Abstract:
The renal sinus is a space that forms the medial border of the kidney and is surrounded by the renal parenchyma laterally. The renal arteries, veins, lymphatic vessels, nerve fibers, renal pelvis and major and minor calices are located within the renal sinus. 50 images were used in the study. The study conducted at radiology department in Royal Care hospital using CT scan 64 slice manufactured by Toshiba. Then the image were read by IDL in TIFF format and the user clicks on areas represents the renal cortex, renal sinus fat and psoas muscle fat area in case of test group; in these areas a window 3×3 pixel were generated and textural feature for the classes center were generated. These textural features includes FOS; (coefficient of variation, stander deviation, variance, signal, energy, and entropy) were used. These features were assigned as classification center using the Euclidian distances to classify the whole image.
The result of the study revealed that the classification accuracy result using linear discriminant function, in which 93.4% of original grouped cases correctly classified. Overall classification accuracy = 93.4%. Sensitivity of renal sinus, kidney tissue and abdominal fat = 87.7%, 100%, and 94.0% respectively. In respect to the applied features the mean, SD, energy and entropy on CT images can differentiate between renal sinus fat and rest of the tissue successfully and the best feature is the mean followed by energy, then entropy and the least is SD.
Texture analysis depending on the relative attenuation coefficient of tissues could serve the diagnostic field and overcoming the visual diagnosis that comes with different interpretation.
