Distributed Learning to Protect Privacy in Multi-centric Clinical Studies

Andrea Damiani, Mauro Vallati, Roberto Gatta, Nicola Dinapoli, Arthur Jochems, Timo Deist, Johan van Soest, Andre Dekker, Vincenzo Valentini

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

6 Citations (Scopus)

Abstract

Research in medicine has to deal with the growing amount of data about patients which are made available by modern technologies. All these data might be used to support statistical studies, and for identifying causal relations. To use these data which are spread across hospitals efficient merging techniques, as well as policies to deal with this sensitive information, are strongly needed. In this paper we introduce and empirically test a distributed learning approach, to train Support Vector Machines (SVM), that allows to overcome problems related to privacy and data being spread around. The introduced technique allows to train algorithms without sharing any patients-related information, ensuring privacy and avoids the development of merging tools. We tested this approach on a large dataset and we described results, in terms of convergence and performance; we also provide considerations about the features of an IT architecture designed to support distributed learning computations.
Original languageEnglish
Title of host publicationArtificial Intelligence in Medicine
Subtitle of host publication15th Conference on Artificial Intelligence in Medicine, AIME 2015, Pavia, Italy, June 17-20, 2015. Proceedings
EditorsJohn H. Holmes, Riccardo Bellazzi, Lucia Sacchi, Niels Peek
PublisherSpringer, Cham
Pages65-75
Number of pages11
Volume9105
ISBN (Electronic)9783319195513
ISBN (Print)9783319195506
DOIs
Publication statusPublished - 2015
Event15th Conference on Artificial Intelligence in Medicine - Pavia, Italy
Duration: 17 Jun 201520 Jun 2015
Conference number: 15
http://aime15.aimedicine.info/ (Link to Conference Website)

Publication series

NameLecture Notes in Computer Science
PublisherSpringer International
Volume9105
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference15th Conference on Artificial Intelligence in Medicine
Abbreviated titleAIME 2015
CountryItaly
CityPavia
Period17/06/1520/06/15
Internet address

Fingerprint

Merging
Medicine
Support vector machines

Cite this

Damiani, A., Vallati, M., Gatta, R., Dinapoli, N., Jochems, A., Deist, T., ... Valentini, V. (2015). Distributed Learning to Protect Privacy in Multi-centric Clinical Studies. In J. H. Holmes, R. Bellazzi, L. Sacchi, & N. Peek (Eds.), Artificial Intelligence in Medicine: 15th Conference on Artificial Intelligence in Medicine, AIME 2015, Pavia, Italy, June 17-20, 2015. Proceedings (Vol. 9105, pp. 65-75). (Lecture Notes in Computer Science; Vol. 9105). Springer, Cham. https://doi.org/10.1007/978-3-319-19551-3_8
Damiani, Andrea ; Vallati, Mauro ; Gatta, Roberto ; Dinapoli, Nicola ; Jochems, Arthur ; Deist, Timo ; van Soest, Johan ; Dekker, Andre ; Valentini, Vincenzo. / Distributed Learning to Protect Privacy in Multi-centric Clinical Studies. Artificial Intelligence in Medicine: 15th Conference on Artificial Intelligence in Medicine, AIME 2015, Pavia, Italy, June 17-20, 2015. Proceedings. editor / John H. Holmes ; Riccardo Bellazzi ; Lucia Sacchi ; Niels Peek. Vol. 9105 Springer, Cham, 2015. pp. 65-75 (Lecture Notes in Computer Science).
@inproceedings{ba510ac8a6a346a98f626e695d42b7cd,
title = "Distributed Learning to Protect Privacy in Multi-centric Clinical Studies",
abstract = "Research in medicine has to deal with the growing amount of data about patients which are made available by modern technologies. All these data might be used to support statistical studies, and for identifying causal relations. To use these data which are spread across hospitals efficient merging techniques, as well as policies to deal with this sensitive information, are strongly needed. In this paper we introduce and empirically test a distributed learning approach, to train Support Vector Machines (SVM), that allows to overcome problems related to privacy and data being spread around. The introduced technique allows to train algorithms without sharing any patients-related information, ensuring privacy and avoids the development of merging tools. We tested this approach on a large dataset and we described results, in terms of convergence and performance; we also provide considerations about the features of an IT architecture designed to support distributed learning computations.",
keywords = "Distributed learning, Multi-centric clinical studies, Patient privacy preserving",
author = "Andrea Damiani and Mauro Vallati and Roberto Gatta and Nicola Dinapoli and Arthur Jochems and Timo Deist and {van Soest}, Johan and Andre Dekker and Vincenzo Valentini",
year = "2015",
doi = "10.1007/978-3-319-19551-3_8",
language = "English",
isbn = "9783319195506",
volume = "9105",
series = "Lecture Notes in Computer Science",
publisher = "Springer, Cham",
pages = "65--75",
editor = "Holmes, {John H.} and Riccardo Bellazzi and Lucia Sacchi and Niels Peek",
booktitle = "Artificial Intelligence in Medicine",

