Predicting First-Episode Psychosis Associated with Cannabis Use with Artificial Neural Networks and Deep Learning

Daniel Stamate, Wajdi Alghamdi, Daniel Stahl, Ida Pu, Fionn Murtagh, Danielle Belgrave, Robin Murray, Marta di Forti

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In recent years, a number of researches started to investigate the existence of links between cannabis use and psychotic disorder. More recently, artificial neural networks and in particular deep learning have set a revolutionary wave in pattern recognition and machine learning. This study proposes a novel machine learning approach based on neural network and deep learning algorithms, to developing highly accurate predictive models for the onset of first-episode psychosis. Our approach is based also on a novel methodology of optimising and post-processing the predictive models in a computationally intensive framework. A study of the trade-off between the volume of the data and the extent of uncertainty due to missing values, both of which influencing the predictive performance, enhanced this approach. Furthermore, we extended our approach by proposing and encapsulating a novel post-processing k-fold cross-testing method in order to further optimise, and test these models. The results show that the average accuracy in predicting first-episode psychosis achieved by our models in intensive Monte Carlo simulation, is about 89%.

LanguageEnglish
Title of host publicationInformation Processing and Management of Uncertainty in Knowledge-Based Systems. Applications
Subtitle of host publication17th International Conference, IPMU 2018, Cádiz, Spain, June 11-15, 2018, Proceedings, Part III
EditorsJesús Medina, Manuel Ojeda-Aciego, José Luis Verdegay, Irina Perfilieva, Bernadette Bouchon-Meunier, Ronald R. Yager
Place of PublicationCham
PublisherSpringer Verlag
Pages691-702
Number of pages12
ISBN (Electronic)9783319914794
ISBN (Print)9783319914787
DOIs
Publication statusPublished - 18 May 2018
Event17th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems - Cadiz, Spain
Duration: 11 Jun 201815 Jun 2018
Conference number: 17
http://www.wikicfp.com/cfp/servlet/event.showcfp?eventid=63149&copyownerid=98094 (Link to Conference Information)

Publication series

NameCommunications in Computer and Information Science
Volume855
ISSN (Print)1865-0929

Conference

Conference17th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems
Abbreviated titleIPMU 2018
CountrySpain
CityCadiz
Period11/06/1815/06/18
Internet address

Fingerprint

Predictive Model
Post-processing
Artificial Neural Network
Machine Learning
Neural networks
Missing Values
Pattern Recognition
Learning systems
Disorder
Learning Algorithm
Fold
Monte Carlo Simulation
Trade-offs
Optimise
Neural Networks
Uncertainty
Testing
Methodology
Processing
Model

