Edge detection and feature extraction of gastrointestinal radiographs using adaptive differential filter and level set method

To improve the detectability of the gastric lesions in the X-ray examinations, the computer image processing methods of radiographs of gastric cancer lesions are discussed. We propose an adaptive differential filter to the edge detection. Moreover, the level set method is adopted to the segmentation. To verify the effectiveness of the proposed method, the numerical examples are carried out, using the case radiograph of the gastric intestines medical picture database produced by National Kyushu Cancer Center. The numerical results show that the number of iterations in the level set method can be reduced using the binary image obtained by the adaptive differential filter.