Abstract
This article concerns the rise of a new paradigm in AI - “foundation models,” which are pre-trained on broad data at scale and subsequently adapted to particular downstream tasks. In particular, it explores the issue from the perspective of healthcare and biomedicine, focusing on the benefits of foundation models, as well as their propensity to encode bias, which threatens to exacerbate discriminatory practices already experienced by patients in Europe. Section 1 offers a brief introduction concerning the use of AI in healthcare and biomedicine and the problem of divergencies in access to and quality of healthcare across Europe. Section 2 familiarises the reader with the technical qualities of foundation models and recent developments in the field. Section 3 explains how the new health data strategy proposed by the EU could foster the development of foundation models in healthcare. Section 4 elaborates on their benefits in healthcare and biomedicine, while Sect. 5 explores the risk of bias exhibited by foundation models. Section 6 comments on the uncertain status of foundation models under the proposed Artificial Intelligence Act and offers brief recommendations concerning future regulation. Section 7 concludes.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Abdi, J., et al.: Scoping review on the use of socially assistive robot technology in elderly care. BMJ Open 8(2), e018815 (2018). https://doi.org/10.1136/bmjopen-2017-018815
Abid, A., Farooqi, M., Zou, J.: Persistent anti-Muslim bias in large language models. arXiv:2101.05783. arXiv (2021).https://doi.org/10.48550/arXiv.2101.05783
Alpaydin, E.: Machine Learning, Revised and Updated Edition. The MIT Press (The MIT Press Essential Knowledge Series), Cambridge, Massachusetts (2021)
Alsentzer, E., et al.: Publicly available clinical BERT embeddings. arXiv:1904.03323. arXiv (2019). https://doi.org/10.48550/arXiv.1904.03323
Auffray, C., et al.: Making sense of big data in health research: towards an EU action plan. Genome Med. 8(1), 71 (2016). https://doi.org/10.1186/s13073-016-0323-y
Barera, M.: Mind the gap: addressing structural equity and inclusion on Wikipedia. http://hdl.handle.net/10106/29572 (2020). Accessed 20 Mar 2022
Bender, E.M., et al.: On the dangers of stochastic parrots: can language models be too big?. In: Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, pp. 610–623. Association for Computing Machinery (FAccT ’21), New York, NY, USA (2021). doi:https://doi.org/10.1145/3442188.3445922
Birhane, A., Prabhu, V.U., Kahembwe, E.: Multimodal datasets: misogyny, pornography, and malignant stereotypes. arXiv:2110.01963. arXiv (2021). https://doi.org/10.48550/arXiv.2110.01963
Bolukbasi, T., et al.: ‘Man is to computer programmer as woman is to homemaker? debiasing word embeddings In: Lee, D., et al. (eds.) Advances in Neural Information Processing Systems’, Vol. 29. Curran Associates, Inc. https://proceedings.neurips.cc/paper/2016/file/a486cd07e4ac3d270571622f4f316ec5-Paper.pdf (2016). Accessed 20 Mar 2022
Bommasani, R., et al.: On the opportunities and risks of foundation models. arXiv:2108.07258. arXiv (2021). https://doi.org/10.48550/arXiv.2108.07258
Broadband Commission for Sustainable Development: Working group on digital and AI in health: reimagining global health through artificial intelligence: the roadmap to AI maturity (Sep 2020)
Brown, T.B., et al.: Language models are few-shot learners. arXiv:2005.14165. arXiv (2020). https://doi.org/10.48550/arXiv.2005.14165
Caruana, R., et al.: Intelligible models for healthcare: predicting pneumonia risk and hospital 30-day readmission. In: Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 1721–1730. Association for Computing Machinery (KDD ’15), New York, NY, USA (2015). https://doi.org/10.1145/2783258.2788613
Chaix, B., et al.: When chatbots meet patients: one-year prospective study of conversations between patients with breast cancer and a chatbot. JMIR Cancer 5(1), e12856 (2019). https://doi.org/10.2196/12856
Chowdhery, A., et al.: PaLM: scaling language modeling with pathways. arXiv. http://arxiv.org/abs/2204.02311 (2022)
Communication from the Commission to the European Parliament and the Council: A European health data space: harnessing the power of health data for people, patients and innovation COM, 196 final (2022)
Communication from the Commission to the European Parliament: the council, the european economic and social committee and the committee of the regions. A European strategy for data COM, 66 final (2020)
Council of Europe: Recommendation Rec: 10 of the committee of ministers to member states on better access to health care for Roma and Travellers in Europe (Adopted by the Committee of Ministers on 12 July 2006 at the 971st meeting of the Ministers’ Deputies) (2006)
Council of the European Union: Proposal for a regulation of the european parliament and of the council laying down harmonised rules on artificial intelligence (Artificial Intelligence Act) and amending certain Union legislative acts - Slovenian Presidency compromise text. Brussels, 2021/0106(COD) (29 Nov 2021)
