Abstract
The risk of work disability can be predicted with machine learning (ML). All stakeholders should understand how the estimation is done to trust the results. Thus, these methods should be not only sufficiently accurate but also transparent and explainable. Explainability is one topic in artificial intelligence (AI) Ethics. Explainable AI (XAI) is especially important in health-related topics. We compared the accuracy of two ML methods that use occupational health care data (MHealth) and pension decision register data (MPension). Method MHealth uses deep neural networks and natural language processing (NLP) algorithms and method MPension uses different decision tree algorithms. We can assume that both methods are black box predictors because the reasoning behind the function is not understandable by humans. We observed in our previous study that both methods are sufficiently accurate to support experts in decision making. Our aim in this study was to determine if these methods are sufficiently explainable for clinical use. The two main approaches to estimating the explainability of the ML methods are transparency design and post-hoc explanations. We could not access the data for these methods and had to limit our research to the post-hoc explanation approach. We formulated the visualizations for the methods MHealth and MPension and discussed how understandable they are. We also determined that the explainability is better in MPension but the deep learning algorithm in MHealth is also possible to visualize.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
OECD: Sickness, Disability and Work: Breaking the Barriers. OECD Publishing, Paris (2010)
Saarela, K., Huhta-Koivisto, V., Nurminen, J.K.: Ethical aspects of work disability risk prediction using machine learning. In: Nagar, A.K., Singh Jat, D., Kumar Mishra, D., Joshi, A. (eds.) Intelligent Sustainable Systems: Selected Papers of WorldS4 2022, vo. 1, pp. 499–509. Springer Nature, Singapore (2023). https://doi.org/10.1007/978-981-19-7660-5_43
Airaksinen, J., et al.: Development and validation of a risk prediction model for work disability: multicohort study. Sci. Rep. 7(1), 13578 (2017)
Salonen, L., Blomgren, J., Laaksonen, M.: From long-term sickness absence to disability retirement: diagnostic and occupational class differences within the working-age Finnish population. BMC Publ. Health 20(1), 1–11 (2020)
Salonen, L., Alexanderson, K., Rugulies, R., et al.: Combinations of job demands and job control and future trajectories of sickness absence and disability pension an 11-year follow-up of two million employees in Sweden. J. Occup. Env. Med. 62(10), 795–802 (2020)
Bethge, M., Spanier, K., Köhn, S., Schlumbohm, A.: Self-reported work ability predicts health-related exit and absence from work, work participation, and death: longitudinal findings from a sample of German employees. Int. Arch. Occup. Environ. Health 94(4), 591–599 (2020)
Shiri, R., et al.: A screening tool for the risk of disability retirement due to musculoskeletal disorders. Scand. J. Work Environ. Health 44(1), 37–46 (2018)
van Amelsvoort, L.G.P.M., Kant, I.J., Beurskens, A.J.H.M., Schröer, C.A.P., Swaen, G.M.H.: Fatigue as a predictor of work disability. Occup, Env. Med. 59(10), 712–713 (2002)
Lindström, I., Pallasaho, P., Remes, J., et al.: Does lung function predict the risk of disability pension? An 11-year register-based follow-up study. BMC Public Health 20, 165 (2020)
Pham, A.D., et al.: Natural language processing of radiology reports for the detection of thromboembolic diseases and clinically relevant incidental findings. BMC Bioinform. 15(1), 266 (2014)
Huhta-Koivisto, T.: Work disability risk prediction with machine learning. Master Thesis. Aalto University, School of Electrical Engineering, 56 p. Espoo (2020)
Saarela, K., Huhta-Koivisto, V., Nurminen, J.K.: Work disability risk prediction using machine learning, comparison of two methods. In: Daimi, K., Al Sadoon, A. (eds.) Proceedings of the ICR’22 International Conference on Innovations in Computing Research, pp. 13–21. Springer International Publishing, Cham (2022). https://doi.org/10.1007/978-3-031-14054-9_2
Jobin, A., Ienca, M., Vayena, E.: The global landscape of AI ethics guidelines. Nat. Mach. Intell. 1, 389–399 (2019)
Nickel, J.: Human Rights. In: Zalta, E.N. (ed.) The Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/archives/fall2021/entries/rights-human/ (2021)
Parliament, E.: Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46. OJEU 59, 294 (2016)
Miller K.: AI decisions: do we deserve an explanation? 29 Jun 2020. https://www.futurity.org/ai-decisions-right-to-explanation-2394872-2/. Accessed 11 Nov 2021
Ghassemi, M., Oakden-Rayner, L., Beam, A.L.: The false hope of current approaches to explainable artificial intelligence in health care. The Lancet Dig. Health 3(11), e745–e750 (2021)
Doshi-Velez, F., Been, K.: Towards a rigorous science of interpretable machine learning. arXiv preprint arXiv:1702.08608 (2017)
Xu, F., Uszkoreit, H., Du, Y., Fan, W., Zhao, D., Zhu, J.: Explainable AI: a brief survey on history, research areas, approaches and challenges. In: Tang, J., Kan, M.-Y., Zhao, D., Li, S., Zan, H. (eds.) NLPCC 2019. LNCS (LNAI), vol. 11839, pp. 563–574. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-32236-6_51
Angelov, P.P., Soares, E.A., Jiang, R., Arnold, N.I., Atkinson, P.M.: Explainable artificial intelligence: an analytical review. Wiley Interdisc. Rev.: Data Min. Knowl. Discovery 11(5), e1424 (2021). https://doi.org/10.1002/widm.1424
Wells, L., Bednarz, T.: Explainable ai and reinforcement learning—a systematic review of current approaches and trends. Front. Artif. Intell. 4 (2021)
Ananny, M., Crawford, K.: Seeing without knowing: Limitations of the transparency ideal and its application to algorithmic accountability. New Media Soc. 20(3), 973–989 (2018)
Mittelstadt, B., Russell, C., Wachter, S.: Explaining Explanations in AI. arXiv:1811.01439v1 [cs.AI]. 4 Nov 2018
Rudin, C.: Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nat. Mach. Intell. 1(5), 206–215 (2019)
European Comission: Digital strategy – managing health data. https://digital-strategy.ec.europa.eu/en/policies/health-data. Accessed 26 Nov 2022
Varis, J.: “Eläketurvakeskuksen koneoppimiskokeilu – näin se tehtiin!”, Finnish Centre for Pensions, blog post, https://www.etk.fi/blogit/elaketurvakeskuksen-koneoppimiskokeilu-nain-se-tehtiin/. 17 Apr 2018
European Pensions: Finnish AI testing successfully identifies future retirees facing disability pension. https://www.europeanpensions.net/ep/Finnish-AI-successfully-identifies-future-retirees-facing-disability-pension.php. 17 Apr 2018
Guidotti, R., Monreale, A., Ruggieri, S., Turini, F., Giannotti, F., Pedreschi, D.: A survey of methods for explaining black box models. ACM Comput. Surv. 51(5), 1–42 (2018)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Saarela, K., Huhta-Koivisto, V., Kemell, KK., Nurminen, J.K. (2023). Explainability of Machine Learning in Work Disability Risk Prediction. In: Daimi, K., Al Sadoon, A. (eds) Proceedings of the 2023 International Conference on Advances in Computing Research (ACR’23). ACR 2023. Lecture Notes in Networks and Systems, vol 700. Springer, Cham. https://doi.org/10.1007/978-3-031-33743-7_20
Download citation
DOI: https://doi.org/10.1007/978-3-031-33743-7_20
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-33742-0
Online ISBN: 978-3-031-33743-7
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)