Detection of injured kidney in computed tomography

Abstract

Timely and accurate diagnosis of intraabdominal organ injuries due to trauma is critical. Computer Assisted Detection (CAD) systems are rapidly developing techniques to segment the organs or to detect the pathologies in medical applications; either automatically or semi-automatically. In this work, our aim is to propose and validate a CAD system which classifies injured kidney in Computed Tomography (CT) images. Sixteen cases containing nineteen injured and thirteen intact kidneys were considered for the validation of the method. The classification of the injured kidney was satisfactorily performed with 100% sensitivity ratio.