ARPnet: Antidepressant response prediction network for major depressive disorder

Buru Chang, Yonghwa Choi, Minji Jeon, Junhyun Lee, Kyu Man Han, Aram Kim, Byung Joo Ham, Jaewoo Kang

Research output: Contribution to journalArticle

Abstract

Treating patients with major depressive disorder is challenging because it takes 6-8 weeks for antidepressants prescribed for the patients to take effect. This limitation may result in increased risks and treatment costs. To address this limitation, an accurate antidepressant response prediction model is needed. Recently, several studies have proposed models that extract useful features such as neuroimaging biomarkers and genetic variants from patient data, and use them as predictors for predicting the antidepressant responses of patients. However, it is impossible to utilize all the different types of predictors when making a clinical decision on what drugs to prescribe for a patient. Although a machine learning-based antidepressant response prediction model has been proposed to overcome this problem, the model cannot find the most effective antidepressant for a patient. Based on a neural network, we propose an Antidepressant Response Prediction Network (ARPNet) model capturing high-dimensional patterns from useful features. Based on a literature survey and data-driven feature selection, we extract useful features from patient data, and use the features as predictors. In ARPNet, the patient representation layer captures patient features and the antidepressant prescription representation layer captures antidepressant features. Utilizing the patient and antidepressant prescription representation vectors, ARPNet predicts the degree of antidepressant response. The experimental evaluation results demonstrate that our proposed ARPNet model outperforms machine learning-based models in predicting antidepressant response. Moreover, we demonstrate the applicability of ARPNet in downstream applications in use case scenarios. Our source code that implements ARPNet is available at http://github.com/dmis-lab/arpnet.

Original languageEnglish
Article number907
JournalGenes
Volume10
Issue number11
DOIs
Publication statusPublished - 2019 Nov

Fingerprint

Major Depressive Disorder
Antidepressive Agents
Prescriptions
Neuroimaging
Health Care Costs

Keywords

  • Antidepressant response prediction
  • Major depressive disorder
  • Neural network
  • Patient representation

ASJC Scopus subject areas

  • Genetics
  • Genetics(clinical)

Cite this

ARPnet : Antidepressant response prediction network for major depressive disorder. / Chang, Buru; Choi, Yonghwa; Jeon, Minji; Lee, Junhyun; Han, Kyu Man; Kim, Aram; Ham, Byung Joo; Kang, Jaewoo.

In: Genes, Vol. 10, No. 11, 907, 11.2019.

Research output: Contribution to journalArticle

Chang, Buru ; Choi, Yonghwa ; Jeon, Minji ; Lee, Junhyun ; Han, Kyu Man ; Kim, Aram ; Ham, Byung Joo ; Kang, Jaewoo. / ARPnet : Antidepressant response prediction network for major depressive disorder. In: Genes. 2019 ; Vol. 10, No. 11.
@article{5d4eda1b6e8646b29ff6e4d75b9ea939,
title = "ARPnet: Antidepressant response prediction network for major depressive disorder",
abstract = "Treating patients with major depressive disorder is challenging because it takes 6-8 weeks for antidepressants prescribed for the patients to take effect. This limitation may result in increased risks and treatment costs. To address this limitation, an accurate antidepressant response prediction model is needed. Recently, several studies have proposed models that extract useful features such as neuroimaging biomarkers and genetic variants from patient data, and use them as predictors for predicting the antidepressant responses of patients. However, it is impossible to utilize all the different types of predictors when making a clinical decision on what drugs to prescribe for a patient. Although a machine learning-based antidepressant response prediction model has been proposed to overcome this problem, the model cannot find the most effective antidepressant for a patient. Based on a neural network, we propose an Antidepressant Response Prediction Network (ARPNet) model capturing high-dimensional patterns from useful features. Based on a literature survey and data-driven feature selection, we extract useful features from patient data, and use the features as predictors. In ARPNet, the patient representation layer captures patient features and the antidepressant prescription representation layer captures antidepressant features. Utilizing the patient and antidepressant prescription representation vectors, ARPNet predicts the degree of antidepressant response. The experimental evaluation results demonstrate that our proposed ARPNet model outperforms machine learning-based models in predicting antidepressant response. Moreover, we demonstrate the applicability of ARPNet in downstream applications in use case scenarios. Our source code that implements ARPNet is available at http://github.com/dmis-lab/arpnet.",
keywords = "Antidepressant response prediction, Major depressive disorder, Neural network, Patient representation",
author = "Buru Chang and Yonghwa Choi and Minji Jeon and Junhyun Lee and Han, {Kyu Man} and Aram Kim and Ham, {Byung Joo} and Jaewoo Kang",
year = "2019",
month = "11",
doi = "10.3390/genes10110907",
language = "English",
volume = "10",
journal = "Genes",
issn = "2073-4425",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "11",

