Low Power VLSI based Design and Implementation of MLP-BP Neural Network for Detection of Kidney Stone in Real Time

In recent years one of the most common problems that occur in the human urinary system is kidney stones or urinary stones. There are many Methods of medical imaging that can be used to examine parameters of human kidneys, for example magnetic resonance imaging (MRI), x-ray computed tomography (CT), ultrasound imaging (US), and many others. This detection is very important for the doctor to determine the status of the kidneys and also to visualize any abnormalities present in the kidney [2]. Any person affected with a problem in kidney suffers with a pain in early stage. The detection of abnormalities of kidney inside the body is a main field of study in medical research by bio-medical image processing[1-4], Due to some abnormalities (speckle noise) in ultrasound or MRI images and artifacts, wrong diagnosis may happen by analyzing the scanned image. Therefore in this work the main focus is on development of new hardware implementation based on neural network architecture for detection of kidney stone in real time by optimizing area, power and speed on FPGA [5]. This algorithm implemented on Vertex-II Pro FPGA device and simulated in matlab [9].