Automated detection of breast tumors in ultrasonic images usingfuzzy reasoning

摘要

This paper proposes a novel method for the automated detection ofbreast tumors in three dimensional ultrasonic images using fuzzyreasoning. The algorithm consists of two stages of processing. In thefirst stage a three dimensional LoG (Laplace of Gaussian) filtering isemployed to generate a membership function for the fuzzy inference. Inthe second stage, each voxel in the three dimensional image isclassified as “tumor”, “normal tissue” or“boundary” by fuzzy reasoning followed by thedefuzzification using relaxation techniques. We apply the method to 10cases of malignant and 10 cases of benign tumors. Both of the two typesof extracted tumors agree well with the manually traced boundaries.Further we define a parameter for evaluation of the uneven complexity ofthe surface roughness for diagnosis of malignant and benign tumors. Theresults show a successful characterization