IDENTIFICATION OF BOWEL CANCER BY RNN THEORM USING HISTOPATHOLOGICAL IMAGES

Abstract

IDENTIFICATION OF BOWEL CANCER BY RNN THEORM USING HISTOPATHOLOGICAL IMAGES ABSTRACT Colorectal cancer is a one of the highest causes of cancer – related death, especially in men. polyps are the main causes of colorectal cancer and early detection of polyps by RNN and Navie Bayes theorem using histopathological images could result in successful treatment. Diagnosis of polyps using histopathological images is a challenging task due to variation of size and shape of polyps. Performance of the method is enhanced by two ways based on RNN & Navie Bayes theorem and histopathological images and CT scans. First, we perform the computer tomography method to scan the polyp. And second, we perform the histopathological images method that will check with microscopic images and compare and execute the results. By using this method, the annual death rate by colon cancer is reduced by 16%.

Specification

Description:DESCRIPTIONS:
Cancer is now the biggest cause of death in the world. There different types of the cancer disease are affected to the human body like skin cancer, blood cancer etc. Among all these cancers the colon cancer is deadly occurring men and women. To reduce these death rates, occur because of cancer we need some methodology to detect the cancer in its early stage. Not only the detection of disease in its early stage may reduce the death rate but also proper diet and doctor's treatment help in it. Cancer cells can be carried away from the colon in blood, or polyp fluid that surrounds intestine tissue. Polyp flows through polypatic vessels, which drain into polyp nodes located in the colon and in the centre of the large intestine. Colon cancer often spreads toward the centre of the intestine because the natural flow of polyp out of the bowel is toward the center of the intestine. Eating fast food and consuming alcohol is the main reason of causing colon cancer. With changing life style, the people in age group of 18 to 30 consume alcohol and fast food regularly and because of that the percentage of colon cancer patients in this age group is high. The rank order of cancers for both males and females among Jordanians in 2008 indicated that there were 356 cases of colon cancer accounting for (7.7 %) of all newly diagnosed cancer cases in 2008. Colon cancer affected 297 (13.1 %) males and 59 (2.5%) females with a male to female ratio of 5:1 which colon cancer ranked fourth among males and 15th among females. Presently, CT are said to be more effective than plain abdomen x-ray in detecting and diagnosing the colon cancer. Colon cancer is diagnosed from the CT image of colon. Normally a doctor analyses the CT image of abdomen and detect the presence of cancer in colon. In this manual diagnose method may have the chance of false detection. False detection is due to the presence of normal polyp in bowel, and blood vessels etc. So, it is essential to develop a method of computerized detection of cancer. Image processing techniques are the best tool for developing such a computerized method for lung cancer detection. CT image of colon is processed and finds whether the presence of cancer nodule is present or not. For this purpose, there are many image processing tools are used. Colon cancer detection system contains four basic stages. The first stage starts with taking a collection of CT images. The second stage applies several techniques of image enhancement, to get best level of quality and clearness. The third stage applies image segmentation algorithms which play an effective rule in image processing stages, and the fourth stage obtains the general features from enhanced segmented image which gives indicators of normality or abnormality of images. Polyp Segmentation in Colonoscopy Images Using fully Convolutional Network In this method the segmentation of polyp is done by FCN-8S network and thresholding. They used Median filter as noise removal for Image Pre-processing, Histogram Equalization for Image Enhancement, Region Growing is used for image segmentation then the feature extraction gives region of interest of tumor. At the end they describe the various classification techniques for detecting colon cancer. In existing method, they use Convolutional neural network with the FCN-8S and OTSU Thresholding for segmentation and classification Of the polyp. Also, broad images are does not work well in CNN segmentation. It is mostly High noise sensitive. High boost filter which is used for Enhancement technique is used only for High frequency component. In our method using the combination of RNN & Navie Bayes theorem and histopathological images the result accuracy is 96%. We will compare the CT scanned images and histopathological images together and the output is executed. Here, Histopathological image equalization is used for the image enhancement. Region Growing method is used for Image segmentation. The third and essential step in system is feature extraction to recognize the shape and size of nodule present in image using LDRN.
, Claims:CLAIMS:
1. Cancer is the largely occurred disease in the world. From a variety of cancer disease colon cancer is mainly found in men and rarely in women.
2. In medical exercise, some iterative alteration of parameters is inevitable: the notion of clinical acceptability which varies from clinic to clinic or even planner to planner is extremely difficult to pose either as an objective function or a solid restriction. Only early detection of disease may help to decrease the death rate due to the colon cancer.
3. Image processing techniques are widely used in several medical areas for image improvement in earlier detection and treatment stages, where the time factor is very important to discover the abnormality issues in target images.
4. The CT image is processed and the region of interest Image quality and accuracy is the core factors of this research, image quality assessment as well as enhancement stage where were adopted on low pre-processing Practices and techniques based on Median filter.
5. The proposed technique is efficient for segmentation principles to be a region of interest foundation for feature extraction obtaining. The proposed technique gives very promising results comparing with other used techniques. Relying on general features, a normality comparison is made.
6. The main detected features for accurate images comparison are pixels percentage and mask-labelling with high accuracy and robust operation

Documents