An adversorial approach to enable re-use of machine learning models and collaborative research efforts using synthetic unstructured free-text medical data

Suranga N. Kasthurirathne, Gregory Dexter, Shaun Grannis

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

Abstract

We leverage Generative Adversarial Networks (GAN) to produce synthetic free-text medical data with low re-identification risk, and apply these to replicate machine learning solutions. We trained GAN models to generate free-text cancer pathology reports. Decision models were trained using synthetic datasets reported performance metrics that were statistically similar to models trained using original test data. Our results further the use of GANs to generate synthetic data for collaborative research and re-use of machine learning models.

Original languageEnglish (US)
Title of host publicationMEDINFO 2019
Subtitle of host publicationHealth and Wellbeing e-Networks for All - Proceedings of the 17th World Congress on Medical and Health Informatics
EditorsBrigitte Seroussi, Lucila Ohno-Machado, Lucila Ohno-Machado, Brigitte Seroussi
PublisherIOS Press
Pages1510-1511
Number of pages2
ISBN (Electronic)9781643680026
DOIs
StatePublished - Aug 21 2019
Event17th World Congress on Medical and Health Informatics, MEDINFO 2019 - Lyon, France
Duration: Aug 25 2019Aug 30 2019

Publication series

NameStudies in Health Technology and Informatics
Volume264
ISSN (Print)0926-9630
ISSN (Electronic)1879-8365

Conference

Conference17th World Congress on Medical and Health Informatics, MEDINFO 2019
CountryFrance
CityLyon
Period8/25/198/30/19

Fingerprint

Learning systems
Research
Pathology
Neoplasms
Identification (control systems)
Machine Learning
Datasets

Keywords

  • Dataset
  • Machine Learning
  • Neural Networks (Computer)

ASJC Scopus subject areas

  • Biomedical Engineering
  • Health Informatics
  • Health Information Management

Cite this

Kasthurirathne, S. N., Dexter, G., & Grannis, S. (2019). An adversorial approach to enable re-use of machine learning models and collaborative research efforts using synthetic unstructured free-text medical data. In B. Seroussi, L. Ohno-Machado, L. Ohno-Machado, & B. Seroussi (Eds.), MEDINFO 2019: Health and Wellbeing e-Networks for All - Proceedings of the 17th World Congress on Medical and Health Informatics (pp. 1510-1511). (Studies in Health Technology and Informatics; Vol. 264). IOS Press. https://doi.org/10.3233/SHTI190509

An adversorial approach to enable re-use of machine learning models and collaborative research efforts using synthetic unstructured free-text medical data. / Kasthurirathne, Suranga N.; Dexter, Gregory; Grannis, Shaun.

MEDINFO 2019: Health and Wellbeing e-Networks for All - Proceedings of the 17th World Congress on Medical and Health Informatics. ed. / Brigitte Seroussi; Lucila Ohno-Machado; Lucila Ohno-Machado; Brigitte Seroussi. IOS Press, 2019. p. 1510-1511 (Studies in Health Technology and Informatics; Vol. 264).

