Prognosis of kidney disease on ultrasound images using machine learning

Title

Prognosis of kidney disease on ultrasound images using machine learning

Creator

Mino, George

Contributor

H B, Anita

Description

Kidney diseases can affect the ability to clean the blood, filter extra water out of your blood. The kidneys failure will affect the control over blood pressure and sugar level. It can also affect red blood cell production and vitamin D metabolism which is very important for bone health. When your kidneys are damaged, waste products and fluid can build up in the body. This is harmful to the health. This damages the kidney function, can get worse over time, and when the kidneys stop working completely, this is called kidney failure or end-stage renal disease. Not all patients with kidney disease progress to kidney failure. This disease has emerged as one of the most prominent reasons of death and suffering in this century. Recent studies states that, kidney disease affects most of the population and over two million people require kidney replacement. To help prevent Chronic Kidney Diseases and lower the risk for kidney failure, control risk factors for CKD, get tested yearly, make lifestyle changes, take medicine as needed. The detection of kidney abnormalities at their early stages helps to avoid the impairment of newlinekidney. The US imaging is considered as preliminary diagnostic tool in finding various kidney diseases in the clinical imaging field. This is one of the commonly used imaging modalities due to the inexpensiveness and non-ionization nature. The presence of noise in US images, degrade newlinethe quality and clarity of the images. Also, the heterogeneous structure of kidney, makes it very difficult to detect and measure the size of stones and cysts. Hence, an automatic kidney disease detection system is highly in demand. The proposed model can assist the radiologist in accurate abnormality detection. The proposed model includes different phases such as, pre-processing, features extraction, classification and newlinesegmentation. The pre-processing phase include cropping and noise removal. Further, the GLCM and intensity-based features are extracted for the classification of abnormal kidney images.

Source

Author's Submission

Date

2024-11-08

Publisher

Christ(Deemed to be University)

Subject

Computer Science

Rights

Open Access

Relation

61000348

Format

PDF

Language

English

Type

PhD

Identifier

http://hdl.handle.net/10603/575078