Implementation of predictive data mining techniques for identifying risk factors of early AVF failure in hemodialysis patients

Rezapour, M. and Khavanin Zadeh, M. and Sepehri, M.M. (2013) Implementation of predictive data mining techniques for identifying risk factors of early AVF failure in hemodialysis patients. Computational and Mathematical Methods in Medicine, 2013.

[img] Text
830745.pdf

Download (1MB)
Official URL: https://www.scopus.com/inward/record.uri?eid=2-s2....

Abstract

Arteriovenous fistula (AVF) is an important vascular access for hemodialysis (HD) treatment but has 20-60 rate of early failure. Detecting association between patient's parameters and early AVF failure is important for reducing its prevalence and relevant costs. Also predicting incidence of this complication in new patients is a beneficial controlling procedure. Patient safety and preservation of early AVF failure is the ultimate goal. Our research society is Hasheminejad Kidney Center (HKC) of Tehran, which is one of Iran's largest renal hospitals. We analyzed data of 193 HD patients using supervised techniques of data mining approach. There were 137 male (70.98) and 56 female (29.02) patients introduced into this study. The average of age for all the patients was 53.87 ± 17.47 years. Twenty eight patients had smoked and the number of diabetic patients and nondiabetics was 87 and 106, respectively. A significant relationship was found between "diabetes mellitus," "smoking," and "hypertension" with early AVF failure in this study. We have found that these mentioned risk factors have important roles in outcome of vascular surgery, versus other parameters such as "age." Then we predicted this complication in future AVF surgeries and evaluated our designed prediction methods with accuracy rates of 61.66-75.13. © 2013 Mohammad Rezapour et al.

Item Type: Article
Additional Information: cited By 10
Uncontrolled Keywords: adolescent; adult; arteriovenous fistula; article; data mining; diabetes mellitus; disease association; female; heart failure; hemodialysis; human; hypertension; major clinical study; male; outcome assessment; patient safety; risk assessment; smoking; vascular surgery; aged; algorithm; arteriovenous shunt; biology; chronic kidney failure; decision tree; diabetes mellitus; hypertension; methodology; middle aged; renal replacement therapy; risk factor; smoking; statistics; treatment failure; arteriovenous shunt; complication; data mining; Diabetes Complications; Kidney Failure, Chronic; procedures; renal replacement therapy; statistics and numerical data, Fistula, Adult; Aged; Algorithms; Arteriovenous Shunt, Surgical; Computational Biology; Data Mining; Decision Trees; Diabetes Complications; Female; Humans; Hypertension; Kidney Failure, Chronic; Male; Middle Aged; Renal Dialysis; Risk Factors; Smoking; Treatment Failure, Adult; Aged; Algorithms; Arteriovenous Shunt, Surgical; Computational Biology; Data Mining; Decision Trees; Diabetes Complications; Female; Humans; Hypertension; Kidney Failure, Chronic; Male; Middle Aged; Renal Dialysis; Risk Factors; Smoking; Treatment Failure
Subjects: WJ Urogenital System
Depositing User: somayeh pourmorteza
Date Deposited: 14 May 2019 09:48
Last Modified: 14 May 2019 09:48
URI: http://eprints.iums.ac.ir/id/eprint/9676

Actions (login required)

View Item View Item