The Research About Imagery Processing System Based On BP Nerve Network Of Lung Cancer Cell

With the development of computer technology, computer-aided medical surgical techniques by people's attention, medical microscopic image processing application is much concern from the public. Digital image processing, pattern recognition and expert system technology in the biomedical field gaining increasing use of computer technology and expert medical knowledge and experience by combining the characteristics used in lung cancer cells from the areas of knowledge that the rules can be effectively assisted experts for diagnosis. According to the development trend of modern medicine, there is an urgent need for us to the traditional optical image displayed on the computer screen, then the observation cell parameters were analyzed statistics, a medical doctor-assisted diagnosis.Cancer experts predict that in the next 20 to 30 years, the incidence of lung cancer will continue to rise sharply, and China in the near future to become the highest incidence and death rates of malignant tumors. Early diagnosis for lung cancer patients is the key. Early diagnosis of lung cancer to issue the medical profession has made some research, and have achieved some tangible results.However, because of the many types of lung cancer, and a variety of related factors, making existing diagnostic accuracy and practicality in all aspects there are considerable limitations, such as modeling complex and difficult, fault-tolerant capability is not strong, the scope of application is not extensive; Incomplete data on these incomplete data are usually difficult to handle .In recent years, with the medical imaging technology rapid development of a real-time ultrasound imaging (us), computer tomography (CT), magnetic resonance imaging (MRI), vascular create imaging of hepatocellular carcinoma, radionuclide imaging, interventional, and other effects of technology, special other comprehensive diagnostic imaging is the application of the success of the early diagnosis of lung cancer rate has been greatly increased. On the other hand, computer image processing projects in meteorology, remote sensing technology, medical diagnosis and biological engineering, and other fields access a wide range of applications, and have received very good results. Cell Image processing is a computer image processing specific applications, including cell image by significant technology, and cellular technologies and image transform cell image analysis and identification technology.Countries "Eighth Five" technology key topic in the lung cancer cells automatic identification Color Image Processing System basic research, the lung cancer cells to resolve the feasibility of automatic identification problems, in order to further develop lung cancer cells automatic identification color image processing system, laid the foundation for. This basic research for the automatic identification of cancer cells to promote and develop research, laid the foundation so that it has feasibility. At the same time, for further study provides a new research content. Lung cancer cells color image processing system based on this technology is the environment, is of great scientific value and good social benefits.Pathology experts rely mainly on the current stage through the naked eye on the cell pathology slice images were observed and estimated. On the one hand, experienced fewer number of pathological experts, on the other hand due to fatigue, and other factors, the diagnosis of pathological experts will be subjective factors interference affected the diagnosis. Lung cancer cells Color Image Processing System study is the early lung cancer screening, diagnosis and effective way is to be solved in the current issue. A digital image processing, pattern recognition and neural network-based computer-aided diagnosis of attention gradually carried out effective research. Using computer image processing and analysis, can effectively help doctors and other diseases such as a tumor diagnosis. On the one hand, computers can replace people to do those time-consuming, boring repetitive counting work, raise work efficiency; On the other hand, identify cancer cells, sometimes need to draw quantitative results, the human eye can hardly capable of such work, and using computer image processing and pattern recognition technology to complete microsurgical Image analysis and identification has made great progress.At the same time, lung cancer cells Color Image Processing System will study artificial intelligence, pattern recognition theory and expert medical knowledge and experience, applied cancer cell recognition, shape, chromatin, and other features in image processing can be extracted, according to the characteristics of cancer cells extracted, which can be identified is true Recognition classification rules, so that we can work for image recognition.This paper is the analysis of mathematical morphology on the basis of the principle against lung cancer cells images fuzzy unclear and uncertain features, the use of statistical pattern recognition Fisher Linear judgment, dealt with fairly good image segmentation lung cancer cells. In the list of the medical basis for lung cancer cell analysis on the basis of the analysis focused on the morphological analysis methods: 8 - chain code tracking algorithm and direction code algorithm. In this paper the final, the lung cancer cells from the five shape characteristic values and adopted a three-tier approach is fully integrated Bp neural network model to establish the diagnosis, after training, network meet the training objectives.Through the integrated use of computer image processing technology, BP neural network technology and the statistical model identification method, the cell lung cancer diagnosis and image processing technologies for thefurther study.