Transparency and reproducibility in artificial intelligence

doi:10.1038/s41586-020-2766-y

Comment

. 2020 Oct;586(7829):E14-E16.

doi: 10.1038/s41586-020-2766-y. Epub 2020 Oct 14.

Transparency and reproducibility in artificial intelligence

Benjamin Haibe-Kains^{1

2

3

4

5}, George Alexandru Adam^{6

7}, Ahmed Hosny^{8

9}, Farnoosh Khodakarami^{10

11}; Massive Analysis Quality Control (MAQC) Society Board of Directors; Levi Waldron¹², Bo Wang^{11

6

7

13

14}, Chris McIntosh^{11

7

13}, Anna Goldenberg^{6

7

15

16}, Anshul Kundaje^{17

18}, Casey S Greene^{19

20}, Tamara Broderick²¹, Michael M Hoffman^{10

11

6

7}, Jeffrey T Leek²², Keegan Korthauer^{23

24}, Wolfgang Huber²⁵, Alvis Brazma²⁶, Joelle Pineau^{27

28}, Robert Tibshirani^{29

30}, Trevor Hastie^{29

30}, John P A Ioannidis^{29

30

31

32

33}, John Quackenbush^{34

35

36}, Hugo J W L Aerts^{8

9

37

38}

Collaborators, Affiliations

Collaborators

Massive Analysis Quality Control (MAQC) Society Board of Directors:
Thakkar Shraddha, Rebecca Kusko, Susanna-Assunta Sansone, Weida Tong, Russ D Wolfinger, Christopher E Mason, Wendell Jones, Joaquin Dopazo, Cesare Furlanello

Affiliations

¹ Princess Margaret Cancer Centre, University Health Network, Toronto, Ontario, Canada. bhaibeka@uhnresearch.ca.
² Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada. bhaibeka@uhnresearch.ca.
³ Department of Computer Science, University of Toronto, Toronto, Ontario, Canada. bhaibeka@uhnresearch.ca.
⁴ Ontario Institute for Cancer Research, Toronto, Ontario, Canada. bhaibeka@uhnresearch.ca.
⁵ Vector Institute for Artificial Intelligence, Toronto, Ontario, Canada. bhaibeka@uhnresearch.ca.
⁶ Department of Computer Science, University of Toronto, Toronto, Ontario, Canada.
⁷ Vector Institute for Artificial Intelligence, Toronto, Ontario, Canada.
⁸ Artificial Intelligence in Medicine (AIM) Program, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA.
⁹ Radiation Oncology and Radiology, Dana-Farber Cancer Institute, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA.
¹⁰ Princess Margaret Cancer Centre, University Health Network, Toronto, Ontario, Canada.
¹¹ Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada.
¹² Department of Epidemiology and Biostatistics and Institute for Implementation Science in Population Health, CUNY Graduate School of Public Health and Health Policy, New York, NY, USA.
¹³ Peter Munk Cardiac Centre, University Health Network, Toronto, Ontario, Canada.
¹⁴ Department of Laboratory Medicine and Pathobiology, University of Toronto, Ontario, Canada.
¹⁵ SickKids Research Institute, Toronto, Ontario, Canada.
¹⁶ Child and Brain Development Program, CIFAR, Toronto, Ontario, Canada.
¹⁷ Department of Genetics, Stanford University School of Medicine, Stanford, CA, USA.
¹⁸ Department of Computer Science, Stanford University, Stanford, CA, USA.
¹⁹ Dept. of Systems Pharmacology and Translational Therapeutics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA.
²⁰ Childhood Cancer Data Lab, Alex's Lemonade Stand Foundation, Philadelphia, PA, USA.
²¹ Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, USA.
²² Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA.
²³ Department of Statistics, University of British Columbia, Vancouver, British Columbia, Canada.
²⁴ BC Children's Hospital Research Institute, Vancouver, British Columbia, Canada.
²⁵ European Molecular Biology Laboratory, Genome Biology Unit, Heidelberg, Germany.
²⁶ European Molecular Biology Laboratory, European Bioinformatics Institute, EMBL-EBI, Hinxton, UK.
²⁷ McGill University, Montreal, Quebec, Canada.
²⁸ Montreal Institute for Learning Algorithms, Quebec, Canada.
²⁹ Department of Statistics, Stanford University School of Humanities and Sciences, Stanford, CA, USA.
³⁰ Department of Biomedical Data Science, Stanford University School of Medicine, Stanford, CA, USA.
³¹ Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA.
³² Meta-Research Innovation Center at Stanford (METRICS), Stanford, CA, USA.
³³ Department of Epidemiology and Population Health, Stanford University School of Medicine, Stanford, CA, USA.
³⁴ Department of Biostatistics, Harvard T.H Chan School of Public Health, Boston, MA, USA.
³⁵ Channing Division of Network Medicine, Brigham and Women's Hospital, Boston, MA, USA.
³⁶ Department of Data Science, Dana-Farber Cancer Institute, Boston, MA, USA.
³⁷ Radiology and Nuclear Medicine, Maastricht University, Maastricht, The Netherlands.
³⁸ Cardiovascular Imaging Research Center, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA.