Cite this

Stamate, D., Alghamdi, W., Stahl, D., Pu, I., Murtagh, F., Belgrave, D., ... di Forti, M. (2018). Predicting First-Episode Psychosis Associated with Cannabis Use with Artificial Neural Networks and Deep Learning. In J. Medina, M. Ojeda-Aciego, J. L. Verdegay, I. Perfilieva, B. Bouchon-Meunier, & R. R. Yager (Eds.), Information Processing and Management of Uncertainty in Knowledge-Based Systems. Applications: 17th International Conference, IPMU 2018, Cádiz, Spain, June 11-15, 2018, Proceedings, Part III (pp. 691-702). (Communications in Computer and Information Science; Vol. 855). Cham: Springer Verlag. https://doi.org/10.1007/978-3-319-91479-4_57
Stamate, Daniel ; Alghamdi, Wajdi ; Stahl, Daniel ; Pu, Ida ; Murtagh, Fionn ; Belgrave, Danielle ; Murray, Robin ; di Forti, Marta. / Predicting First-Episode Psychosis Associated with Cannabis Use with Artificial Neural Networks and Deep Learning. Information Processing and Management of Uncertainty in Knowledge-Based Systems. Applications: 17th International Conference, IPMU 2018, Cádiz, Spain, June 11-15, 2018, Proceedings, Part III. editor / Jesús Medina ; Manuel Ojeda-Aciego ; José Luis Verdegay ; Irina Perfilieva ; Bernadette Bouchon-Meunier ; Ronald R. Yager. Cham : Springer Verlag, 2018. pp. 691-702 (Communications in Computer and Information Science).
@inproceedings{e1e3054bc9f742508bbf780f115eb3d4,
title = "Predicting First-Episode Psychosis Associated with Cannabis Use with Artificial Neural Networks and Deep Learning",
abstract = "In recent years, a number of researches started to investigate the existence of links between cannabis use and psychotic disorder. More recently, artificial neural networks and in particular deep learning have set a revolutionary wave in pattern recognition and machine learning. This study proposes a novel machine learning approach based on neural network and deep learning algorithms, to developing highly accurate predictive models for the onset of first-episode psychosis. Our approach is based also on a novel methodology of optimising and post-processing the predictive models in a computationally intensive framework. A study of the trade-off between the volume of the data and the extent of uncertainty due to missing values, both of which influencing the predictive performance, enhanced this approach. Furthermore, we extended our approach by proposing and encapsulating a novel post-processing k-fold cross-testing method in order to further optimise, and test these models. The results show that the average accuracy in predicting first-episode psychosis achieved by our models in intensive Monte Carlo simulation, is about 89{\%}.",
keywords = "Cannabis Use, Classification, Deep Learning, First-Episode Psychosis, Missing Data Based Uncertainty, Monte Carlo Simulation, Neural Network, Post-Processing, Precision Medicine, Prediction Modelling",
author = "Daniel Stamate and Wajdi Alghamdi and Daniel Stahl and Ida Pu and Fionn Murtagh and Danielle Belgrave and Robin Murray and {di Forti}, Marta",
note = "The final authenticated version is available online at https://doi.org/10.1007/978-3-319-91479-4_57.",
year = "2018",
month = "5",
day = "18",
doi = "10.1007/978-3-319-91479-4_57",
language = "English",
isbn = "9783319914787",
series = "Communications in Computer and Information Science",
publisher = "Springer Verlag",
pages = "691--702",
editor = "Jes{\'u}s Medina and Manuel Ojeda-Aciego and Verdegay, {Jos{\'e} Luis} and Irina Perfilieva and Bernadette Bouchon-Meunier and Yager, {Ronald R.}",
booktitle = "Information Processing and Management of Uncertainty in Knowledge-Based Systems. Applications",

}

Stamate, D, Alghamdi, W, Stahl, D, Pu, I, Murtagh, F, Belgrave, D, Murray, R & di Forti, M 2018, Predicting First-Episode Psychosis Associated with Cannabis Use with Artificial Neural Networks and Deep Learning. in J Medina, M Ojeda-Aciego, JL Verdegay, I Perfilieva, B Bouchon-Meunier & RR Yager (eds), Information Processing and Management of Uncertainty in Knowledge-Based Systems. Applications: 17th International Conference, IPMU 2018, Cádiz, Spain, June 11-15, 2018, Proceedings, Part III. Communications in Computer and Information Science, vol. 855, Springer Verlag, Cham, pp. 691-702, 17th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, Cadiz, Spain, 11/06/18. https://doi.org/10.1007/978-3-319-91479-4_57

Predicting First-Episode Psychosis Associated with Cannabis Use with Artificial Neural Networks and Deep Learning. / Stamate, Daniel; Alghamdi, Wajdi; Stahl, Daniel; Pu, Ida; Murtagh, Fionn; Belgrave, Danielle; Murray, Robin; di Forti, Marta.

