Deep Neural Network to Identify Patients with Alcohol Use Disorder

Ali Ebrahimi; Uffe Kock Wiil; Marjan Mansourvar; Amin Naemi; Kjeld Andersen; Anette Søgaard Nielsen

doi:10.3233/SHTI210156

Deep Neural Network to Identify Patients with Alcohol Use Disorder

Ali Ebrahimi, Uffe Kock Wiil, Marjan Mansourvar, Amin Naemi, Kjeld Andersen, Anette Søgaard Nielsen

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

48 Downloads (Pure)

Abstract

This paper presents an application of deep neural networks (DNN) to identify patients with Alcohol Use Disorder based on historical electronic health records. Our methodology consists of four stages including data collection, preprocessing, predictive model development, and validation. Data are collected from two sources and labeled into three classes including Normal, Hazardous, and Harmful drinkers. Moreover, problems such as imbalanced classes, noise, and categorical variables were handled. A four-layer fully-connected feedforward DNN architecture was designed and developed to predict Normal, Hazardous, and Harmful drinkers. Results show that our proposed method could successfully classify about 96%, 82%, and 89% of Normal, Hazardous, and Harmful drinkers, respectively, which is better than classical machine learning approaches.

Original language	English
Title of host publication	Public Health and Informatics
Editors	John Mantas, Lăcrămioara Stoicu-Tivadar, Catherine Chronaki, Arie Hasman, Patrick Weber, Parisis Gallos, Mihaela Crişan-Vida, Emmanouil Zoulias, Oana Sorina Chirila
Volume	281
Publication date	27. May 2021
Pages	238-242
ISBN (Print)	978-1-64368-184-9
ISBN (Electronic)	978-1-64368-185-6
DOIs	https://doi.org/10.3233/SHTI210156
Publication status	Published - 27. May 2021
Event	MIE 2021: 31st Medical Informatics Europe Conference Online - Duration: 29. May 2021 → 31. May 2021

Conference

Conference	MIE 2021: 31st Medical Informatics Europe Conference Online
Period	29/05/2021 → 31/05/2021

Series	Studies in Health Technology and Informatics
Volume	281
ISSN	0926-9630

Keywords

Alcohol Use Disorder
Deep Neural Networks
Multiclass Classification

Documents & Links

10.3233/SHTI210156Licence: CC BY-NC

Open Access VersionFinal published version, 244 KBLicence: CC BY-NC

SDU Link

Check access to the material in SDU Library's search engine

Cite this

Ebrahimi, A., Wiil, U. K., Mansourvar, M., Naemi, A., Andersen, K., & Nielsen, A. S. (2021). Deep Neural Network to Identify Patients with Alcohol Use Disorder. In J. Mantas, L. Stoicu-Tivadar, C. Chronaki, A. Hasman, P. Weber, P. Gallos, M. Crişan-Vida, E. Zoulias, & O. S. Chirila (Eds.), Public Health and Informatics (Vol. 281, pp. 238-242) https://doi.org/10.3233/SHTI210156

Ebrahimi, Ali ; Wiil, Uffe Kock ; Mansourvar, Marjan et al. / Deep Neural Network to Identify Patients with Alcohol Use Disorder. Public Health and Informatics. editor / John Mantas ; Lăcrămioara Stoicu-Tivadar ; Catherine Chronaki ; Arie Hasman ; Patrick Weber ; Parisis Gallos ; Mihaela Crişan-Vida ; Emmanouil Zoulias ; Oana Sorina Chirila. Vol. 281 2021. pp. 238-242 (Studies in Health Technology and Informatics, Vol. 281).

@inproceedings{c7ec79adeaf3469fb9bfbf6b38df0a56,

title = "Deep Neural Network to Identify Patients with Alcohol Use Disorder",

abstract = "This paper presents an application of deep neural networks (DNN) to identify patients with Alcohol Use Disorder based on historical electronic health records. Our methodology consists of four stages including data collection, preprocessing, predictive model development, and validation. Data are collected from two sources and labeled into three classes including Normal, Hazardous, and Harmful drinkers. Moreover, problems such as imbalanced classes, noise, and categorical variables were handled. A four-layer fully-connected feedforward DNN architecture was designed and developed to predict Normal, Hazardous, and Harmful drinkers. Results show that our proposed method could successfully classify about 96%, 82%, and 89% of Normal, Hazardous, and Harmful drinkers, respectively, which is better than classical machine learning approaches.",

keywords = "Alcohol Use Disorder, Deep Neural Networks, Multiclass Classification",

author = "Ali Ebrahimi and Wiil, {Uffe Kock} and Marjan Mansourvar and Amin Naemi and Kjeld Andersen and Nielsen, {Anette S{\o}gaard}",

year = "2021",

month = may,

day = "27",

doi = "10.3233/SHTI210156",

language = "English",

isbn = "978-1-64368-184-9",

volume = "281",

series = "Studies in Health Technology and Informatics",

publisher = "IOS Press BV",

pages = "238--242",

editor = "John Mantas and L{\u a}cr{\u a}mioara Stoicu-Tivadar and Catherine Chronaki and Arie Hasman and Patrick Weber and Parisis Gallos and Mihaela Cri{\c s}an-Vida and Emmanouil Zoulias and Chirila, {Oana Sorina}",

booktitle = "Public Health and Informatics",

note = "MIE 2021: 31st Medical Informatics Europe Conference Online ; Conference date: 29-05-2021 Through 31-05-2021",

