Classification of Chronic Kidney Disease based on health care records using machine learning with Support Vector Machine

Aug 6, 2026

Upload by Himbo

Chronic Kidney Disease (CKD) is an increasing global health burden that requires early and accurate diagnosis to support effective management. This study applies machine learning, using Support Vector Machine (SVM), to classify CKD from health care records. A dataset containing clinical and demographic information is preprocessed, relevant features are extracted, and SVM is trained and tested on separate data splits. The resulting model achieves 98% accuracy for predicting whether a patient is diagnosed with CKD.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages7

Reading Time18 min

AccessFree to download

Abstract

Introduction

Background and epidemiology of CKD

Diagnosis indicators and causes

Burden of CKD worldwide

Medical records as data for classification

Support Vector Machine approach

Available : [https://journal.isi-padangpanjang.ac.id/index.php/JTST](https://journal.isi-padangpanjang.ac.id/index.php/JTST) Vol 2, No 2, Dec 2023 E-ISSN: 2962-5378 Classification of Chronic Kidney Disease based on health care records using machine learning with Support Vector Machine Abdurrahman Niarman 1 , iswandi2 Amuharnis 3 Department of Informatics Management, UIN Mahmud Yunus Batusangkar1, Department of Informatics, Universitas Metamedia 2 [aabniarman@uinmybatusangkar.ac.id](aabniarman@uinmybatusangkar.ac.id) , [iswandi@uinmybatusangkar.ac.id](iswandi@uinmybatusangkar.ac.id) , [amuharnis@metamedia.ac.id](amuharnis@metamedia.ac.id) ABSTRACT Chronic Kidney Disease (CKD) is a global health concern with a rising prevalence that necessitates early and accurate diagnosis for effective management. This study proposes the application of Machine Learning (ML), specifically Support Vector Machine (SVM), to classify CKD based on health care records. Leveraging a comprehensive dataset of patient health records, including clinical and demographic information, the research aims to develop a predictive model that can assist in the timely identification of individuals at risk of CKD. The methodology involves preprocessing the health care records, extracting relevant features, and implementing the SVM algorithm for classification. The dataset is divided into training and testing sets to evaluate the model's performance. The SVM classification model that was developed after going through the data preprocessing process produced results that were good enough to be able to classify whether a patient was diagnosed with CKD or not with an accuracy level of 98% and a total of 400 lines of data and 25 features. Keywords: Chronic Kidney Disease, Machine Learning, Support Vector Machine, Health Care Records, Classification, Predictive Modeling INTRODUCTION Chronic kidney failure is a progressive and slow development of kidney failure, and usually lasts for one year. The kidneys lose the ability to maintain the volume and composition of body fluids under normal food intake (Slyvia Anderson et al. , 2006) . Chronic Kidney Disease in the world is currently increasing and becoming a serious health problem, the results of the 2010 Global Burden of Disease research, chronic kidney disease was the 27th leading cause of death in the world in 1990 and increased to 18th in 2010. In 2013, as many as 2 per 1000 population or 499,800 Indonesians suffered from kidney failure. As many as 6 per 1000 population or 1,499,400 Indonesians suffer from kidney stones (Kemenkes, 2013) . Chronic kidney disease (CKD) arises from many heterogeneous disease pathways that alter the function and structure of the kidney irreversibly, over months or years. The diagnosis of CKD rests on establishing a chronic reduction in kidney function and structural kidney damage. The best available indicator of overall kidney function is glomerular filtration rate (GFR), which equals the total amount of fluid filtered through all of the functioning nephrons per unit of time. Generally, CKD is caused by diffuse and chronic intrinsic kidney disease. Glomerulonephritis, essential hypertension, and pyelonephritis are the most common causes of chronic renal failure, accounting for approximately 60%(Sukandar, 2006) . The burden of CKD is substantial. According to WHO global health estimates, 864 226 deaths (or 1·5% of deaths worldwide) were attributable to this condition in 2012. Ranked fourteenth in the list of leading causes of death, CKD accounted for 12·2 deaths per 100 000 people. Since 1990, only deaths from complications of HIV infection have increased at a faster rate than deaths from CKD. Projections from the Global Health Observatory suggest that although the death rate from HIV will decrease in the next 15 years, the death rate from CKD will continue to increase to reach 14 per 100 000 people by 2030. CKD is also associated with substantial morbidity. World

FAQs

What problem does the study address about Chronic Kidney Disease (CKD)?

CKD is described as a global health concern with rising prevalence, requiring early and accurate diagnosis to improve management.

Which machine learning method is used to classify CKD in this study?

The study uses Support Vector Machine (SVM) to build a predictive classification model based on health care records.

How is the model evaluated and what performance is reported?

The dataset is divided into training and testing sets to evaluate performance, and the SVM model reaches an accuracy level of 98% for classifying CKD diagnosis.

Aug 6, 2026

Upload by Himbo