Detection of Tumor in Liver Using Image Segmentation and Registration Technique

Now-a-days, there is rise in the death rate of patients suffering from liver cancer. The liver cancer rate is increasing year by year. Generally, liver cancer’s death rate is very high because the disease causes no symptoms, so it’s often not caught until it's in final stages. The Canadian Cancer Society says, if we catch someone's liver cancer early, their chance of defeating the disease is 70 to 80 per cent. If the disease is caught late, the average person survives about a year after diagnosis. We propose an algorithm for liver cancer detection which is based on concepts of fuzzy logic and neural network. Neuro-fuzzy (NF) systems are suitable tools to deal with uncertainty found in the process of extracting useful information from images. In this work, the liver tumor is detected through the medical images in three phases, pre-processing phase, processing phase and detection phase. Initially in the pre-processing phase, a set of medical images is filtered for removing noise. Then the filtered image is segmented automatically using fuzzy logic, neural network and windowing technique. In the detection phase neuro-fuzzified segmented images of CT and MRI is registered to obtain the tumor. The result is obtained for few different set of database.