Identifying first episodes of psychosis in psychiatric patient records using machine learning

Genevieve Gorrell; Sherifat Oduola; Angus Roberts; Thomas Craig; Craig Morgan; Robert Stewart

Identifying first episodes of psychosis in psychiatric patient records using machine learning

Genevieve Gorrell, Sherifat Oduola, Angus Roberts, Thomas Craig, Craig Morgan, Robert Stewart

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

3 Citations (Scopus)

Abstract

Natural language processing is being pressed into use to facilitate the selection of cases for medical research in electronic health record databases, though study inclusion criteria may be complex, and the linguistic cues indicating eligibility may be subtle. Finding cases of first episode psychosis raised a number of problems for automated approaches, providing an opportunity to explore how machine learning technologies might be used to overcome them. A system was delivered that achieved an AUC of 0.85, enabling 95% of relevant cases to be identified whilst halving the work required in manually reviewing cases. The techniques that made this possible are presented.

Original language	English
Title of host publication	BioNLP 2016 - Proceedings of the 15th Workshop on Biomedical Natural Language Processing
Editors	Kevin Bretonnel Cohen, Dina Demner-Fushman, Sophia Ananiadou, Jun-ichi Tsujii
Publisher	Association for Computational Linguistics (ACL)
Pages	196-205
Number of pages	10
ISBN (Electronic)	9781945626128
Publication status	Published - 2016
Event	15th Workshop on Biomedical Natural Language Processing, BioNLP 2016 - Berlin, Germany Duration: 12 Aug 2016 → …

Publication series

Name	BioNLP 2016 - Proceedings of the 15th Workshop on Biomedical Natural Language Processing

Conference

Conference	15th Workshop on Biomedical Natural Language Processing, BioNLP 2016
Country/Territory	Germany
City	Berlin
Period	12/08/2016 → …

Cite this

Gorrell, G., Oduola, S., Roberts, A., Craig, T., Morgan, C., & Stewart, R. (2016). Identifying first episodes of psychosis in psychiatric patient records using machine learning. In K. B. Cohen, D. Demner-Fushman, S. Ananiadou, & J. Tsujii (Eds.), BioNLP 2016 - Proceedings of the 15th Workshop on Biomedical Natural Language Processing (pp. 196-205). (BioNLP 2016 - Proceedings of the 15th Workshop on Biomedical Natural Language Processing). Association for Computational Linguistics (ACL).

Gorrell, Genevieve ; Oduola, Sherifat ; Roberts, Angus et al. / Identifying first episodes of psychosis in psychiatric patient records using machine learning. BioNLP 2016 - Proceedings of the 15th Workshop on Biomedical Natural Language Processing. editor / Kevin Bretonnel Cohen ; Dina Demner-Fushman ; Sophia Ananiadou ; Jun-ichi Tsujii. Association for Computational Linguistics (ACL), 2016. pp. 196-205 (BioNLP 2016 - Proceedings of the 15th Workshop on Biomedical Natural Language Processing).

@inbook{223630589f0045cfa9fb71d8dfafbee9,

title = "Identifying first episodes of psychosis in psychiatric patient records using machine learning",

abstract = "Natural language processing is being pressed into use to facilitate the selection of cases for medical research in electronic health record databases, though study inclusion criteria may be complex, and the linguistic cues indicating eligibility may be subtle. Finding cases of first episode psychosis raised a number of problems for automated approaches, providing an opportunity to explore how machine learning technologies might be used to overcome them. A system was delivered that achieved an AUC of 0.85, enabling 95% of relevant cases to be identified whilst halving the work required in manually reviewing cases. The techniques that made this possible are presented.",

author = "Genevieve Gorrell and Sherifat Oduola and Angus Roberts and Thomas Craig and Craig Morgan and Robert Stewart",

year = "2016",

language = "English",

series = "BioNLP 2016 - Proceedings of the 15th Workshop on Biomedical Natural Language Processing",

publisher = "Association for Computational Linguistics (ACL)",

pages = "196--205",

editor = "Cohen, {Kevin Bretonnel} and Dina Demner-Fushman and Sophia Ananiadou and Jun-ichi Tsujii",

booktitle = "BioNLP 2016 - Proceedings of the 15th Workshop on Biomedical Natural Language Processing",

