Dissertation > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device

Research and Implementation of CAD on the Pathological Slices of Gastric Adenocarcinoma

Author CaiHaiYang
Tutor ZhouLiHua
School Xi'an University of Electronic Science and Technology
Course Computer System Architecture
Keywords Threshold segmentation Nucleus segmentation Watershed transform GVF Snake model Discriminant analysis
CLC TP391.41
Type Master's thesis
Year 2008
Downloads 20
Quotes 0
Download Dissertation

Through the computer, medical image processing and analysis is an important application of image processing technology, combined with the characteristics of medical imaging . The computer-aided diagnosis system for medical research and clinical diagnosis , especially of great significance in the case of lack of experts . Existing computer-aided diagnosis of cancer are basically smears , for the most common pathological cell image , its organizational complexity , difficulty in cell division , has been little research . Gastric adenocarcinoma biopsy computer-aided diagnosis system . The composite cell image pre - processing technology, the composition of the nucleus contour segmentation algorithm and nuclear feature extraction and discrimination is the main work of this paper . Image preprocessing stage , according to the features of complex tissue biopsy of gastric cancer cell image analysis of pretreatment technology , composite pretreatment technology based on threshold segmentation to eliminate the impact of other organizations . In the nucleus segmentation stage , the combination of algorithms based on the the watershed transform the the ( watershed transformation ) GVFSnake model ( GVF Snake model ) . Watershed transform the initial contour of the nucleus , and then use GVF Snake model approximation of the true contours . In the nucleus feature extraction and determination phase , to give up the idea of a common , try to use statistically Fisher discriminant analysis of large sample data . Iterative screening and solving the characteristic parameter coefficient characteristic parameters , the eventual establishment of a cell sorting Fisher discriminant function expression . Experimental results show that this paper based on the improved gastric adenocarcinoma biopsy established computer-aided diagnosis system with the ability of cell division and better cell classification and discrimination ability to play a role in the pathological diagnosis for improving the reading level of junior doctors has a certain value . In this study, in theory enrich the contents of the computer-aided diagnosis has laid a good foundation for further study .

Related Dissertations
More Dissertations