Prediction Model of Type 2 Diabetes Mellitus for Oman Prediabetes Patients Using Artificial Neural Network and Six Machine Learning Classifiers - 文章摘要与方法结果

Aug 5, 2026

Upload by Anda

Early diagnosis of type 2 diabetes mellitus enables timely intervention to slow disease progression and reduce premature mortality. This study builds artificial intelligence and machine learning prediction models for diagnosing T2DM in Oman using a specially created clinical dataset. Six algorithms including K-NN, SVM, naive Bayes, decision tree, random forest, LDA, and ANN are implemented in MATLAB and compared with the Pima Indian Diabetes dataset. With Oman data, random forest and decision tree achieve 98.38% accuracy, and Oman performance exceeds PID by 9.1% under matched settings.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages22

Reading Time55 min

AccessFree to download

Introduction

Disease burden and need for early diagnosis

Related work and dataset context

Methods

Dataset creation in Oman

Pre-processing and feature selection

Algorithms and implementation (MATLAB)

Results

Model comparison against Pima dataset

Accuracy and impact of feature count

Conclusion

applied sciences Article Prediction Model of Type 2 Diabetes Mellitus for Oman Prediabetes Patients Using Artiﬁcial Neural Network and Six Machine Learning Classiﬁers Khoula Al Sadi * and Wamadeva Balachandran Citation: Al Sadi, K.; Balachandran, W. Prediction Model of Type 2 Diabetes Mellitus for Oman Prediabetes Patients Using Artiﬁcial Neural Network and Six Machine Learning Classiﬁers. Appl. Sci. 2023, 13, 2344. [https://doi.org/10.3390/](https://doi.org/10.3390/)[ ](https://doi.org/10.3390/)app13042344 Academic Editor: Yang Kuang Received: 12 December 2022 Revised: 3 February 2023 Accepted: 9 February 2023 Published: 11 February 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) . Department of Electronic and Electrical Engineering Research, Brunel University London, Uxbridge UB8 3PH, UK * Correspondence: [khoula.alsadi@brunel.ac.uk](khoula.alsadi@brunel.ac.uk) Abstract: The early diagnosis of type 2 diabetes mellitus (T2DM) will provide an early treatment intervention to control disease progression and minimise premature death. This paper presents artiﬁcial intelligence and machine learning prediction models for diagnosing T2DM in the Omani population more accurately and with less processing time using a specially created dataset. Six machine learning algorithms: K-nearest neighbours (K-NN), support vector machine (SVM), naive Bayes (NB), decision tree, random forest (RF), linear discriminant analysis (LDA), and artiﬁcial neural networks (ANN) were applied in MATLAB. All data used were clinical data collected manually from a prediabetes register and the Al Shifa health system of South Al Batinah Province in Oman. The results were compared with the most widely used Pima Indian Diabetes dataset. Eleven clinical features were taken into consideration for predicting T2DM. The random forest and decision tree models performed better than all the other algorithms, providing an accuracy of 98.38% for Oman data. When the same model and number of features were used, the accuracy obtained with the Oman dataset exceeded PID by 9.1% . The analysis showed that T2DM diagnosis efﬁciency increased with more features, which is of help in the case of many missing values. Keywords: K-nearest neighbours (K-NN); support vector machine (SVM); naive Bayes (NB); decision tree; random forest (RF); linear discriminant analysis (LDA); artiﬁcial neural network (ANN); type 2 diabetes mellitus (T2DM); Pima Indian Diabetes (PID) dataset; machine learning (ML) 1. Introduction Non-communicable diseases (NCDs) are estimated to comprise 72% of all deaths in Oman. Type 2 diabetes mellitus is the fourth most common reason for death, according to the World Health Organization and World Health Matrix [1] . An early diagnosis of T2DM can prevent further medical complications such as heart disease, kidney failure, retinopathy, depression, and hypertension [2] . The Institute for Health Metrics and Evaluation (IHME) in Washington presented data on the worldwide burden of 369 diseases and injuries in 204 nations between 1990 and 2019 in 2020 . Non-communicable and injury YLDs (years lived with disability) accounted for more than half of disease burdens in 11 countries in 2019 . From 24% in 1990, Oman's T2DM cases climbed to 49%[3] . By 2025, Oman's T2DM rate will climb 174% from 10 .4% to 21 . 1% in those over 20 [4] . Numerous bioinformatics researchers have sought to forecast T2DM and to develop tools and systems that will aid in forecasting. The accuracy rates achieved by various techniques ranged from 77.86% to 95.7% . They either developed prediction models utilising a variety of machine learning algorithms, such as assoc

FAQs

本研究使用了哪些机器学习算法进行T2DM预测？

研究应用了K-nearest neighbours（K-NN）、support vector machine（SVM）、naive Bayes（NB）、decision tree、random forest（RF）、linear discriminant analysis（LDA）以及artificial neural networks（ANN）等算法，并在MATLAB中实现与比较。

Oman数据集与Pima Indian Diabetes（PID）数据集的主要区别是什么？

Oman数据集由作者在阿曼的初级与次级医疗系统中手工收集并用于首次研究，包含11个临床特征；PID数据集为公开数据，文中用于对照。

模型性能结果如何，随机森林与决策树的表现如何？

在阿曼数据上，random forest与decision tree的表现优于其他算法，准确率最高达到98.38%。在相同模型与相同特征数量条件下，Oman数据集的准确率比PID高出9.1%。

Aug 5, 2026

Upload by Anda