Automatic determination of small bowel phantom length in computed tomography images

In this study, processes on 4 different colors and lengths of pipe images of computerized tomography images were done in Radiology Department of Health Sciences University Gülhane Medical Faculty. Intestinal length knowledge is needed in bowel diseases that are shortened by various causes such as short bowel syndrome, bowel insufficiency. There is no study about length of the small intestine in the literature. Very little work has been done about small intestinal segmentation. It is aimed to find the length of the small intestinal phantoms by adding a different direction. First, pre-processing was done by applying morphological processes to the image. Center points of image which has been converted to binary using thresholding methods have also been determined by using region growing method. By writing the equation created in the second order of the center points, the correct ordering of the center points was obtained by finding the errors in each cycle and ordering the minimum ones. Thus, by calculating and summing the distances between the center points, the length of the small intestinal phantom was found. This study first was applied the phantom with the shortest length and then by applying to the other three a result with high accuracy was obtained .This thesis study has an important place in terms of automatic finding of intestinal length and ease of doctors without too much intervention in the human body. Keywords: Small intestinal segmentation, short bowel syndrome, thresholding, intestinal insufficiency, computed tomography