Usefulness of textural analysis as a tool for noninvasive liver fibrosis staging
- 174 Downloads
Noninvasive diagnosis of liver fibrosis is a popular topic in the medical literature. Textural analysis on B-mode ultrasound is viewed as a noninvasive tool for fibrosis staging. A liver tissue model is proposed and used to simulate ultrasound images.
One hundred and twenty-five patients with chronic hepatitis C were included in this study. Patients were investigated using B-mode ultrasound and liver biopsy (Metavir scoring). A texture analysis tool consisting of 12 algorithms and a logistic regression classifier was implemented and validated. Tissue model parameters were varied and ultrasound images were generated.
Texture analysis can discriminate between stages F0 and F4 using actual patient data (accuracy 69.5%) and synthetic images (accuracy 76.6%). A human expert is less sensitive than texture analysis in discriminating subtle changes in ultrasound images. High fibrosis detection accuracies are correlated with larger differences in portal space density (r 2 = 0.5). Accuracies measured when we varied only the fibrosis stage and kept the rest of the tissue parameters constant showed high detection rates only in a narrow parameter interval.
The texture analysis system shows limited performance in staging fibrosis and it cannot be used for accurate monitoring of fibrosis evolution over time.
KeywordsTissue model Fibrosis staging Noninvasive diagnosis Texture analysis
Part of this work was funded by the National Council for Scientific Research in Higher Education Grant No. 41-071/2007: SONOFIBROCAST.
- 1.Oosterveld B, Thijssen M, Romijnt R. Ultrasound attenuation and texture analysis of diffuse liver disease: methods and preliminary results. Phys Med. 1991;36:1034–64.Google Scholar
- 2.Fujii Y, Taniguchi N, Wang Y, et al. Clinical application of a new method that segments the region of interest into multiple layers for RF amplitude histogram analysis in the cirrhotic liver. J Med Ultrason. 2004;31:91–8.Google Scholar
- 5.Li GD, Min LQ, Zang HY, et al. Some information of B-scan image that detected by CNN. In: Proceedings of the 2006 IEEE international conference on information acquisition, vol 1 and 2; 2006. p. 819–23.Google Scholar
- 10.Jeong JW, Lee S, Lee JW, et al. The echotextural characteristics for the diagnosis of the liver cirrhosis using the sonographic images. Proc Annu Int IEEE Eng Med Biol Soc. 2007;2007:1343–5.Google Scholar
- 12.Ahmadian A, Mostafa A, Abolhassani MD, et al. A texture classification method for diffused liver diseases using Gabor wavelets. Proc Annu Int IEEE Eng Med Biol Soc. 2005;2005:1567–70.Google Scholar
- 18.Jensen J, Munk P. Computer phantoms for simulating ultrasound B-mode and cfm images. Acoust Imaging. 1997;23:75–80.Google Scholar
- 23.[No authors listed]. Intraobserver and interobserver variations in liver biopsy interpretation in patients with chronic hepatitis C. The French METAVIR Cooperative Study Group. Hepatology. 1994;20:15–20.Google Scholar
- 24.Cao G, Shi P, Hu B. Ultrasonic liver characterization using phase congruency. Proc Annu Int IEEE Eng Med Biol Soc. 2005;2005:6356–9.Google Scholar
- 25.Cao GT, Shi PF, Hu B. Liver fibrosis identification based on ultrasound images. Proc Annu Int IEEE Eng Med Biol Soc. 2005;2005:6317–20.Google Scholar
- 29.Jan J. Medical image processing, reconstruction, and restoration: concepts and methods. Boca Raton: CRC; 2006.Google Scholar
- 30.De Berg M, Cheong O, Van Kreveld M, et al. Delaunay triangulations. Computational geometry: algorithms and applications. New York: Springer; 2008.Google Scholar
- 33.Materka A, Strzelecki M. Texture analysis methods: a review. Lodz: Technical University of Lodz, Institute of Electronics; 1998.Google Scholar
- 36.Lee W, Hsien K, Chen Y. A study of ultrasonic liver images classification with artificial neural networks based on fractal geometry and multiresolution analysis. Biomed Eng Appl Basis Commun. 2004;16:17–25.Google Scholar
- 38.Friedman J, Hastie T, Tibshirani R. The elements of statistical learning: data mining, inference, and prediction. Berlin: Springer; 2001.Google Scholar
- 41.Grandvalet Y, Bengio Y. Hypothesis testing for cross-validation. Technical report 1285. Montreal: Universite de Montreal; 2006.Google Scholar
- 42.Bengio Y, Grandvalet Y. No unbiased estimator of the variance of k-fold cross-validation. J Mach Learn Res. 2004;5:1089–105.Google Scholar
- 43.Brodatz P. Textures: a photographic album for artists and designers. New York: Dover; 1966.Google Scholar
- 44.Witten I, Frank E. Data mining: practical machine learning tools and techniques. Menlo Park: Morgan Kaufmann; 2005.Google Scholar
- 45.Reichl T, Passenger J, Acosta O, et al. Ultrasound goes GPU: real-time simulation using CUDA. Proc SPIE. 2009;7261;726116. doi: 10.1117/12.812486