}

Ebrahimi, A , Wiil, UK, Mansourvar, M, Naemi, A , Andersen, K & Nielsen, AS 2021, Deep Neural Network to Identify Patients with Alcohol Use Disorder. in J Mantas, L Stoicu-Tivadar, C Chronaki, A Hasman, P Weber, P Gallos, M Crişan-Vida, E Zoulias & OS Chirila (eds), Public Health and Informatics. vol. 281, Studies in Health Technology and Informatics, vol. 281, pp. 238-242, MIE 2021: 31st Medical Informatics Europe Conference Online, 29/05/2021. https://doi.org/10.3233/SHTI210156

Deep Neural Network to Identify Patients with Alcohol Use Disorder. / Ebrahimi, Ali ; Wiil, Uffe Kock; Mansourvar, Marjan et al.
Public Health and Informatics. ed. / John Mantas; Lăcrămioara Stoicu-Tivadar; Catherine Chronaki; Arie Hasman; Patrick Weber; Parisis Gallos; Mihaela Crişan-Vida; Emmanouil Zoulias; Oana Sorina Chirila. Vol. 281 2021. p. 238-242 (Studies in Health Technology and Informatics, Vol. 281).

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

TY - GEN

T1 - Deep Neural Network to Identify Patients with Alcohol Use Disorder

AU - Ebrahimi, Ali

AU - Wiil, Uffe Kock

AU - Mansourvar, Marjan

AU - Naemi, Amin

AU - Andersen, Kjeld

AU - Nielsen, Anette Søgaard

PY - 2021/5/27

Y1 - 2021/5/27

N2 - This paper presents an application of deep neural networks (DNN) to identify patients with Alcohol Use Disorder based on historical electronic health records. Our methodology consists of four stages including data collection, preprocessing, predictive model development, and validation. Data are collected from two sources and labeled into three classes including Normal, Hazardous, and Harmful drinkers. Moreover, problems such as imbalanced classes, noise, and categorical variables were handled. A four-layer fully-connected feedforward DNN architecture was designed and developed to predict Normal, Hazardous, and Harmful drinkers. Results show that our proposed method could successfully classify about 96%, 82%, and 89% of Normal, Hazardous, and Harmful drinkers, respectively, which is better than classical machine learning approaches.

AB - This paper presents an application of deep neural networks (DNN) to identify patients with Alcohol Use Disorder based on historical electronic health records. Our methodology consists of four stages including data collection, preprocessing, predictive model development, and validation. Data are collected from two sources and labeled into three classes including Normal, Hazardous, and Harmful drinkers. Moreover, problems such as imbalanced classes, noise, and categorical variables were handled. A four-layer fully-connected feedforward DNN architecture was designed and developed to predict Normal, Hazardous, and Harmful drinkers. Results show that our proposed method could successfully classify about 96%, 82%, and 89% of Normal, Hazardous, and Harmful drinkers, respectively, which is better than classical machine learning approaches.

KW - Alcohol Use Disorder

KW - Deep Neural Networks

KW - Multiclass Classification

U2 - 10.3233/SHTI210156

DO - 10.3233/SHTI210156

M3 - Article in proceedings

C2 - 34042741

SN - 978-1-64368-184-9

VL - 281

T3 - Studies in Health Technology and Informatics

SP - 238

EP - 242

BT - Public Health and Informatics

A2 - Mantas, John

A2 - Stoicu-Tivadar, Lăcrămioara

A2 - Chronaki, Catherine

A2 - Hasman, Arie

A2 - Weber, Patrick

A2 - Gallos, Parisis

A2 - Crişan-Vida, Mihaela

A2 - Zoulias, Emmanouil

A2 - Chirila, Oana Sorina

T2 - MIE 2021: 31st Medical Informatics Europe Conference Online

Y2 - 29 May 2021 through 31 May 2021

ER -

Ebrahimi A , Wiil UK, Mansourvar M, Naemi A , Andersen K , Nielsen AS. Deep Neural Network to Identify Patients with Alcohol Use Disorder. In Mantas J, Stoicu-Tivadar L, Chronaki C, Hasman A, Weber P, Gallos P, Crişan-Vida M, Zoulias E, Chirila OS, editors, Public Health and Informatics. Vol. 281. 2021. p. 238-242. (Studies in Health Technology and Informatics, Vol. 281). doi: 10.3233/SHTI210156

Deep Neural Network to Identify Patients with Alcohol Use Disorder

Abstract

Conference

Keywords

Documents & Links

SDU Link

Fingerprint

Cite this