}

Gorrell, G, Oduola, S , Roberts, A , Craig, T , Morgan, C & Stewart, R 2016, Identifying first episodes of psychosis in psychiatric patient records using machine learning. in KB Cohen, D Demner-Fushman, S Ananiadou & J Tsujii (eds), BioNLP 2016 - Proceedings of the 15th Workshop on Biomedical Natural Language Processing. BioNLP 2016 - Proceedings of the 15th Workshop on Biomedical Natural Language Processing, Association for Computational Linguistics (ACL), pp. 196-205, 15th Workshop on Biomedical Natural Language Processing, BioNLP 2016, Berlin, Germany, 12/08/2016.

Identifying first episodes of psychosis in psychiatric patient records using machine learning. / Gorrell, Genevieve; Oduola, Sherifat ; Roberts, Angus et al.
BioNLP 2016 - Proceedings of the 15th Workshop on Biomedical Natural Language Processing. ed. / Kevin Bretonnel Cohen; Dina Demner-Fushman; Sophia Ananiadou; Jun-ichi Tsujii. Association for Computational Linguistics (ACL), 2016. p. 196-205 (BioNLP 2016 - Proceedings of the 15th Workshop on Biomedical Natural Language Processing).

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

TY - CHAP

T1 - Identifying first episodes of psychosis in psychiatric patient records using machine learning

AU - Gorrell, Genevieve

AU - Oduola, Sherifat

AU - Roberts, Angus

AU - Craig, Thomas

AU - Morgan, Craig

AU - Stewart, Robert

PY - 2016

Y1 - 2016

N2 - Natural language processing is being pressed into use to facilitate the selection of cases for medical research in electronic health record databases, though study inclusion criteria may be complex, and the linguistic cues indicating eligibility may be subtle. Finding cases of first episode psychosis raised a number of problems for automated approaches, providing an opportunity to explore how machine learning technologies might be used to overcome them. A system was delivered that achieved an AUC of 0.85, enabling 95% of relevant cases to be identified whilst halving the work required in manually reviewing cases. The techniques that made this possible are presented.

AB - Natural language processing is being pressed into use to facilitate the selection of cases for medical research in electronic health record databases, though study inclusion criteria may be complex, and the linguistic cues indicating eligibility may be subtle. Finding cases of first episode psychosis raised a number of problems for automated approaches, providing an opportunity to explore how machine learning technologies might be used to overcome them. A system was delivered that achieved an AUC of 0.85, enabling 95% of relevant cases to be identified whilst halving the work required in manually reviewing cases. The techniques that made this possible are presented.

UR - http://www.scopus.com/inward/record.url?scp=85065919845&partnerID=8YFLogxK

M3 - Conference paper

AN - SCOPUS:85065919845

T3 - BioNLP 2016 - Proceedings of the 15th Workshop on Biomedical Natural Language Processing

SP - 196

EP - 205

BT - BioNLP 2016 - Proceedings of the 15th Workshop on Biomedical Natural Language Processing

A2 - Cohen, Kevin Bretonnel

A2 - Demner-Fushman, Dina

A2 - Ananiadou, Sophia

A2 - Tsujii, Jun-ichi

PB - Association for Computational Linguistics (ACL)

T2 - 15th Workshop on Biomedical Natural Language Processing, BioNLP 2016

Y2 - 12 August 2016

ER -

Gorrell G, Oduola S , Roberts A , Craig T , Morgan C , Stewart R. Identifying first episodes of psychosis in psychiatric patient records using machine learning. In Cohen KB, Demner-Fushman D, Ananiadou S, Tsujii J, editors, BioNLP 2016 - Proceedings of the 15th Workshop on Biomedical Natural Language Processing. Association for Computational Linguistics (ACL). 2016. p. 196-205. (BioNLP 2016 - Proceedings of the 15th Workshop on Biomedical Natural Language Processing).

Identifying first episodes of psychosis in psychiatric patient records using machine learning

Abstract

Publication series

Conference

Other files and links

Fingerprint

Cite this