Article Preview
Top1. Introduction
Medical image processing contributes a lot to the process of diagnosis of diseases and also to the subsequent appropriate medical treatment. It facilitates the doctor’s effort to see the interior portion of the body in order to pinpoint the cause of a disease. Medical image processing helps the physician to diagnose the cause precisely or identify the affected portion of the patient’s anatomy. The primary purpose of medical image processing is to observe the objects that are not seen by the naked eye, distinguish and then create a better image. Imaging methods like Magnetic Resonance Imaging (MRI), X-ray, Computed Tomography, Endoscopy Image, Ultrasound, etc. have led to the use of many image processing techniques today in the medical field.
Endoscopy, also known as Gastroscopy, is widely used now for diagnosing certain diseases in hospitals. It helps in identifying a health problem called the irritable bowel syndrome. When the camera has been inserted into the stomach via the esophagus, the doctor will then examine and possibly photograph or take biopsies of the patient’s food-pipe, stomach and/or small intestine for any signs of the disease. The list of possible findings includes irritation of the stomach lining, gastric or duodenal ulcers, infection in the small bowel, inflammation and pre-cancerous conditions. Rarely can cancer also be diagnosed incidentally.
Irritable Bowel Syndrome is an acute intestinal disorder that effects changes in the bowel system. Identifying this disorder is not easy just because it does not cause any significant damage in the bowel pattern. In general, Irritable Bowel Syndrome causes cramping, constipation, bleeding, abdominal pain, bloating, gas, diarrhea, celiac disease, gastric cancer, ulcer, etc. The incidence of this disorder is more in South America and less in Southeast Asia. Both men and women before the age of 45 are mostly susceptible to this disorder. This study has taken up four major issues, namely bleeding, celiac disease, gastric cancer and ulcer with a view to identifying the disease using an improved imaging technique proposed and experimented in this project.
Gastrointestinal bleeding, GI bleeding in short, is also known as gastrointestinal hemorrhage. It refers to various forms of bleeding in the gastrointestinal tract from the mouth to the rectum. Some of the color-based features (Brzeski et al., 2013) that are used to detect bleeding are HSV histogram based on dominant colors, Local HSV histograms, RGB ratio color values, HSV color space and HSI color spaces. Celiac disease is a common disorder that affects genetically predisposed individuals on ingestion of gluten. Some of the features used to detect celiac disease are color-based, texture-based and spatial-based. Those features include shape curvature histogram, frame-to-frame pixel brightness, dominant period analysis and local binary pattern.
Figure 1.
The Endoscopic images of bleeding, cancer, celiac disease and ulcer
In the disease called Gastric (stomach) cancer, malignant cells form in the lining of the stomach in Figure 1. Some of the features used to detect gastric cancer are color-based, texture-based and shape-based, and they are color contrast, HSV color space, grey level co-occurrence matrix, GLRLM, invariant moments, intensity histogram, edge detection and shape curvature histogram. Another disease known as peptic ulcer refers to painful sores or ulcers in the lining of the stomach or the first segment of the small intestine. Some of the features used to detect a gastric ulcer are color-based, texture-based and shape-based, and they are RGB color space, HSV color space and color coherence vector, gray level co-occurrence matrix, energy, contrast, entropy, correlation, homogeneity and inertia.
The method adopted in this study consists of five stages of the image processing technique to detect Irritable Bowel Syndrome: 1. Image Acquisition, 2. Image Pre-processing, 3. Image Segmentation, 4. Feature Extraction and 5. Image Classification.