Comparison of the predictive outcomes for anti-tuberculosis drug-induced hepatotoxicity by different machine learning techniques

Nai Hua Lai, Wan Chen Shen, Chun Nin Lee, Jui Chia Chang, Man Ching Hsu, Li Na Kuo, Ming Chih Yu, Hsiang Yin Chen

研究成果: 雜誌貢獻文章

摘要

Background: The study compared the predictive outcomes of artificial neural network, support vector machine and random forest on the occurrence of anti-tuberculosis drug-induced hepatotoxicity. Methods: The clinical and genomic data of patients treated with anti-tuberculosis drugs at Taipei Medical University-Wanfang Hospital were used as training sets, and those at Taipei Medical University-Shuang Ho Hospital served as test sets. Features were selected through a univariate risk factor analysis and literature evaluation. The accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve were calculated to compare the traditional, genomic, and combined models of the three techniques. Results: Nine models were created with 7 clinical factors and 4 genotypes. Artificial neural network with clinical and genomic factors exhibited the best performance, with an accuracy of 88.67%, a sensitivity of 80%, and a specificity of 90.4% for the test set. The area under the receiver operating characteristic curve of this best model reached 0.894 for training set and 0.898 for test set, which was significantly better than 0.801 for training set and 0.728 for test set by support vector machine and 0.724 for training set and 0.718 for test set by random forest. Conclusions: Artificial neural network with clinical and genomic data can become a clinical useful tool in predicting anti-tuberculosis drug-induced hepatotoxicity. The machine learning technique can be an innovation to predict and prevent adverse drug reaction.
原文英語
文章編號105307
期刊Computer Methods and Programs in Biomedicine
188
DOIs
出版狀態已發佈 - 五月 2020

指紋

Learning systems
Tuberculosis
Neural networks
ROC Curve
Support vector machines
Pharmaceutical Preparations
Factor analysis
Drug-Related Side Effects and Adverse Reactions
Statistical Factor Analysis
Innovation
Genotype
Sensitivity and Specificity
Machine Learning
Support Vector Machine
Forests

ASJC Scopus subject areas

  • Software
  • Computer Science Applications
  • Health Informatics

引用此文

@article{3aa1c3f7d76e4d5cb6a3ba4e2caa62da,
title = "Comparison of the predictive outcomes for anti-tuberculosis drug-induced hepatotoxicity by different machine learning techniques",
abstract = "Background: The study compared the predictive outcomes of artificial neural network, support vector machine and random forest on the occurrence of anti-tuberculosis drug-induced hepatotoxicity. Methods: The clinical and genomic data of patients treated with anti-tuberculosis drugs at Taipei Medical University-Wanfang Hospital were used as training sets, and those at Taipei Medical University-Shuang Ho Hospital served as test sets. Features were selected through a univariate risk factor analysis and literature evaluation. The accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve were calculated to compare the traditional, genomic, and combined models of the three techniques. Results: Nine models were created with 7 clinical factors and 4 genotypes. Artificial neural network with clinical and genomic factors exhibited the best performance, with an accuracy of 88.67{\%}, a sensitivity of 80{\%}, and a specificity of 90.4{\%} for the test set. The area under the receiver operating characteristic curve of this best model reached 0.894 for training set and 0.898 for test set, which was significantly better than 0.801 for training set and 0.728 for test set by support vector machine and 0.724 for training set and 0.718 for test set by random forest. Conclusions: Artificial neural network with clinical and genomic data can become a clinical useful tool in predicting anti-tuberculosis drug-induced hepatotoxicity. The machine learning technique can be an innovation to predict and prevent adverse drug reaction.",
keywords = "Anti-tuberculosis drugs, Artificial neural network, Feature selection, Gene polymorphism, Random forest, Support vector machine, Tuberculosis",
author = "Lai, {Nai Hua} and Shen, {Wan Chen} and Lee, {Chun Nin} and Chang, {Jui Chia} and Hsu, {Man Ching} and Kuo, {Li Na} and Yu, {Ming Chih} and Chen, {Hsiang Yin}",
year = "2020",
month = "5",
doi = "10.1016/j.cmpb.2019.105307",
language = "English",
volume = "188",
journal = "Computer Methods and Programs in Biomedicine",
issn = "0169-2607",
publisher = "Elsevier Ireland Ltd",

