KIDNEY DISEASE PREDICTION USING CONVOLUTIONAL NEURAL NETWORK ALGORITHM
Mr. SATHEESHKUMAR K, MCA, M.Phil,
PUGAZHINI R, ARSHINI B.S, SWATHI T
Department of Computer Science and Engineering
University College Of Engineering, Thirukkuvalai
(A constituent College Of Anna University::Chennai and Approved by AICTE, New Delhi)
-------------------------------------------------------------------------------------------------------------------
ABSTRACT
Renal calculus, more commonly known as kidney disease formation, is characterized by the formation of crystals in the urine caused by substance concentration or genetic susceptibility. The precise segmentation of kidneys and kidney tumors can help medical specialists to diagnose diseases and improve treatment planning, which is highly required in clinical practice. Manual segmentation of the kidneys is extremely time-consuming and prone to variability between different specialists due to their heterogeneity. Because of this hard work, computational techniques, such as deep convolutional neural networks, have become popular in kidney segmentation tasks to assist in the early diagnosis of kidney tumors. All persons are susceptible to kidney stones, even infants, and yet, the majority of kidney stone cases remain undetected except in cases where extreme abdominal pain is exhibited or abnormal urine color is observed. In addition, people with kidney stones exhibit common signs such as fever, pain and nausea that are easily associated to other conditions. Kidney stone detection is important particularly in its early stages to facilitate intervention or to receive proper medical treatment. The presence or the recurring presence of kidney stone decreases kidney functions and dilation of the kidney. This paper presents a technique for detection of kidney stones through different steps of image processing. The first step is the image pre-processing using filters in which image gets smoothed as well as the noise is removed from the image. Next, the image segmentation is performed on the preprocessed image using guided active contour method. Then using Convolutional neural network algorithm to identify the diseases in kidney images. The imaging modality used is CT because it has low noise compared to other modalities such as x-ray and ultrasound.
KEY WORDS: Kidney Stone, Features extraction, Deep learning, Medical images, Convolutional neural network
