International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395 -0056
Volume: 04 Issue: 04 | Apr-2017
p-ISSN: 2395-0072
www.irjet.net
DETECTING DIABETES MELLITUS GRADIENT VECTOR FLOW SNAKE SEGMENTED TECHNIQUE Dr. S. K. Jayanthi1, B.Shanmugapriyanga2 1 Head
and Associate Professor, Dept. of Computer Science, Vellalar College for Women, Erode, Tamilnadu, India Scholar, Department of Computer Science, Vellalar College for Women, Erode, Tamilnadu, India ---------------------------------------------------------------------***--------------------------------------------------------------------2 Research
Abstract -Diabetes is a chronic disease and a major
A fasting plasma glucose (FPG) test is the standard method practiced by many medical professionals to diagnose DM. It is performed after the patient has gone at least 12 hours without food, and requires taking a sample of the patient’s blood (by piercing their finger) in order to analyze its blood glucose levels. Even though this method is accurate, it can be considered invasive, and slightly painful. During the past several years, certain achievements have been made in tongue image and diagnostic classification technologies. Moreover tongue diagnosis is considered to be the most promising direction in the 21st century: no pain and no injury. Tongue image analysis, especially its chromatic features provides plenty of valuable diagnostic information to reveal the disorder or even pathological changes of internal organs. Tongue images were captured using a specially designed in-house device taking color correction into consideration. Segmentation, feature extraction and classification techniques are used to detect diabetes mellitus from tongue image. Gradient Vector Flow has been used to segment tongue image alone. Then color, texture and geometry features are extracted from tongue image. Moreover hybrid classifier of minimum distance, bayes classifier and support vector machine is used to classify whether the tongue image is healthy or diabetes. This paper is organized as follows: Related works are discussed in Section 2 and Section 3 describes the process of diabetes detection using Gradient Vector Flow (GVF) Snake Technique. Section 4 analyses the results based on the evaluation metrics. Finally, Section 5 concludes the work.
public health challenge worldwide. Due to lack of awareness among the people on eating habits, diabetic patient counts have been increased steadily in our country. This motivates researchers to develop a medical system which can screen a large number of people for life-threatening disease such as cardiovascular disease, the retinal disorder in diabetic patients. Tongue has played a prominent role in the diagnosis and the subsequent treatment of diseases. Tongue segmentation using Bi-Elliptical Deformable Contour (BEDC) does not process fake edges and also provides poor results. Hence this paper proposes Gradient Vector Flow (GVF) snake technique to extract the region as it encourages convergence in boundary concavities and also provides better results in detecting diabetes mellitus compared to BEDC. Moreover the hybrid classifier using Minimum Distance, Bayes Classifier and Support Vector Machine have been proposed and developed in this research work and gives promising results. The results are evaluated using performance evaluation metrics, Sensitivity and Specificity and gains an accuracy of 85.5% compared to BEDC which has an accuracy of 60%. Key Words: Diabetes Mellitus (DM), Bi-Elliptical Deformable Contour (BEDC), Gradient Vector Flow (GVF) snake technique, Feature Extraction, Hybrid Classifier. 1. INTRODUCTION Diabetes mellitus is one of the most serious health challenges in both developing and developed countries. Diabetes is due to either the pancreas not producing enough insulin, or the cells of the body not responding properly to the insulin produced. Diabetes mellitus (DM), also known as simply diabetes, is a group of metabolic disorder in which blood sugar levels is high over a prolonged period. This high blood sugar results in the symptoms of frequent urination, increased thirst, and increased hunger. Disease that results in autoimmune destruction of insulin-producing beta cells of the pancreas is diabetes mellitus type-I. Metabolic disorder that is characterized by high blood glucose in the context of insulin fighting and relative insulin deficiency is diabetes mellitus type-II.
Š 2017, IRJET
|
Impact Factor value: 5.181
2. LITERATURE REVIEW Diabetes mellitus have been detected using segmentation, feature extraction and classification. Particularly, there have been a considerable number of efforts that rely on segmentation to detect diabetes mellitus. It is beneficial to evaluate and examine the existing systems for better understanding of detecting diabetes mellitus. Hence, recent approaches and methodologies in the area of detecting diabetes mellitus have been discussed.
|
ISO 9001:2008 Certified Journal
|
Page 1238