Adaptive Feature Selection for Ultrasound Image Processing and its Application to Liver Cirrhosis Detection

Abstract:

An approach to improve the regions of interesting (ROIs) selection accuracy automatically for medical images is proposed. The aim of the study is to select the most interesting regions of image features that good for diffuse objects detection or classification. We use the AHP (Analytic Hierarchy Process) to obtain physicians high-level diagnosis vectors and are clustered using the well-known K-Means clustering algorithm. The system also automatically extracts low-level image features for improving to detect liver diseases from ultrasound images. The weights of low-level features are adaptively updated according the feature variances in the class. Finally, the high-level diagnosis decision is made based on the high-level diagnosis vectors for the top K near neighbors from the medical experts classified database. Experimental results show the effectiveness of the system.

• Periodical: Applied Mechanics and Materials (Volumes 411-414)
• Pages: 1372-1376
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.411-414.1372
• Online since: September 2013
• Authors: Wei Tin Lin, Shyi Chyi Cheng, Chih Lang Lin, Chen Kuei Yang
• Keywords: Clustering, Feature Selection, Object Classification, Region of Interest, ROI

References

[1] YJ Cjoi, SK Baik, DH Park, et al. Comparision of Doppler Ultrasonograpgy and the Hepatic Venous pressure gradient in assessing portal hypertension in liver cirrhosis, J Gastroenterol Hepatol, vol. 18 (2003), pp.424-429.

DOI: 10.1046/j.1440-1746.2003.02992.x

[2] K Ishak, A Baptista, L Bianchi, et al. Histological grading and staging of chronic hepatitis. J Hepatol vol. 22 (1995), pp.696-699.

[3] SK Bailk, JW Kim, HS Kim, et al. Recent variceal bleeding: DUS hepatic vein waveform in assessment of severity of portal hypertension and vasoactive drug response, Radiology vol. 240 (2006), pp.574-580.

DOI: 10.1148/radiol.2402051142

[4] P Taourel, P Blanc, M Bolohnesi, et al. Doppler study of mesenteric, hepatic, and portal circulation in alcoholic cirrhosis: relationship between quantitative Doppler measurements and the severity of portal hypertension and hepatic failure, Hepatology, vol. 28 (1998).

DOI: 10.1002/hep.510280406

[5] TC Chou, SC Cheng, Design and Implementation of A Semantic Image Classification and Retrieval of Organization Memory Information System Using Analytical Hierarchy Process, Omega: The Int. J of Management Science, vol. 34 (2006) pp.125-134.

DOI: 10.1016/j.omega.2004.08.004

[6] TL Stay, The Analytic Hierarchy Process, McGraw-Hill, New York (1980).

[7] L Fürst, S Fidler, A Leonardis, Selecting features for object detection using an AdaBoost -compatible evaluation function. Pattern Recognition Letters vol. 29 (2008) pp.1603-1612.

DOI: 10.1016/j.patrec.2008.03.020

[8] CC Chang, CJ Lin, LIBSVM : a library for support vector machines. ACM Transactions on Intelligent Systems and Technology vol. 2 (2011) pp.1-27.

DOI: 10.1145/1961189.1961199

[9] ED Mitrea, S Nedevschi, B Fratila, et al. Exploring texture-based parameters for noninvasive detection of diffuse liver diseases and liver cancer from ultrasound images. Proc. of the 8th WSEAS Int. Conf. on Mathematical Methods and Computational Techniques in Electrical Engineering (2006).

[10] Nobuyuki Otsu, A Tlreshold Selection Method from Gray-Level Histograms, IEEE Trans. on Systems, man, and Cybernetic, vol. 9-1 (1979) pp.62-66.

DOI: 10.1109/tsmc.1979.4310076