Skip to content
SDGs Center UDINUS

Evidence record · UDINUS · SDGs

Improvement of Heart Failure Prediction Using Binary Preprocessing

Rektorat UDINUS 2023 Published 2026-10-30

Evidence record · UDINUS · SDGs

Unit
Rektorat UDINUS
Period
2023
Goals
Record
63af876a-e90b-4404-ad0b-e9b68352e684
Published
2026-10-30

Universitas Dian Nuswantoro conducted a research activity on improving heart failure prediction through binary preprocessing. The activity involved UDINUS researchers Lutfi, Dien Kartika, and Shidik, Guruh Fajar, who participated in examining heart failure prediction using machine learning methods.

The activity addressed heart disease and heart failure prediction. The source identifies several potential factors related to the onset of heart failure, including the nature of chest pain, blood pressure at rest, cholesterol levels in the bloodstream, fasting blood sugar levels, resting electrocardiogram results, attainment of maximum heart rate, exercise-induced angina, oldpeak value, and the incline of the ST segment during peak exercise. The study used a heart failure prediction dataset provided through Kaggle, a platform widely utilized by researchers.

The research compared binary preprocessing techniques with several machine learning algorithms. The algorithms used were Gaussian Naive Bayes, SVM, KNN, Logistic Regression, Deep Learning, Random Forest Classifier, Decision Tree Classifier, and Extra Trees Classifier. The study also applied Random Oversampling, Random Undersampling, Tomek Links, SMOTE, and ENN as sampling techniques. These comparisons sought to obtain a high accuracy rate for heart failure prediction using validated data on potential factors associated with heart failure occurrence.

The reported finding was that preprocessing the data using Robust Scaler and applying ENN Sampling with the KNN algorithm achieved the highest accuracy rate, namely 92.61%, for heart failure prediction cases. This activity relates to the stated goal through its focus on health research and computational prediction of heart failure. The activity is presented as an institutional research contribution by Universitas Dian Nuswantoro in the field of health-related machine learning.

Reference https://www.scopus.com/pages/publications/85186531016?origin=resultslist