Predicting Chronic Kidney Diseases in Bangladesh Using Machine Learning

Date
2022
Research Supervisor
Editor
Journal Title
Volume
Issue
Journal Title
Journal ISSN
Volume Title
Abstract
Chronic Kidney Disease (CKD) has been considered a worldwide public health prob lem. Unfortunately, the symptoms of CKD do not appear until very late in the progression period, often reaching terminal stage. A reliable CKD prediction model built by ML strategies can be beneficial for its early diagnosis. To predict and an alyze CKD in the context of Bangladesh, we collected data of 200 patients from Mitford Hospital, Dhaka. We applied eight different ML algorithms to our data: Decision Tree, K-nearest Neighbour (KNN), Support Vector Machine, Logistic Re gression, Random Forest, XGBoost, Perceptron and Gaussian Naive Bayes. We achieved satisfactory results for all the classifiers, among which Support Vector Ma chine (SVM) gave the best results with 93% accuracy and recall. We also conducted some Exploratory Data Analysis (EDA) in this research using Explainable Artificial Intelligence (XAI) to draw meaningful insights about the features.
Description
Keywords
Citation
Department Name
Electrical and Computer Engineering
Publisher
North South University
Printed Thesis
DOI
ISSN
ISBN