Conference Paper2024Cited by 2

Performance Evaluation of Machine Learning and Deep Learning Models for Predicting Type-2 Diabetes on Balanced and Imbalanced Data

Nayema Khatun, Nabarun Halder, Ashraful Islam, Md Zahangir Alam

IEEE 2024 International Advances in Science & Engineering Technology (ASET) Multi-Conferences

IEEE, pp. 1-9

  • Good Health and Well-being — Relevance: 93%
DOI: 10.1109/aset60340.2024.10708720URL

Abstract

Diabetes is a chronic metabolic illness defined by high blood glucose levels caused by the body's inability to produce or utilize insulin. The pancreas does not create enough insulin, thus cells do not respond well. Diabetes mellitus, formerly termed as adult-onset diabetes, can start in childhood or adulthood. Type 2 diabetes typically affects older people. Over time, high blood sugar levels associated with diabetes can lead to damage to various organs, increasing the risk of conditions such as heart disease, kidney disease, and eye problems. Diabetes can be managed if a person becomes aware early about their health condition. This study incorporates dataset balancing techniques and 10-Fold cross-validation to develop a robust predictive model. The study contributes to the understanding of health behavior through the use of deep learning algorithms known as Multilayer Perceptrons (MLPs). Employing ML methods such as Logistic Regression (LR), Random Forest (RF), XGBoost, and AdaBoost (AdaBoost) increased predicted accuracy even further. Various hybrid data preparation strategies, such as the Synthetic Minority Over-sampling Technique (SMOTE), SMOTE with Edited Nearest Neighbors (SMOTE-ENN), and SMOTE-Tomek, improve diabetes prediction accuracy. Our model achieved accuracy levels of up to 99.9%. The findings offer insights for leveraging large-scale health surveys in informed decision-making and public health interventions.

Keywords

  • AdaBoost
  • SMOTE-Tomek
  • type 2 diabetes prediction
  • Random forest
  • SMOTE-ENN
  • Logistic regression
  • Multilayer perceptron
  • SMOTE
  • 10-fold cross-validation
  • XGBoost

CCDS Authors

Nabarun Halder
Nabarun Halder

Nabarun Halder

  • Research Assistant, CCDS

Published at

UbiComp, IJCNN, HCII, IEEE Access

Research Interests

Machine Learning, Deep Learning, Natural Language Processing, Health Informatics, Human-Computer Interaction

Ashraful Islam, PhD
Ashraful Islam

Ashraful Islam, PhD

  • Center Director, CCDS
  • Director, HCI Wing

Assistant Professor

Department of Computer Science and Engineering

Independent University, Bangladesh (IUB)

Published at

CHI, UIST, UbiComp, IJCNN, ICIP, HCII, ICDMW, BLP, IEEE Internet of Things Journal, JMIR, JMIR Human Factors, JMIR Formative Research, IEEE Access, Smart Agricultural Technology, Diabetes Research and Clinical Practice, PLOS ONE

Research Interests

Human-Computer Interaction, AI for Social Good, AI for Public Health, AI for Impact

References

  1. 1.Ahmedbahaaaldin Ibrahem Ahmed Osman, Ali Najah Ahmed, Chow Ming Fai, Yuk Feng Huang, Ahmed El‐Shafie. (2021). Extreme gradient boosting (Xgboost) model to predict the groundwater levels in Selangor Malaysia. Ain Shams Engineering Journal, 12(2), 1545–1556[10.1016/j.asej.2020.11.011]
  2. 2.Aishwarya Madhav Mujumdar, V. Vaidehi. (2019). Diabetes Prediction using Machine Learning Algorithms. Procedia Computer Science, 165, 292–299[10.1016/j.procs.2020.01.047]
  3. 3.Victor Chang, Meghana Ashok Ganatra, Karl Hall, Lewis Golightly, Qianwen Xu. (2022). An assessment of machine learning models and algorithms for early prediction and diagnosis of diabetes using health indicators. Healthcare Analytics, 2, 100118[10.1016/j.health.2022.100118]
  4. 4.Xuchun Wang, Jiahui Ren, Hao Ren, Wenzhu Song, Yuchao Qiao, Ying Zhao, Liqin Linghu, Yu Cui, Zhiyang Zhao, Limin Chen, Lixia Qiu. (2023). Diabetes mellitus early warning and factor analysis using ensemble Bayesian networks with SMOTE-ENN and Boruta. Scientific Reports, 13(1), 12718[10.1038/s41598-023-40036-5]
  5. 5.Ahmed I. B. ElSeddawy, Faten Khalid Karim, Aisha Mohamed Hussein, Doaa Sami Khafaga. (2022). Predictive Analysis of Diabetes-Risk with Class Imbalance. Computational Intelligence and Neuroscience, 2022, 1–16[10.1155/2022/3078025]
  6. 6.Edward Yellakuor Baagyere, Agebure Apambila Moses, Qin Zhen, Oyetunji Elkanah Olaosebikan, Zhiguang Qin. (2020). A Multi-Spiking Neural Network Learning Model for Data Classification. IEEE Access, 8, 72360–72371[10.1109/access.2020.2985257]
  7. 7.(2021). Towards a Smart Healthcare System. International Journal of Healthcare Information Systems and Informatics, 16(4), 0[10.4018/ijhisi.20211001oa08]

Cited By

  1. 1.Elahe Goli, Melike Șah. (2025). Diabetes Classification using Deep Learning With a Hybrid Data Imbalance Handling Approach. , 1–8[10.1109/isas66241.2025.11101940]
  2. 2.Talitha Olga Ramadhani, Guruh Fajar Shidik, Adelfia Milka Novtagravina, Aisyah Nuraini, Erna Amalia Widhiyanti, Edi Jaya Kusum. (2025). Deep Learning Approaches for Medical Data: Early-Stage Diabetes Detection with Class Imbalance Handling Techniques. , 139–144[10.1109/isemantic67418.2025.11292429]
← All publications