}

Damiani, A, Vallati, M, Gatta, R, Dinapoli, N, Jochems, A, Deist, T, van Soest, J, Dekker, A & Valentini, V 2015, Distributed Learning to Protect Privacy in Multi-centric Clinical Studies. in JH Holmes, R Bellazzi, L Sacchi & N Peek (eds), Artificial Intelligence in Medicine: 15th Conference on Artificial Intelligence in Medicine, AIME 2015, Pavia, Italy, June 17-20, 2015. Proceedings. vol. 9105, Lecture Notes in Computer Science, vol. 9105, Springer, Cham, pp. 65-75, 15th Conference on Artificial Intelligence in Medicine, Pavia, Italy, 17/06/15. https://doi.org/10.1007/978-3-319-19551-3_8

Distributed Learning to Protect Privacy in Multi-centric Clinical Studies. / Damiani, Andrea; Vallati, Mauro; Gatta, Roberto; Dinapoli, Nicola; Jochems, Arthur; Deist, Timo; van Soest, Johan; Dekker, Andre; Valentini, Vincenzo.

Artificial Intelligence in Medicine: 15th Conference on Artificial Intelligence in Medicine, AIME 2015, Pavia, Italy, June 17-20, 2015. Proceedings. ed. / John H. Holmes; Riccardo Bellazzi; Lucia Sacchi; Niels Peek. Vol. 9105 Springer, Cham, 2015. p. 65-75 (Lecture Notes in Computer Science; Vol. 9105).

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

TY - GEN

T1 - Distributed Learning to Protect Privacy in Multi-centric Clinical Studies

AU - Damiani, Andrea

AU - Vallati, Mauro

AU - Gatta, Roberto

AU - Dinapoli, Nicola

AU - Jochems, Arthur

AU - Deist, Timo

AU - van Soest, Johan

AU - Dekker, Andre

AU - Valentini, Vincenzo

PY - 2015

Y1 - 2015

N2 - Research in medicine has to deal with the growing amount of data about patients which are made available by modern technologies. All these data might be used to support statistical studies, and for identifying causal relations. To use these data which are spread across hospitals efficient merging techniques, as well as policies to deal with this sensitive information, are strongly needed. In this paper we introduce and empirically test a distributed learning approach, to train Support Vector Machines (SVM), that allows to overcome problems related to privacy and data being spread around. The introduced technique allows to train algorithms without sharing any patients-related information, ensuring privacy and avoids the development of merging tools. We tested this approach on a large dataset and we described results, in terms of convergence and performance; we also provide considerations about the features of an IT architecture designed to support distributed learning computations.

AB - Research in medicine has to deal with the growing amount of data about patients which are made available by modern technologies. All these data might be used to support statistical studies, and for identifying causal relations. To use these data which are spread across hospitals efficient merging techniques, as well as policies to deal with this sensitive information, are strongly needed. In this paper we introduce and empirically test a distributed learning approach, to train Support Vector Machines (SVM), that allows to overcome problems related to privacy and data being spread around. The introduced technique allows to train algorithms without sharing any patients-related information, ensuring privacy and avoids the development of merging tools. We tested this approach on a large dataset and we described results, in terms of convergence and performance; we also provide considerations about the features of an IT architecture designed to support distributed learning computations.

KW - Distributed learning

KW - Multi-centric clinical studies

KW - Patient privacy preserving

UR - http://www.scopus.com/inward/record.url?scp=84947929379&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-19551-3_8

DO - 10.1007/978-3-319-19551-3_8

M3 - Conference contribution

SN - 9783319195506

VL - 9105

T3 - Lecture Notes in Computer Science

SP - 65

EP - 75

BT - Artificial Intelligence in Medicine

A2 - Holmes, John H.

A2 - Bellazzi, Riccardo

A2 - Sacchi, Lucia

A2 - Peek, Niels

PB - Springer, Cham

ER -

Damiani A, Vallati M, Gatta R, Dinapoli N, Jochems A, Deist T et al. Distributed Learning to Protect Privacy in Multi-centric Clinical Studies. In Holmes JH, Bellazzi R, Sacchi L, Peek N, editors, Artificial Intelligence in Medicine: 15th Conference on Artificial Intelligence in Medicine, AIME 2015, Pavia, Italy, June 17-20, 2015. Proceedings. Vol. 9105. Springer, Cham. 2015. p. 65-75. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-319-19551-3_8