Main Article Content
In today’s world almost 5.6% of people were affected by the gastric cancer. The right measure at right time has to be taken to prevent the people from severe problem. In this paper we proposed an idea to detect the cancer at early stage most accurately.At present,doctors detect the gastric cancer by endoscopic image of patient manually, whereas in our project the system will automatically detect the disease. Bringing automation in disease detection will increase accuracy and efficiency. For this automation, endoscopy images of stomach where taken and trained up ourneural network. Mask R-CNN is used to segment the lesion place in the image frame whereas the use of Grabcut is for further refinement of segmented lesion formed by Mask R-CNN.