Artificial intelligence in clinical decision support: challenges for evaluating AI and practical implications

Farah Magrabi, Elske Ammenwerth, Jytte Brender McNair, Nicolet F. De Keizer, Hannele Hyppönen, Pirkko Nykänen, Michael Rigby, Philip J. Scott, Tuulikki Vehko, Zoie Shui Yee Wong, Andrew Georgiou

Research output: Contribution to journalArticleResearchpeer-review

Abstract

Objectives: This paper draws attention to: i) key considerations for evaluating artificial intelligence (AI) enabled clinical decision support; and ii) challenges and practical implications of AI design, development, selection, use, and ongoing surveillance. Method: A narrative review of existing research and evaluation approaches along with expert perspectives drawn from the International Medical Informatics Association (IMIA) Working Group on Technology Assessment and Quality Development in Health Informatics and the European Federation for Medical Informatics (EFMI) Working Group for Assessment of Health Information Systems. Results: There is a rich history and tradition of evaluating AI in healthcare. While evaluators can learn from past efforts, and build on best practice evaluation frameworks and methodologies, questions remain about how to evaluate the safety and effectiveness of AI that dynamically harness vast amounts of genomic, biomarker, phenotype, electronic record, and care delivery data from across health systems. This paper first provides a historical perspective about the evaluation of AI in healthcare. It then examines key challenges of evaluating AI-enabled clinical decision support during design, development, selection, use, and ongoing surveillance. Practical aspects of evaluating AI in healthcare, including approaches to evaluation and indicators to monitor AI are also discussed. Conclusion: Commitment to rigorous initial and ongoing evaluation will be critical to ensuring the safe and effective integration of AI in complex sociotechnical settings. Specific enhancements that are required for the new generation of AI-enabled clinical decision support will emerge through practical application.

LanguageEnglish
Pages128-134
Number of pages7
JournalYearbook of medical informatics
Volume28
Issue number1
DOIs
Publication statusPublished - 1 Aug 2019

Fingerprint

Clinical Decision Support Systems
Artificial Intelligence
Medical Informatics
Delivery of Health Care
Health Information Systems
Biomedical Technology Assessment
Informatics
Internationality
Health
Practice Guidelines
Biomarkers
History

Bibliographical note

Copyright the Publisher 2019. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

Keywords

  • Artificial intelligence
  • machine learning
  • clinical decision support
  • evaluation studies
  • program evaluation

Cite this

Magrabi, Farah ; Ammenwerth, Elske ; McNair, Jytte Brender ; De Keizer, Nicolet F. ; Hyppönen, Hannele ; Nykänen, Pirkko ; Rigby, Michael ; Scott, Philip J. ; Vehko, Tuulikki ; Wong, Zoie Shui Yee ; Georgiou, Andrew. / Artificial intelligence in clinical decision support : challenges for evaluating AI and practical implications. In: Yearbook of medical informatics. 2019 ; Vol. 28, No. 1. pp. 128-134.
@article{297dedc249d44bba90f2c5e855383835,
title = "Artificial intelligence in clinical decision support: challenges for evaluating AI and practical implications",
abstract = "Objectives: This paper draws attention to: i) key considerations for evaluating artificial intelligence (AI) enabled clinical decision support; and ii) challenges and practical implications of AI design, development, selection, use, and ongoing surveillance. Method: A narrative review of existing research and evaluation approaches along with expert perspectives drawn from the International Medical Informatics Association (IMIA) Working Group on Technology Assessment and Quality Development in Health Informatics and the European Federation for Medical Informatics (EFMI) Working Group for Assessment of Health Information Systems. Results: There is a rich history and tradition of evaluating AI in healthcare. While evaluators can learn from past efforts, and build on best practice evaluation frameworks and methodologies, questions remain about how to evaluate the safety and effectiveness of AI that dynamically harness vast amounts of genomic, biomarker, phenotype, electronic record, and care delivery data from across health systems. This paper first provides a historical perspective about the evaluation of AI in healthcare. It then examines key challenges of evaluating AI-enabled clinical decision support during design, development, selection, use, and ongoing surveillance. Practical aspects of evaluating AI in healthcare, including approaches to evaluation and indicators to monitor AI are also discussed. Conclusion: Commitment to rigorous initial and ongoing evaluation will be critical to ensuring the safe and effective integration of AI in complex sociotechnical settings. Specific enhancements that are required for the new generation of AI-enabled clinical decision support will emerge through practical application.",
keywords = "Artificial intelligence, machine learning, clinical decision support, evaluation studies, program evaluation",
author = "Farah Magrabi and Elske Ammenwerth and McNair, {Jytte Brender} and {De Keizer}, {Nicolet F.} and Hannele Hypp{\"o}nen and Pirkko Nyk{\"a}nen and Michael Rigby and Scott, {Philip J.} and Tuulikki Vehko and Wong, {Zoie Shui Yee} and Andrew Georgiou",
note = "Copyright the Publisher 2019. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.",
year = "2019",
month = "8",
day = "1",
doi = "10.1055/s-0039-1677903",
language = "English",
volume = "28",
pages = "128--134",
journal = "Yearbook of medical informatics",
issn = "0943-4747",
number = "1",