}

TY - JOUR

T1 - Comparison of the predictive outcomes for anti-tuberculosis drug-induced hepatotoxicity by different machine learning techniques

AU - Lai, Nai Hua

AU - Shen, Wan Chen

AU - Lee, Chun Nin

AU - Chang, Jui Chia

AU - Hsu, Man Ching

AU - Kuo, Li Na

AU - Yu, Ming Chih

AU - Chen, Hsiang Yin

PY - 2020/5

Y1 - 2020/5

N2 - Background: The study compared the predictive outcomes of artificial neural network, support vector machine and random forest on the occurrence of anti-tuberculosis drug-induced hepatotoxicity. Methods: The clinical and genomic data of patients treated with anti-tuberculosis drugs at Taipei Medical University-Wanfang Hospital were used as training sets, and those at Taipei Medical University-Shuang Ho Hospital served as test sets. Features were selected through a univariate risk factor analysis and literature evaluation. The accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve were calculated to compare the traditional, genomic, and combined models of the three techniques. Results: Nine models were created with 7 clinical factors and 4 genotypes. Artificial neural network with clinical and genomic factors exhibited the best performance, with an accuracy of 88.67%, a sensitivity of 80%, and a specificity of 90.4% for the test set. The area under the receiver operating characteristic curve of this best model reached 0.894 for training set and 0.898 for test set, which was significantly better than 0.801 for training set and 0.728 for test set by support vector machine and 0.724 for training set and 0.718 for test set by random forest. Conclusions: Artificial neural network with clinical and genomic data can become a clinical useful tool in predicting anti-tuberculosis drug-induced hepatotoxicity. The machine learning technique can be an innovation to predict and prevent adverse drug reaction.

AB - Background: The study compared the predictive outcomes of artificial neural network, support vector machine and random forest on the occurrence of anti-tuberculosis drug-induced hepatotoxicity. Methods: The clinical and genomic data of patients treated with anti-tuberculosis drugs at Taipei Medical University-Wanfang Hospital were used as training sets, and those at Taipei Medical University-Shuang Ho Hospital served as test sets. Features were selected through a univariate risk factor analysis and literature evaluation. The accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve were calculated to compare the traditional, genomic, and combined models of the three techniques. Results: Nine models were created with 7 clinical factors and 4 genotypes. Artificial neural network with clinical and genomic factors exhibited the best performance, with an accuracy of 88.67%, a sensitivity of 80%, and a specificity of 90.4% for the test set. The area under the receiver operating characteristic curve of this best model reached 0.894 for training set and 0.898 for test set, which was significantly better than 0.801 for training set and 0.728 for test set by support vector machine and 0.724 for training set and 0.718 for test set by random forest. Conclusions: Artificial neural network with clinical and genomic data can become a clinical useful tool in predicting anti-tuberculosis drug-induced hepatotoxicity. The machine learning technique can be an innovation to predict and prevent adverse drug reaction.

KW - Anti-tuberculosis drugs

KW - Artificial neural network

KW - Feature selection

KW - Gene polymorphism

KW - Random forest

KW - Support vector machine

KW - Tuberculosis

UR - http://www.scopus.com/inward/record.url?scp=85077332542&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85077332542&partnerID=8YFLogxK

U2 - 10.1016/j.cmpb.2019.105307

DO - 10.1016/j.cmpb.2019.105307

M3 - Article

AN - SCOPUS:85077332542

VL - 188

JO - Computer Methods and Programs in Biomedicine

JF - Computer Methods and Programs in Biomedicine

SN - 0169-2607

M1 - 105307

ER -