosis of fatty liver.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.