International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 11 Issue: 06 | Jun 2024
www.irjet.net
p-ISSN: 2395-0072
PROGNOSIS FOR DIABETES USING: MACHINE LEARNING 1Tulsi Bhalani, 2Ami Mehta 1PG Scholar, Computer Science and Engineering, Dr. Subhash University, Gujarat, India
2Assistant Professor, Computer Engineering, Dr. Subhash University, Gujarat, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Diabetes Mellitus, a chronic metabolic disorder affecting millions worldwide, presents substantial challenges to both healthcare systems and individual human beings. In this study the use of machine learning algorithms to prognosis diabetes by focusing on long term outcomes such as disease progression, complications etc. by using the dataset of diabetes patients records, various machine learning models are trained and evaluated like as SVM, Naïve Bayes, decision tree, random forest, logistic regression, gradient boosting etc. for compaction of treatment strategies and to get improved patient outcome. The highest accuracy is obtained from the hybrid of Correlation Feature Selection technique with XGB Algorithm that is 89%.
lifestyle factors, and genetic information, to predict disease outcomes with high accuracy. This study aims to leverage machine learning to enhance the prognosis of diabetes. By employing supervised Learning algorithms, we seek to develop a predictive model that can identify patients at high risk of disease progression and suggest personalized intervention strategies. The following sections will delve into the methodology, dataset, and specific machine learning techniques used, followed by an analysis of the results and their implications for clinical practice.
1.1 Problem Statement:
Key Words: Diabetes Mellitus, Prognosis, Chronic metabolic disorder, Machine learning, Disease progression, Support Vector Machine (SVM), Decision tree, Random forest, Logistic regression, Gradient boosting, Patient outcomes, XGB Algorithm, Hybrid model, Healthcare, Correlation Feature Selection
Diabetes prognostic models are usually detached treating patients as homogeneous entities within the groups that are uniform. Machine learning algorithms have the potential to identify specific variations in health parameters and lifestyle, leading to more customized and effective prognostic models. A more accurate and comprehensive prognosis is made feasible by the integration of genetic information, wearable data, and electronic health records in healthcare. This method provides an in-depth view of an individual's health. The analysis of long-term information is made possible by the use of machine learning algorithms, whereas traditional models may not be able to fully capture temporal patterns. This method contributes to the identification of patterns and trends, resulting in a more thorough prognosis.
1.INTRODUCTION Diabetes mellitus is a pervasive and chronic health condition that affects millions of individuals worldwide. Characterized by high blood glucose levels, diabetes can lead to severe complications, including cardiovascular disease, nerve damage, and kidney failure, if not managed properly. The World Health Organization estimates that the prevalence of diabetes has been steadily increasing, underscoring the urgent need for effective management and intervention strategies.
1.2 Need of Research:
A critical aspect of managing diabetes is the ability to accurately predict its progression. Early prognosis allows for timely intervention, potentially preventing severe complications and improving the quality of life for patients. However, traditional prognostic methods often fall short due to the complex and multifactorial nature of the disease.
According to the research India ranks second behind China in Diabetes. The number of people suffering from Diabetes is increasing, there is continuous change in the statistics and frequent updates are required. The cause of Diabetes between the ages of 20-79 years in India is recorded 8.9%. As per the current research Diabetes cases in adults are expected to reach 100 million by 2030. Currently 1 in 6 adults are recorded with diabetes.
In recent years, machine learning has emerged as a powerful tool in the field of healthcare, offering advanced techniques for analyzing large datasets and uncovering patterns that may not be apparent through conventional methods. Machine learning algorithms can process vast amounts of patient data, including medical history,
As the statistics are increasing rapidly we have proposed the research to predict the occurrence of diabetes by analyzing the symptoms such as Pregnancies, Glucose, Blood Pressure, Skin Thickness, Insulin and many more.
© 2024, IRJET
|
Impact Factor value: 8.226
|
ISO 9001:2008 Certified Journal
|
Page 106