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

Kasthurirathne, SN, Dexter, G & Grannis, S 2019, An adversorial approach to enable re-use of machine learning models and collaborative research efforts using synthetic unstructured free-text medical data. in B Seroussi, L Ohno-Machado, L Ohno-Machado & B Seroussi (eds), MEDINFO 2019: Health and Wellbeing e-Networks for All - Proceedings of the 17th World Congress on Medical and Health Informatics. Studies in Health Technology and Informatics, vol. 264, IOS Press, pp. 1510-1511, 17th World Congress on Medical and Health Informatics, MEDINFO 2019, Lyon, France, 8/25/19. https://doi.org/10.3233/SHTI190509
Kasthurirathne SN, Dexter G, Grannis S. An adversorial approach to enable re-use of machine learning models and collaborative research efforts using synthetic unstructured free-text medical data. In Seroussi B, Ohno-Machado L, Ohno-Machado L, Seroussi B, editors, MEDINFO 2019: Health and Wellbeing e-Networks for All - Proceedings of the 17th World Congress on Medical and Health Informatics. IOS Press. 2019. p. 1510-1511. (Studies in Health Technology and Informatics). https://doi.org/10.3233/SHTI190509
Kasthurirathne, Suranga N. ; Dexter, Gregory ; Grannis, Shaun. / An adversorial approach to enable re-use of machine learning models and collaborative research efforts using synthetic unstructured free-text medical data. MEDINFO 2019: Health and Wellbeing e-Networks for All - Proceedings of the 17th World Congress on Medical and Health Informatics. editor / Brigitte Seroussi ; Lucila Ohno-Machado ; Lucila Ohno-Machado ; Brigitte Seroussi. IOS Press, 2019. pp. 1510-1511 (Studies in Health Technology and Informatics).
@inproceedings{1015ccf9a9b443028883445876602687,
title = "An adversorial approach to enable re-use of machine learning models and collaborative research efforts using synthetic unstructured free-text medical data",
abstract = "We leverage Generative Adversarial Networks (GAN) to produce synthetic free-text medical data with low re-identification risk, and apply these to replicate machine learning solutions. We trained GAN models to generate free-text cancer pathology reports. Decision models were trained using synthetic datasets reported performance metrics that were statistically similar to models trained using original test data. Our results further the use of GANs to generate synthetic data for collaborative research and re-use of machine learning models.",
keywords = "Dataset, Machine Learning, Neural Networks (Computer)",
author = "Kasthurirathne, {Suranga N.} and Gregory Dexter and Shaun Grannis",
year = "2019",
month = "8",
day = "21",
doi = "10.3233/SHTI190509",
language = "English (US)",
series = "Studies in Health Technology and Informatics",
publisher = "IOS Press",
pages = "1510--1511",
editor = "Brigitte Seroussi and Lucila Ohno-Machado and Lucila Ohno-Machado and Brigitte Seroussi",
booktitle = "MEDINFO 2019",
address = "United States",

}

TY - GEN

T1 - An adversorial approach to enable re-use of machine learning models and collaborative research efforts using synthetic unstructured free-text medical data

AU - Kasthurirathne, Suranga N.

AU - Dexter, Gregory

AU - Grannis, Shaun

PY - 2019/8/21

Y1 - 2019/8/21

N2 - We leverage Generative Adversarial Networks (GAN) to produce synthetic free-text medical data with low re-identification risk, and apply these to replicate machine learning solutions. We trained GAN models to generate free-text cancer pathology reports. Decision models were trained using synthetic datasets reported performance metrics that were statistically similar to models trained using original test data. Our results further the use of GANs to generate synthetic data for collaborative research and re-use of machine learning models.

AB - We leverage Generative Adversarial Networks (GAN) to produce synthetic free-text medical data with low re-identification risk, and apply these to replicate machine learning solutions. We trained GAN models to generate free-text cancer pathology reports. Decision models were trained using synthetic datasets reported performance metrics that were statistically similar to models trained using original test data. Our results further the use of GANs to generate synthetic data for collaborative research and re-use of machine learning models.

KW - Dataset

KW - Machine Learning

KW - Neural Networks (Computer)

UR - http://www.scopus.com/inward/record.url?scp=85071498564&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071498564&partnerID=8YFLogxK

U2 - 10.3233/SHTI190509

DO - 10.3233/SHTI190509

M3 - Conference contribution

C2 - 31438206

AN - SCOPUS:85071498564

T3 - Studies in Health Technology and Informatics

SP - 1510

EP - 1511

BT - MEDINFO 2019

A2 - Seroussi, Brigitte

A2 - Ohno-Machado, Lucila

A2 - Ohno-Machado, Lucila

A2 - Seroussi, Brigitte

PB - IOS Press

ER -