Wavelet based texture analysis of Liver tumor from Computed Tomography images for characterization using Linear Vector Quantization Neural Network

The use of medical imaging and tissue characterization techniques is popular in diagnosis, treatment and research. The objective of this work is to automatically extract the liver tumor from the liver region of the CT abdominal image and to characterize the liver tumor as benign or malignant using wavelet based texture analysis and Linear Vector Quantization (LVQ) neural network. The system is tested with 100 images. The accuracy obtained is 92%. Performance of the system for the different parameters of LVQ like learning rate, number of hidden neurons and the number of epochs are analyzed. To evaluate the performance of the system, parameters like sensitivity, specificity, positive predicting value and negative predicting value are calculated. The results are evaluated with radiologists.