PMID: 33057217
PMCID: PMC8144864
DOI: 10.1038/s41586-020-2766-y

Comment

Transparency and reproducibility in artificial intelligence

Benjamin Haibe-Kains et al. Nature. 2020 Oct.

. 2020 Oct;586(7829):E14-E16.

doi: 10.1038/s41586-020-2766-y. Epub 2020 Oct 14.

Authors

Collaborators

Massive Analysis Quality Control (MAQC) Society Board of Directors:
Thakkar Shraddha, Rebecca Kusko, Susanna-Assunta Sansone, Weida Tong, Russ D Wolfinger, Christopher E Mason, Wendell Jones, Joaquin Dopazo, Cesare Furlanello

Affiliations

¹ Princess Margaret Cancer Centre, University Health Network, Toronto, Ontario, Canada. bhaibeka@uhnresearch.ca.
² Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada. bhaibeka@uhnresearch.ca.
³ Department of Computer Science, University of Toronto, Toronto, Ontario, Canada. bhaibeka@uhnresearch.ca.
⁴ Ontario Institute for Cancer Research, Toronto, Ontario, Canada. bhaibeka@uhnresearch.ca.
⁵ Vector Institute for Artificial Intelligence, Toronto, Ontario, Canada. bhaibeka@uhnresearch.ca.
⁶ Department of Computer Science, University of Toronto, Toronto, Ontario, Canada.
⁷ Vector Institute for Artificial Intelligence, Toronto, Ontario, Canada.
⁸ Artificial Intelligence in Medicine (AIM) Program, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA.
⁹ Radiation Oncology and Radiology, Dana-Farber Cancer Institute, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA.
¹⁰ Princess Margaret Cancer Centre, University Health Network, Toronto, Ontario, Canada.
¹¹ Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada.
¹² Department of Epidemiology and Biostatistics and Institute for Implementation Science in Population Health, CUNY Graduate School of Public Health and Health Policy, New York, NY, USA.
¹³ Peter Munk Cardiac Centre, University Health Network, Toronto, Ontario, Canada.
¹⁴ Department of Laboratory Medicine and Pathobiology, University of Toronto, Ontario, Canada.
¹⁵ SickKids Research Institute, Toronto, Ontario, Canada.
¹⁶ Child and Brain Development Program, CIFAR, Toronto, Ontario, Canada.
¹⁷ Department of Genetics, Stanford University School of Medicine, Stanford, CA, USA.
¹⁸ Department of Computer Science, Stanford University, Stanford, CA, USA.
¹⁹ Dept. of Systems Pharmacology and Translational Therapeutics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA.
²⁰ Childhood Cancer Data Lab, Alex's Lemonade Stand Foundation, Philadelphia, PA, USA.
²¹ Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, USA.
²² Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA.
²³ Department of Statistics, University of British Columbia, Vancouver, British Columbia, Canada.
²⁴ BC Children's Hospital Research Institute, Vancouver, British Columbia, Canada.
²⁵ European Molecular Biology Laboratory, Genome Biology Unit, Heidelberg, Germany.
²⁶ European Molecular Biology Laboratory, European Bioinformatics Institute, EMBL-EBI, Hinxton, UK.
²⁷ McGill University, Montreal, Quebec, Canada.
²⁸ Montreal Institute for Learning Algorithms, Quebec, Canada.
²⁹ Department of Statistics, Stanford University School of Humanities and Sciences, Stanford, CA, USA.
³⁰ Department of Biomedical Data Science, Stanford University School of Medicine, Stanford, CA, USA.
³¹ Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA.
³² Meta-Research Innovation Center at Stanford (METRICS), Stanford, CA, USA.
³³ Department of Epidemiology and Population Health, Stanford University School of Medicine, Stanford, CA, USA.
³⁴ Department of Biostatistics, Harvard T.H Chan School of Public Health, Boston, MA, USA.
³⁵ Channing Division of Network Medicine, Brigham and Women's Hospital, Boston, MA, USA.
³⁶ Department of Data Science, Dana-Farber Cancer Institute, Boston, MA, USA.
³⁷ Radiology and Nuclear Medicine, Maastricht University, Maastricht, The Netherlands.
³⁸ Cardiovascular Imaging Research Center, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA.