Information Processing and Management of Uncertainty in Knowledge-Based Systems. Applications: 17th International Conference, IPMU 2018, Cádiz, Spain, June 11-15, 2018, Proceedings, Part III. ed. / Jesús Medina; Manuel Ojeda-Aciego; José Luis Verdegay; Irina Perfilieva; Bernadette Bouchon-Meunier; Ronald R. Yager. Cham : Springer Verlag, 2018. p. 691-702 (Communications in Computer and Information Science; Vol. 855).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Predicting First-Episode Psychosis Associated with Cannabis Use with Artificial Neural Networks and Deep Learning

AU - Stamate, Daniel

AU - Alghamdi, Wajdi

AU - Stahl, Daniel

AU - Pu, Ida

AU - Murtagh, Fionn

AU - Belgrave, Danielle

AU - Murray, Robin

AU - di Forti, Marta

N1 - The final authenticated version is available online at https://doi.org/10.1007/978-3-319-91479-4_57.

PY - 2018/5/18

Y1 - 2018/5/18

N2 - In recent years, a number of researches started to investigate the existence of links between cannabis use and psychotic disorder. More recently, artificial neural networks and in particular deep learning have set a revolutionary wave in pattern recognition and machine learning. This study proposes a novel machine learning approach based on neural network and deep learning algorithms, to developing highly accurate predictive models for the onset of first-episode psychosis. Our approach is based also on a novel methodology of optimising and post-processing the predictive models in a computationally intensive framework. A study of the trade-off between the volume of the data and the extent of uncertainty due to missing values, both of which influencing the predictive performance, enhanced this approach. Furthermore, we extended our approach by proposing and encapsulating a novel post-processing k-fold cross-testing method in order to further optimise, and test these models. The results show that the average accuracy in predicting first-episode psychosis achieved by our models in intensive Monte Carlo simulation, is about 89%.

AB - In recent years, a number of researches started to investigate the existence of links between cannabis use and psychotic disorder. More recently, artificial neural networks and in particular deep learning have set a revolutionary wave in pattern recognition and machine learning. This study proposes a novel machine learning approach based on neural network and deep learning algorithms, to developing highly accurate predictive models for the onset of first-episode psychosis. Our approach is based also on a novel methodology of optimising and post-processing the predictive models in a computationally intensive framework. A study of the trade-off between the volume of the data and the extent of uncertainty due to missing values, both of which influencing the predictive performance, enhanced this approach. Furthermore, we extended our approach by proposing and encapsulating a novel post-processing k-fold cross-testing method in order to further optimise, and test these models. The results show that the average accuracy in predicting first-episode psychosis achieved by our models in intensive Monte Carlo simulation, is about 89%.

KW - Cannabis Use

KW - Classification

KW - Deep Learning

KW - First-Episode Psychosis

KW - Missing Data Based Uncertainty

KW - Monte Carlo Simulation

KW - Neural Network

KW - Post-Processing

KW - Precision Medicine

KW - Prediction Modelling

UR - http://www.scopus.com/inward/record.url?scp=85048040554&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-91479-4_57

DO - 10.1007/978-3-319-91479-4_57

M3 - Conference contribution

SN - 9783319914787

T3 - Communications in Computer and Information Science

SP - 691

EP - 702

BT - Information Processing and Management of Uncertainty in Knowledge-Based Systems. Applications

A2 - Medina, Jesús

A2 - Ojeda-Aciego, Manuel

A2 - Verdegay, José Luis

A2 - Perfilieva, Irina

A2 - Bouchon-Meunier, Bernadette

A2 - Yager, Ronald R.

PB - Springer Verlag

CY - Cham

ER -

Stamate D, Alghamdi W, Stahl D, Pu I, Murtagh F, Belgrave D et al. Predicting First-Episode Psychosis Associated with Cannabis Use with Artificial Neural Networks and Deep Learning. In Medina J, Ojeda-Aciego M, Verdegay JL, Perfilieva I, Bouchon-Meunier B, Yager RR, editors, Information Processing and Management of Uncertainty in Knowledge-Based Systems. Applications: 17th International Conference, IPMU 2018, Cádiz, Spain, June 11-15, 2018, Proceedings, Part III. Cham: Springer Verlag. 2018. p. 691-702. (Communications in Computer and Information Science). https://doi.org/10.1007/978-3-319-91479-4_57