}

TY - JOUR

T1 - ARPnet

T2 - Antidepressant response prediction network for major depressive disorder

AU - Chang, Buru

AU - Choi, Yonghwa

AU - Jeon, Minji

AU - Lee, Junhyun

AU - Han, Kyu Man

AU - Kim, Aram

AU - Ham, Byung Joo

AU - Kang, Jaewoo

PY - 2019/11

Y1 - 2019/11

N2 - Treating patients with major depressive disorder is challenging because it takes 6-8 weeks for antidepressants prescribed for the patients to take effect. This limitation may result in increased risks and treatment costs. To address this limitation, an accurate antidepressant response prediction model is needed. Recently, several studies have proposed models that extract useful features such as neuroimaging biomarkers and genetic variants from patient data, and use them as predictors for predicting the antidepressant responses of patients. However, it is impossible to utilize all the different types of predictors when making a clinical decision on what drugs to prescribe for a patient. Although a machine learning-based antidepressant response prediction model has been proposed to overcome this problem, the model cannot find the most effective antidepressant for a patient. Based on a neural network, we propose an Antidepressant Response Prediction Network (ARPNet) model capturing high-dimensional patterns from useful features. Based on a literature survey and data-driven feature selection, we extract useful features from patient data, and use the features as predictors. In ARPNet, the patient representation layer captures patient features and the antidepressant prescription representation layer captures antidepressant features. Utilizing the patient and antidepressant prescription representation vectors, ARPNet predicts the degree of antidepressant response. The experimental evaluation results demonstrate that our proposed ARPNet model outperforms machine learning-based models in predicting antidepressant response. Moreover, we demonstrate the applicability of ARPNet in downstream applications in use case scenarios. Our source code that implements ARPNet is available at http://github.com/dmis-lab/arpnet.

AB - Treating patients with major depressive disorder is challenging because it takes 6-8 weeks for antidepressants prescribed for the patients to take effect. This limitation may result in increased risks and treatment costs. To address this limitation, an accurate antidepressant response prediction model is needed. Recently, several studies have proposed models that extract useful features such as neuroimaging biomarkers and genetic variants from patient data, and use them as predictors for predicting the antidepressant responses of patients. However, it is impossible to utilize all the different types of predictors when making a clinical decision on what drugs to prescribe for a patient. Although a machine learning-based antidepressant response prediction model has been proposed to overcome this problem, the model cannot find the most effective antidepressant for a patient. Based on a neural network, we propose an Antidepressant Response Prediction Network (ARPNet) model capturing high-dimensional patterns from useful features. Based on a literature survey and data-driven feature selection, we extract useful features from patient data, and use the features as predictors. In ARPNet, the patient representation layer captures patient features and the antidepressant prescription representation layer captures antidepressant features. Utilizing the patient and antidepressant prescription representation vectors, ARPNet predicts the degree of antidepressant response. The experimental evaluation results demonstrate that our proposed ARPNet model outperforms machine learning-based models in predicting antidepressant response. Moreover, we demonstrate the applicability of ARPNet in downstream applications in use case scenarios. Our source code that implements ARPNet is available at http://github.com/dmis-lab/arpnet.

KW - Antidepressant response prediction

KW - Major depressive disorder

KW - Neural network

KW - Patient representation

UR - http://www.scopus.com/inward/record.url?scp=85074711872&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85074711872&partnerID=8YFLogxK

U2 - 10.3390/genes10110907

DO - 10.3390/genes10110907

M3 - Article

C2 - 31703457

AN - SCOPUS:85074711872

VL - 10

JO - Genes

JF - Genes

SN - 2073-4425

IS - 11

M1 - 907

ER -