PMID: 33057217
PMCID: PMC8144864
DOI: 10.1038/s41586-020-2766-y

Abstract

Breakthroughs in artificial intelligence (AI) hold enormous potential as it can automate complex tasks and go even beyond human performance. In their study, McKinney et al. showed the high potential of AI for breast cancer screening. However, the lack of methods’ details and algorithm code undermines its scientific value. Here, we identify obstacles hindering transparent and reproducible AI research as faced by McKinney et al., and provide solutions to these obstacles with implications for the broader field.

PubMed Disclaimer

Comment in

Reply to: Transparency and reproducibility in artificial intelligence.
McKinney SM, Karthikesalingam A, Tse D, Kelly CJ, Liu Y, Corrado GS, Shetty S. McKinney SM, et al. Nature. 2020 Oct;586(7829):E17-E18. doi: 10.1038/s41586-020-2767-x. Nature. 2020. PMID: 33057218 No abstract available.

Comment on

International evaluation of an AI system for breast cancer screening.
McKinney SM, Sieniek M, Godbole V, Godwin J, Antropova N, Ashrafian H, Back T, Chesus M, Corrado GS, Darzi A, Etemadi M, Garcia-Vicente F, Gilbert FJ, Halling-Brown M, Hassabis D, Jansen S, Karthikesalingam A, Kelly CJ, King D, Ledsam JR, Melnick D, Mostofi H, Peng L, Reicher JJ, Romera-Paredes B, Sidebottom R, Suleyman M, Tse D, Young KC, De Fauw J, Shetty S. McKinney SM, et al. Nature. 2020 Jan;577(7788):89-94. doi: 10.1038/s41586-019-1799-6. Epub 2020 Jan 1. Nature. 2020. PMID: 31894144

Cited by

Deep learning-based Alzheimer's disease detection: reproducibility and the effect of modeling choices.
Turrisi R, Verri A, Barla A. Turrisi R, et al. Front Comput Neurosci. 2024 Sep 20;18:1360095. doi: 10.3389/fncom.2024.1360095. eCollection 2024. Front Comput Neurosci. 2024. PMID: 39371524 Free PMC article.
Artificial Intelligence to Close the Gap between Pharmacokinetic/Pharmacodynamic Targets and Clinical Outcomes in Critically Ill Patients: A Narrative Review on Beta Lactams.
Gonçalves Pereira J, Fernandes J, Mendes T, Gonzalez FA, Fernandes SM. Gonçalves Pereira J, et al. Antibiotics (Basel). 2024 Sep 6;13(9):853. doi: 10.3390/antibiotics13090853. Antibiotics (Basel). 2024. PMID: 39335027 Free PMC article. Review.
Expression of microRNA induced by postoperative delirium-like behavior is associated with long-term default mode network disruption: Sequencing and a secondary analysis of resting-state fMRI data.
Liu Y, Feng H, Fu H, Nie B, Wang T. Liu Y, et al. CNS Neurosci Ther. 2024 Sep;30(9):e70038. doi: 10.1111/cns.70038. CNS Neurosci Ther. 2024. PMID: 39317458 Free PMC article.
Early Prediction of Cardiac Arrest in the Intensive Care Unit Using Explainable Machine Learning: Retrospective Study.
Kim YK, Seo WD, Lee SJ, Koo JH, Kim GC, Song HS, Lee M. Kim YK, et al. J Med Internet Res. 2024 Sep 17;26:e62890. doi: 10.2196/62890. J Med Internet Res. 2024. PMID: 39288404 Free PMC article.
ENCORE: a practical implementation to improve reproducibility and transparency of computational research.
van Kampen AHC, Mahamune U, Jongejan A, van Schaik BDC, Balashova D, Lashgari D, Pras-Raves M, Wever EJM, Dane AD, García-Valiente R, Moerland PD. van Kampen AHC, et al. Nat Commun. 2024 Sep 16;15(1):8117. doi: 10.1038/s41467-024-52446-8. Nat Commun. 2024. PMID: 39284801 Free PMC article.