}

Magrabi, F, Ammenwerth, E, McNair, JB, De Keizer, NF, Hyppönen, H, Nykänen, P, Rigby, M, Scott, PJ, Vehko, T, Wong, ZSY & Georgiou, A 2019, 'Artificial intelligence in clinical decision support: challenges for evaluating AI and practical implications', Yearbook of medical informatics, vol. 28, no. 1, pp. 128-134. https://doi.org/10.1055/s-0039-1677903

Artificial intelligence in clinical decision support : challenges for evaluating AI and practical implications. / Magrabi, Farah; Ammenwerth, Elske; McNair, Jytte Brender; De Keizer, Nicolet F.; Hyppönen, Hannele; Nykänen, Pirkko; Rigby, Michael; Scott, Philip J.; Vehko, Tuulikki; Wong, Zoie Shui Yee; Georgiou, Andrew.

In: Yearbook of medical informatics, Vol. 28, No. 1, 01.08.2019, p. 128-134.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Artificial intelligence in clinical decision support

T2 - Yearbook of medical informatics

AU - Magrabi, Farah

AU - Ammenwerth, Elske

AU - McNair, Jytte Brender

AU - De Keizer, Nicolet F.

AU - Hyppönen, Hannele

AU - Nykänen, Pirkko

AU - Rigby, Michael

AU - Scott, Philip J.

AU - Vehko, Tuulikki

AU - Wong, Zoie Shui Yee

AU - Georgiou, Andrew

N1 - Copyright the Publisher 2019. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

PY - 2019/8/1

Y1 - 2019/8/1

N2 - Objectives: This paper draws attention to: i) key considerations for evaluating artificial intelligence (AI) enabled clinical decision support; and ii) challenges and practical implications of AI design, development, selection, use, and ongoing surveillance. Method: A narrative review of existing research and evaluation approaches along with expert perspectives drawn from the International Medical Informatics Association (IMIA) Working Group on Technology Assessment and Quality Development in Health Informatics and the European Federation for Medical Informatics (EFMI) Working Group for Assessment of Health Information Systems. Results: There is a rich history and tradition of evaluating AI in healthcare. While evaluators can learn from past efforts, and build on best practice evaluation frameworks and methodologies, questions remain about how to evaluate the safety and effectiveness of AI that dynamically harness vast amounts of genomic, biomarker, phenotype, electronic record, and care delivery data from across health systems. This paper first provides a historical perspective about the evaluation of AI in healthcare. It then examines key challenges of evaluating AI-enabled clinical decision support during design, development, selection, use, and ongoing surveillance. Practical aspects of evaluating AI in healthcare, including approaches to evaluation and indicators to monitor AI are also discussed. Conclusion: Commitment to rigorous initial and ongoing evaluation will be critical to ensuring the safe and effective integration of AI in complex sociotechnical settings. Specific enhancements that are required for the new generation of AI-enabled clinical decision support will emerge through practical application.

AB - Objectives: This paper draws attention to: i) key considerations for evaluating artificial intelligence (AI) enabled clinical decision support; and ii) challenges and practical implications of AI design, development, selection, use, and ongoing surveillance. Method: A narrative review of existing research and evaluation approaches along with expert perspectives drawn from the International Medical Informatics Association (IMIA) Working Group on Technology Assessment and Quality Development in Health Informatics and the European Federation for Medical Informatics (EFMI) Working Group for Assessment of Health Information Systems. Results: There is a rich history and tradition of evaluating AI in healthcare. While evaluators can learn from past efforts, and build on best practice evaluation frameworks and methodologies, questions remain about how to evaluate the safety and effectiveness of AI that dynamically harness vast amounts of genomic, biomarker, phenotype, electronic record, and care delivery data from across health systems. This paper first provides a historical perspective about the evaluation of AI in healthcare. It then examines key challenges of evaluating AI-enabled clinical decision support during design, development, selection, use, and ongoing surveillance. Practical aspects of evaluating AI in healthcare, including approaches to evaluation and indicators to monitor AI are also discussed. Conclusion: Commitment to rigorous initial and ongoing evaluation will be critical to ensuring the safe and effective integration of AI in complex sociotechnical settings. Specific enhancements that are required for the new generation of AI-enabled clinical decision support will emerge through practical application.

KW - Artificial intelligence

KW - machine learning

KW - clinical decision support

KW - evaluation studies

KW - program evaluation

UR - http://www.scopus.com/inward/record.url?scp=85071700730&partnerID=8YFLogxK

U2 - 10.1055/s-0039-1677903

DO - 10.1055/s-0039-1677903

M3 - Article

VL - 28

SP - 128

EP - 134

JO - Yearbook of medical informatics

JF - Yearbook of medical informatics

SN - 0943-4747

IS - 1

ER -