Volume 29, Issue 4, 2020

---

DOI: 10.24205/03276716.2020.823

Prediction of Mortalityinthe Hemodialysis Patient with Diabetes using Support Vector Machine

Cheng-Hong Yang, Dony Novaliendry, Jin-Bor Chen, Fegie Y Wattimena, Axelon S Renyaan, Yaslinda Lizar, Asriwan Guci, Muhammad Ariyon, Dochi Ramadhani, Unung Verawardina, Irwan, Yenny Desnelita, Wilda Susanti, Gustientiedina, Hastuti Marlina, | Pg. 219

Abstract
Hemodialysis is one of modality to treat end stage kidney disease. This study is aimed to predict the mortality risk of hemodialysis patients. A total of 665 prevalent hemodialysis patients were enrolled in one hemodialysis center in Taiwan. The prediction is based on Support Vector Machine (SVM) which developed under MATLAB. Based on the obtained results, SVM performs better accuracy compared to K-Nearest Neighbor, logistic regression, a lineardiscriminant,Treeand ensemble. In addition, the F1-score of SVM is higher than that from other methods. The highest mortality risk factor is diabetes; the second is cardiovascular diseaseand small influence of related medical variables such as parathyroid surgery, urea reduction ratio,etc.

Keywords
Diabetes, Hemodialysis, Support Vector Machine

Download PDF