See all "Cited by" articles

References

1. McKinney SM, Sieniek M, Godbole V & Godwin J. International evaluation of an AI system for breast cancer screening. Nature (2020). - PubMed
1. Nature Research Editorial Policies. Reporting standards and availability of data, materials, code and protocols. Springer Nature; https://www.nature.com/nature-research/editorial-policies/reporting-stan....
1. Bluemke DA et al. Assessing Radiology Research on Artificial Intelligence: A Brief Guide for Authors, Reviewers, and Readers—From the Radiology Editorial Board. Radiology 192515 (2019) doi:10.1148/radiol.2019192515. - DOI - PubMed
1. Gundersen OE, Gil Y & Aha DW On reproducible AI: Towards reproducible research, open science, and digital scholarship in AI publications. AI Magazine 39, 56–68 (2018).
1. Crane M. Questionable Answers in Question Answering Research: Reproducibility and Variability of Published Results. Transactions of the Association for Computational Linguistics 6, 241–252 (2018).

Publication types

Actions
Actions

MeSH terms

Actions
Actions
Actions

Grants and funding

LinkOut - more resources

Full Text Sources

[1] McKinney SM, Sieniek M, Godbole V & Godwin J. International evaluation of an AI system for breast cancer screening. Nature (2020). - PubMed

[2] McKinney SM, Sieniek M, Godbole V & Godwin J. International evaluation of an AI system for breast cancer screening. Nature (2020). - PubMed

[3] Nature Research Editorial Policies. Reporting standards and availability of data, materials, code and protocols. Springer Nature; https://www.nature.com/nature-research/editorial-policies/reporting-stan....

[4] Nature Research Editorial Policies. Reporting standards and availability of data, materials, code and protocols. Springer Nature; https://www.nature.com/nature-research/editorial-policies/reporting-stan....

[5] Bluemke DA et al. Assessing Radiology Research on Artificial Intelligence: A Brief Guide for Authors, Reviewers, and Readers—From the Radiology Editorial Board. Radiology 192515 (2019) doi:10.1148/radiol.2019192515. - DOI - PubMed

[6] Bluemke DA et al. Assessing Radiology Research on Artificial Intelligence: A Brief Guide for Authors, Reviewers, and Readers—From the Radiology Editorial Board. Radiology 192515 (2019) doi:10.1148/radiol.2019192515. - DOI - PubMed

[7] Gundersen OE, Gil Y & Aha DW On reproducible AI: Towards reproducible research, open science, and digital scholarship in AI publications. AI Magazine 39, 56–68 (2018).

[8] Gundersen OE, Gil Y & Aha DW On reproducible AI: Towards reproducible research, open science, and digital scholarship in AI publications. AI Magazine 39, 56–68 (2018).

[9] Crane M. Questionable Answers in Question Answering Research: Reproducibility and Variability of Published Results. Transactions of the Association for Computational Linguistics 6, 241–252 (2018).

[10] Crane M. Questionable Answers in Question Answering Research: Reproducibility and Variability of Published Results. Transactions of the Association for Computational Linguistics 6, 241–252 (2018).

Save citation to file

Email citation

Add to Collections

Add to My Bibliography

Your saved search

Create a file for external citation management software

Your RSS Feed

Transparency and reproducibility in artificial intelligence

Collaborators

Affiliations

Transparency and reproducibility in artificial intelligence

Authors

Collaborators

Affiliations

Abstract

Comment in

Comment on

Similar articles

Cited by

References

Publication types

MeSH terms

Grants and funding

LinkOut - more resources

Full Text Sources

Abstract

Comment in

Comment on

Similar articles

Cited by

References

Publication types

MeSH terms

Related information

Grants and funding

LinkOut - more resources

Full Text Sources