Application of artificial neural networks for the classification of liver lesions by image texture parameters.

Ultrasound imaging is a powerful tool for characterizing the state of soft tissues; however, in some cases, where only subtle differences in images are seen as in certain liver lesions such as hemangioma and malignancy, existing B-scan methods are inadequate. More detailed analyses of image texture parameters along with artificial neural networks can be utilized to enhance differentiation. From B-scan ultrasound images, 11 texture parameters comprising of first, second and run length statistics have been obtained for normal, hemangioma and malignant livers. Tissue characterization was then performed using a multilayered backpropagation neural network. The results for 113 cases have been compared with a classification based on discriminant analysis. For linear discriminant analysis, classification accuracy is 79.6% and with neural networks the accuracy is 100%. The present results show that neural networks classify better than discriminant analysis, demonstrating a much potential for clinical application.

[1]  Azriel Rosenfeld,et al.  A Comparative Study of Texture Measures for Terrain Classification , 1975, IEEE Transactions on Systems, Man, and Cybernetics.

[2]  Stephen N. Wiener,et al.  Scintigraphy and Ultrasonography of Hepatic Hemangioma1 , 1979 .

[3]  H. Osawa,et al.  Sonographic diagnosis of fatty liver using a histogram technique that compares liver and renal cortical echo amplitudes , 1996, Journal of clinical ultrasound : JCU.

[4]  A Manduca,et al.  Improvement in specificity of ultrasonography for diagnosis of breast tumors by means of artificial intelligence. , 1992, Medical physics.

[5]  B. Garra,et al.  Improving the Distinction between Benign and Malignant Breast Lesions: The Value of Sonographic Texture Analysis , 1993 .

[6]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[7]  R. Linder,et al.  Neural network approach for computer-assisted interpretation of ultrasound images of the gallbladder. , 1993, European journal of radiology.

[8]  J S Ostrem,et al.  Application of neural nets to ultrasound tissue characterization. , 1991, Ultrasonic imaging.

[9]  Bengt Samuelsson,et al.  Raven Press, New York (1995) , 1996 .

[10]  Richard P. Lippmann,et al.  An introduction to computing with neural nets , 1987 .

[11]  A Lorenz,et al.  Echographic Tissue Characterization in Diffuse Parenchymal Liver Disease: Correlation of Image Structure with Histology , 1990, Ultrasonic imaging.

[12]  James F. Greenleaf,et al.  Use of gray value distribution of run lengths for texture analysis , 1990, Pattern Recognit. Lett..

[13]  J M Thijssen,et al.  Application of Neural Networks for the Classification of Diffuse Liver Disease by Quantitative Echography , 1993, Ultrasonic imaging.

[14]  A Lorenz,et al.  Computerized Ultrasound B-Scan Texture Analysis of Experimental Fatty Liver Disease: Influence of Total Lipid Content and Fat Deposit Distribution , 1990, Ultrasonic imaging.

[15]  R. Brooke Jeffrey,et al.  Sonography of the abdomen , 1995 .

[16]  W R Lees,et al.  Diagnostic ultrasound in gastroenterology , 1985 .

[17]  D Zatari,et al.  In vivo liver differentiation by ultrasound using an artificial neural network. , 1994, The Journal of the Acoustical Society of America.

[18]  S. Robbins,et al.  Pathologic basis of disease , 1974 .