Texture-Based Retrieval of Thyroid Gland SPECT Image

With the development of computer science and the appearance of a great number of medical imaging equipments, large amounts of medical image data is produced in hospitals everyday. More and more medical images represent huge quantities of data that need to be safely stored, automatically processed and retrieved in an intelligent way. Thus people raised that Content-based Image Retrieval (CBIR) technology could be applied to medicine. The main idea of CBIR is to establish image feature vectors as their index which based on color, texture, shape and other information contained in images. Since most of medical images are gray images, the general color-based retrieval method is incapable, texture-based and shape-based retrieval methods are widely adopted. This article introduces how the texture feather of a thyroid gland SPECT image is gained by means of gray level co-occurrence matrix, and how the images from database which reveal the same disease as retrieval image be retrieved.