Abstract
Many real-world applications of machine learning involve handling data that is collected over an extended period of time. The longer this time-period, the more likely that the underlying characteristics of this data are to change, potentially leading to a degradation in prediction accuracy and impacting decision-making. This phenomenon, commonly referred to as data drift, poses a risk in the context of medical AI regulation and monitoring. Regulatory bodies must regularly assess previously approved models for their performance on new data, realistically even in scenarios where prediction labels are not yet available, making the tracking of model performance unfeasible. In this paper, our contribution involves introducing a comprehensive framework to estimate the performance drift of a model when evaluated on new unlabelled target data. We introduce a method that assesses both i) the uncertainty in model predictions and ii) the discrimination error between training batches and subsequent test batches, serving as key indicators for identifying drift in AI model performance. We test our framework on simulated drift data where we can control the nature of change, and high-fidelity synthetic primary care data focused on the UK Covid-19 pandemic. Promising results emerge from our experiments, suggesting that the proposed metrics can effectively monitor potential changes in the performance of AI health products post-deployment even in the absence of labelled data.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Gama, J., et al.: A survey on concept drift adaptation. ACM Comput. Surv. (CSUR) 46(4), 1–37 (2014)
Rotalinti, Y., et al.: Detecting drift in healthcare AI models based on data availability. In: Koprinska, I., et al. (eds.) Joint European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML PKDD 2022. CCIS, vol. 1753, pp. 243–258. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-23633-4_17
Hoens, T.R., Polikar, R., Chawla, N.V.: Learning from streaming data with concept drift and imbalance: an overview. Prog. Artif. Intell. 1, 89–101 (2012)
Ditzler, G., et al.: Learning in nonstationary environments: a survey. IEEE Comput. Intell. Mag. 10(4), 12–25 (2015)
Ben-David, S., et al.: Analysis of representations for domain adaptation. In: Advances in Neural Information Processing Systems, vol. 19 (2006)
Gama, J., Medas, P., Castillo, G., Rodrigues, P.: Learning with drift detection. In: Bazzan, A.L.C., Labidi, S. (eds.) SBIA 2004. LNCS (LNAI), vol. 3171, pp. 286–295. Springer, Heidelberg (2004). https://doi.org/10.1007/978-3-540-28645-5_29
Baena-Garcıa, M., et al.: Early drift detection method. In: Fourth International Workshop on Knowledge Discovery from Data Streams, vol. 6 (2006)
Berger, J.O.: Statistical Decision Theory and Bayesian Analysis. Springer, New York (2013). https://doi.org/10.1007/978-1-4757-4286-2
Kelly, M.G., Hand, D.J., Adams, N.M.: The impact of changing populations on classifier performance. In: Proceedings of the Fifth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (1999)
Tsymbal, A.: The problem of concept drift: definitions and related work. Comput. Sci. Dept. Trinity Coll. Dublin 106(2), 58 (2004)
Wares, S., Isaacs, J., Elyan, E.: Data stream mining: methods and challenges for handling concept drift. SN Appl. Sci. 1, 1–19 (2019)
Bifet, A., Gavalda, R.: Learning from time-changing data with adaptive windowing. In: Proceedings of the 2007 SIAM International Conference on Data Mining. Society for Industrial and Applied Mathematics (2007)
Žliobaite, I.: Change with delayed labeling: when is it detectable?. In: 2010 IEEE International Conference on Data Mining Workshops. IEEE (2010)
Ackerman, S., Raz, O., Zalmanovici, M.: FreaAI: automated extraction of data slices to test machine learning models. In: Shehory, O., Farchi, E., Barash, G. (eds.) EDSMLS 2020. CCIS, vol. 1272, pp. 67–83. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-62144-5_6
Clinical Practice Research Datalink, 1 June 2022. https://www.cprd.com
Covid-19 in the UK, 18 November 2023. https://coronavirus.data.gov.uk/
Wolf, A., et al.: Data resource profile: clinical practice research Datalink (CPRD) aurum.". Int. J. Epidemiol. 48(6), 1740–1740g (2019)
Agrawal, R., Imielinski, T., Swami, A.: Database mining: a performance perspective. IEEE Trans. Knowl. Data Eng. 5(6), 914–925 (1993)
Marrs, G.R., Hickey, R.J., Black, M.M.: The impact of latency on online classification learning with concept drift. In: Bi, Y., Williams, M.A. (eds.) Knowledge Science, Engineering and Management: 4th International Conference, KSEM 2010, Belfast, Northern Ireland, UK, 1–3 September 2010, Proceedings, vol. 6291, pp. 459–469. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-15280-1_42
Acknowledgements
This project has been made possible by a grant from the 3.7 million Regulators’ Pioneer Fund launched by the Department for Business, Energy and Industrial Strategy (BEIS).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Rotalinti, Y., Myles, P., Tucker, A. (2024). Predicting Performance Drift in AI Models of Healthcare Without Ground Truth Labels. In: Miliou, I., Piatkowski, N., Papapetrou, P. (eds) Advances in Intelligent Data Analysis XXII. IDA 2024. Lecture Notes in Computer Science, vol 14641. Springer, Cham. https://doi.org/10.1007/978-3-031-58547-0_14
Download citation
DOI: https://doi.org/10.1007/978-3-031-58547-0_14
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-58546-3
Online ISBN: 978-3-031-58547-0
eBook Packages: Computer ScienceComputer Science (R0)