Title :
Tissue characterization of the prostate using Kohonen-maps
Author :
Schmitz, G. ; Ermert, H. ; Senge, T.
fDate :
Oct. 31 1994-Nov. 3 1994
Abstract :
Although transrectal ultrasound is one of the most important tools in the diagnosis and early detection of prostatic cancer, the sensitivity and specificity of the standard sonographic methods are still insufficient. We describe a method which provides the clinician with additional information in the form of color-coded tissue characterization images based on the learning vector quantization (LVQ) algorithm proposed by Kohonen
Keywords :
biomedical ultrasonics; image classification; image colour analysis; medical image processing; self-organising feature maps; vector quantisation; Kohonen-maps; clinician; color-coded tissue characterization images; diagnosis; early detection; learning vector quantization algorithm; prostate; prostatic cancer; standard sonographic methods; tissue characterization; transrectal ultrasound; Biological tissues; Biomedical acoustic imaging; Biomedical signal processing; Image classification; Image color analysis; Neural network applications; Reproductive system; Self-organizing feature maps; Vector quantization;
Conference_Titel :
Ultrasonics Symposium, 1994. Proceedings., 1994 IEEE
Conference_Location :
Cannes, France
Print_ISBN :
0-7803-2012-3
DOI :
10.1109/ULTSYM.1994.401872
