Development of Extraction Technique for Diagnosing Kidney Abnormalities by Using Ultrasound Images

Kidney disease is a health problem consisting of kidney failure, where it is caused by a complication of decreased kidney function. To prevent serious complications, kidney disorders must be diagnosed quickly and automatically. The aim of this study was to develop an extraction method to diagnose kidney abnormalities in ultrasound images from 102 patients. Researchers proposed the development of an extraction technique by using a combination of the Grey Level Co-Occurrence Matrix (GLCM), Scale Invariant Feature Transform (SIFT) method and the Canny method. The statistical-based GLCM feature extraction method used a pixel value matrix that had a certain score and forms the angle of the kidney image pattern. The research shows that the SIFT algorithm is effective for object recognition and it is robust to changes in rotation and scaling of kidney images. Canny is also known as optimal edge detection, localizing edge points and providing only one response for one edge of the image. Based on the tests, it is found that the process of detecting kidney abnormalities quickly and accurately. it increases the efficiency of the medical role in decision making. This automatic system analyses and identifies kidneys based on existing abnormalities with maximum accuracy such as cysts, kidney stones, cancer and tumours as abnormalities in the kidneys. In addition, the automated kidney diagnosis system will help radiologists and nephrologists in reducing inadequate nephrology problems and discrepancies.