New M.E. Thesis Submitted from cse Student


Cancer is a genetic and vital disease. In last decade, many important genes responsible for the creation of various cancers have been discovered, their mutations precisely identified and the pathways through which they act are been characterized. Mammography is the most common technique used by radiologists in the screening and diagnosis of the cancer cells. Although it is seen as the best examination technique early detection of cancer cells but it provides a low accuracy in early prediction in brain . Their interpretation requires skill and experience for proper diagnosis. Computer aided diagnosis systems for detecting malignant texture have been investigated using several techniques. The existing computer aided diagnoses are generally categorized into two main categories namely tumor mass detection and Cluster micro classification.This thesis works presents an extension in computer-aided diagnosis for early prediction of cancer cells in brain using Texture features and neuro classification logic. First the extractions of the texture from the given brain MRI sample is done next is morphological operation followed by neuro classification for prediction of Cancer for a given sample. Results are presented on the images from brain cancer database. The proposed method clearly classifies the different class of brain cancer. On the basis of experimental result it shows that proposed technique using neuro fuzzy classifier is better than fuzzy classifier in predicting the accuracy of cancer. It acts as a second opinion to assist the radiologist in effective diagnosis of abnormality due to tumors in brain.

Leisure Readings :