Perbandingan Decision Tree dan Neural Network dalam Prediksi LVEF pada Pasien Gagal Jantung

Eva Rahmawati; Mieke Nurmalasari; Hosizah Markam; Dhiar Niken Larasati

doi:10.35960/ikomti.v7i2.2438

Authors

Eva Rahmawati Universitas Esa Unggul
Mieke Nurmalasari Universitas Esa Unggul
Hosizah Markam Universitas Esa Unggul
Dhiar Niken Larasati Badan Pusat Statistik

DOI:

https://doi.org/10.35960/ikomti.v7i2.2438

Keywords:

Heart Failure, LVEF, Decision Tree, Neural Network, Data Mining

Abstract

Heart failure is a cardiovascular disease with a high mortality rate, affecting more than 64 million people worldwide, with a one-year fatality rate of 33%. To assess cardiac performance, the Left Ventricular Ejection Fraction (LVEF) indicator was used, which reflects the ability of the left ventricle to pump blood. Therefore, an analytical approach is needed to predict LVEF values more accurately based on patient characteristics. This study aimed to compare the performance of Decision Tree and Neural Network algorithms in predicting LVEF values in patients with heart failure. Data processing was conducted using Orange Data Mining, utilizing risk factor variables as the basis for classification. The approach used was supervised learning, involving 381 heart failure patient records that were split using an 80:20 train-test split technique, resulting in 305 training data and 76 testing data. Class imbalance in the training data was handled using the SMOTE method before the modeling process. The experimental results show that the model built using the Neural Network algorithm provides better performance than the Decision Tree. This was indicated by an AUC value of 64.8%, classification accuracy of 57.9%, F1-score of 55.8%, precision of 57.6%, and recall of 57.9%. In addition, based on the confusion matrix evaluation, the Neural Network algorithm was able to achieve higher accuracy and recall levels in most LVEF categories. Based on these results, it can be concluded that the Neural Network is a more effective method for predicting LVEF values in patients with heart failure. This model is considered more capable of recognizing complex data patterns, thereby producing better predictive accuracy compared with the Decision Tree algorithm.

References

1] WHO, “Cardiovascular diseases (CVDs),” 11 June. [Online]. Available: https://www.who.int/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds)

[2] N. Afni, F. Amalinda, and N. Asriyani, “Analysis of Risk Factors for Heart Failure in Outpatients at The Undata Regional General Hospital, Central Sulawesi Province,” J. Ris. Kualitatif dan Promosi Kesehat., vol. 4, no. 1, pp. 51–67, 2025, doi: 10.61194/jrkpk.v4i1.773.

[3] C. L. Shams P, Malik A, “Heart Failure (Congestive Heart Failure),” Updated 2025 Feb 26. [Online]. Available: https://www.ncbi.nlm.nih.gov/books/NBK430873/

[4] D. Haryadi and D. M. U. Atmaja, “Penerapan Algoritma K-Means Clustering Untuk Pengelompokan Tingkat Risiko Penyakit Jantung,” J. Informatics Commun. Technol., vol. 3, no. 2, pp. 51–66, 2021, doi: 10.52661/j_ict.v3i2.85.

[5] D. T. Larose and C. D. Larose, Discovering Knowledge in Data. 2014. doi: 10.1002/9781118874059.

[6] F. Muhammad, R. Akhdan, A. Ismail, I. A. Mashudi, and A. L. Maukar, “Comparative Analysis of Decision Tree and Artificial Neural Network Methods for Predicting Potential Heart Disease,” vol. 10, no. 1, pp. 29–33, 2025.

[7] M. A. Hassani, R. Tao, M. Kamyab, and M. H. Mohammadi, “An Approach of Predicting Heart Disease Using a Hybrid Neural Network and Decision Tree,” ACM Int. Conf. Proceeding Ser., no. January, pp. 84–89, 2020, doi: 10.1145/3404687.3404704.

[8] H. Purnomo, R. E. Pambudi, and R. Irawan, “Penerapan Decision Tree untuk Klasifikasi Penyakit Berdasarkan Data Rekam Medis,” vol. 7, no. 1, pp. 1–8.

[9] D. Chicco and G. Jurman, “Machine learning can predict survival of patients with heart failure from serum creatinine and ejection fraction alone,” BMC Med. Inform. Decis. Mak., vol. 20, no. 1, pp. 1–16, 2020, doi: 10.1186/s12911-020-1023-5.

[10] P. Ponikowski et al., “2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure,” Eur. Heart J., vol. 37, no. 27, pp. 2129-2200m, 2016, doi: 10.1093/eurheartj/ehw128.

[11] I. A. Ahmad Sabri, M. Man, W. A. W. Abu Bakar, and A. N. Mohd Rose, “Web Data Extraction Approach for Deep Web using WEIDJ,” Procedia Comput. Sci., vol. 163, pp. 417–426, 2019, doi: 10.1016/j.procs.2019.12.124.

[12] A. Indrawati, L. Ilmu, P. Indonesia, and P. I. Diabetes, “Penerapan Teknik Kombinasi Oversampling Dan Undersampling Hybrid Oversampling And Undersampling Techniques To Handling Imbalanced Dataset,” vol. 4, no. 1, pp. 38–43, 2021, doi: 10.33387/jiko.

[13] Z. Fatah and I. Addewiyah, “Penerapan Algoritma Random Forest dan Teknik SMOTE untuk Prediksi Kematian Akibat Gagal Jantung Menggunakan RapidMiner,” vol. 4, 2025.

[14] A. Informatics, R. Saktriawindarta, N. Evitarina, and A. Info, “Pemodelan Pohon Keputusan Menggunakan Algoritma ID3 dalam Pendekatan Data Mining,” vol. 09, no. 1, pp. 53–58, 2026.

[15] S. Demir, H. Selvitopi, and Z. Selvitopi, “An early and accurate diagnosis and detection of the coronary heart disease using deep learning and machine learning algorithms,” J. Big Data, vol. 12, no. 1, 2025, doi: 10.1186/s40537-025-01283-7.

[16] S. A. Hicks et al., “On evaluation metrics for medical applications of artificial intelligence,” Sci. Rep., no. 0123456789, pp. 1–9, 2022, doi: 10.1038/s41598-022-09954-8.

[17] S. Batool, I. A. Taj, and M. Ghafoor, “Ejection Fraction Estimation from Echocardiograms Using Optimal Left Ventricle Feature Extraction Based on Clinical Methods,” Diagnostics, vol. 13, no. 13, 2023, doi: 10.3390/diagnostics13132155.

[18] A. Syukron, S. Sardiarinto, E. Saputro, and P. Widodo, “Penerapan Metode Smote Untuk Mengatasi Ketidakseimbangan Kelas Pada Prediksi Gagal Jantung,” J. Teknol. Inf. dan Terap., vol. 10, no. 1, pp. 47–50, 2023, doi: 10.25047/jtit.v10i1.313.