Advertisement

Modelling the Impact of AI for Clinical Decision Support

  • Mariana R. NevesEmail author
  • D. William R. Marsh
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11526)

Abstract

Many AI (or ML) systems have been proposed for clinical decision support. Clinical usefulness is assessed in an ‘Impact Study’, a form of trial of a completed system. In development, in contrast, the focus is on AI accuracy measures, such as the AUC. To improve impact and to justify the cost of a study, the impact of a proposed AI system should be modelled during its development. We show that an Influence Diagram can be used for this and provide a small set of generic models for diagnostic AI systems. We show that how the AI interacts with clinical decision makers is at least as important as its predictive accuracy.

Keywords

Impact analysis Clinical decision support Influence diagram AI 

Notes

Acknowledgements

Support is acknowledged from EPSRC project EP/P009964/1 PAMBAYESIAN for MR and WM and for WM from the Institutional Links grant 352394702, funded by the UK Department of Business, Energy and Industrial Strategy.

References

  1. 1.
    Wyatt, J.C., Altman, D.G.: Commentary: prognostic models: clinically useful or quickly forgotten? BMJ 311(7019), 1539–1541 (1995)CrossRefGoogle Scholar
  2. 2.
    Altman, D.G., Vergouwe, Y., Royston, P., Moons, K.G.: Prognosis and prognostic research: validating a prognostic model. BMJ 338, b605 (2009)CrossRefGoogle Scholar
  3. 3.
    Wallace, E., et al.: Framework for the impact analysis and implementation of Clinical Prediction Rules (CPRs). BMC Med. Inform. Decis. Mak. 11(1), 62 (2011)CrossRefGoogle Scholar
  4. 4.
    Huang, C., et al.: Enhancing the prediction of acute kidney injury risk after percutaneous coronary intervention using machine learning techniques: a retrospective cohort study. PLOS Med. 15(11) (2018)Google Scholar
  5. 5.
    Hae, H., et al.: Machine learning assessment of myocardial ischemia using angiography: development and retrospective validation. PLOS Med. 15(11), e1002693 (2018)CrossRefGoogle Scholar
  6. 6.
    Yet, B., Neil, M., Fenton, N., Constantinou, A., Dementiev, E.: An improved method for solving hybrid influence diagrams. Int. J. Approx. Reason. 95, 93–112 (2018)MathSciNetCrossRefGoogle Scholar
  7. 7.
    Elmore, J.G., et al.: Evaluation of 12 strategies for obtaining second opinions to improve interpretation of breast histopathology: simulation study. BMJ 353, i3069 (2016)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.School of Electronic Engineering and Computer ScienceQueen Mary University of LondonLondonUK

Personalised recommendations