Cui, C., et al.: Deep multi-modal fusion of image and non-image data in disease diagnosis and prognosis: a review. arXiv:2203.15588. arXiv (2022). https://doi.org/10.48550/arXiv.2203.15588
Demner-Fushman, D., Mrabet, Y., Ben Abacha, A.: Consumer health information and question answering: helping consumers find answers to their health-related information needs. J. Am. Med. Inform. Assoc.: JAMIA 27(2), 194–201 (2020). https://doi.org/10.1093/jamia/ocz152
Dias Oliva, T., Antonialli, D.M., Gomes, A.: Fighting hate speech, silencing drag queens? artificial intelligence in content moderation and risks to LGBTQ voices online. Sex. Cult. 25(2), 700–732 (2020). https://doi.org/10.1007/s12119-020-09790-w
Federico, G.: Access to healthcare in the European Union: are EU patients (Effectively) protected against discriminatory practices? In: Rossi, L.S., Casolari, F. (eds.) The Principle of Equality in EU Law, pp. 229–253. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-66137-7_8
Di Federico, G.: ‘Stuck in the middle with you…wondering what it is I should do. Some Considerations on EU’s Response to COVID-19’ 7 EUROJUS, pp. 60–85 (2020)
Directive (EU): 2019/1024 of the European Parliament and of the Council of 20 June 2019 on Open Data and the Re-use of Public Sector Information OJ L 172, pp. 56–83 (26 Jun 2019)
European Union Agency for Fundamental Rights: Inequalities and Multiple Discrimination in Access to and Quality of Healthcare. Publications Office, LU. (2013). https://doi.org/10.2811/17523. Accessed 12 May 2022
European Union Agency for Fundamental Rights: Coronavirus Pandemic in the EU: Impact on Roma and Travellers, Bulletin #5, 1 March - 30 June 2020. Publications Office, LU (2020). https://doi.org/10.2811/978921. (Accessed 13 May 2022
European Commission: Horizon 2020 Work Programme 2018–2020: Health, Demographic Change and Wellbeing. European Commission Decision C(2020)4029 of (17 Jun 2020)
European Commission: Study to Support the Preparation of the European’s Commission Initiative to Extend the List of EU Crimes in Article 83 of the Treaty on the Functioning of the EU to Hate Speech and Hate crime. Publications Office of the European Union, Luxembourg (2021)
European Parliament: Draft Opinion of the Committee on Legal Affairs for the Committee on the Internal Market and Consumer Protection and the Committee on Civil Liberties, Justice and Home Affairs on the proposal for a regulation of the European Parliament and of the Council laying down Harmonised Rules on Artificial Intelligence (Artificial Intelligence Act) and Amending Certain Union Legislative Acts. Brussels, 2021/0106(COD) (2 Mar 2022)
Equinet: Equality, Diversity and Non-Discrimination in Healthcare: Learning from the Work of Equality Bodies, Brussels (2021)
Funahashi, K.: Big data in health care - predicting your future health’ 94 southern California law review, pp. 355–390 (2021)
Guo, W., Caliskan, A.: Detecting emergent intersectional biases: contextualized word embeddings contain a distribution of human-like biases. In: Proceedings of the 2021 AAAI/ACM Conference on AI, Ethics, and Society, pp. 122–133 (2021). https://doi.org/10.1145/3461702.3462536
Harrer, S., et al.: Artificial intelligence for clinical trial design. Trends Pharmacol. Sci. 40(8), 577–591 (2019). https://doi.org/10.1016/j.tips.2019.05.005
High-Level Conference: Governing the Game Changer – Impacts of artificial intelligence development on human rights, democracy and the rule of law. (Helsinki 2019) CommDH/Speech (2019)
International Coalition of Medicines Regulatory Authorities (ICMRA): Informal Innovation Network Horizon Scanning Assessment Report – Artificial Intelligence (Aug 2021)
Korngiebel, D.M., Mooney, S.D.: ‘Considering the possibilities and pitfalls of generative pre-trained transformer 3 (GPT-3) in healthcare delivery. NPJ Digit. Med. 4(1), 1–3 (2021). https://doi.org/10.1038/s41746-021-00464-x
Kulpa, E., Rahman, A.T., Vahia, I.V.: Approaches to assessing the impact of robotics in geriatric mental health care: a scoping review. Int. Rev. Psychiatry (Abingdon, England) 33(4), 424–434 (2021). https://doi.org/10.1080/09540261.2020.1839391
Lu, K., et al.: Pretrained transformers as universal computation engines. arXiv:2103.05247. arXiv (2021). https://doi.org/10.48550/arXiv.2103.05247
Muller, C., et al.: AIA in-depth #1: objective, scope, definition.https://allai.nl/wp-content/uploads/2022/03/AIA-in-depth-Objective-Scope-and-Definition.pdf (ALLAI 2022). Accessed 20 May 2022
Percha, B.: Modern clinical text mining: a guide and review. Ann. Rev. Biomed. Data Sci. 4, 165–187 (2021). https://doi.org/10.1146/annurev-biodatasci-030421-030931
Proposal for a Regulation of the European Parliament and of the Council on European data governance and amending Regulation (EU) 2018/1724 (Data Governance Act) COM, 767 final (2020)
Proposal for the Regulation of the European Parliament and of the Council laying down harmonized rules on Artificial Intelligence and amending certain Union legislative acts COM, 206 final (2021)
Proposal for a Regulation of the European Parliament and of the Council on harmonised rules on fair access to and use of data (Data Act) COM, 68 final (2022)
Proposal for a Regulation of the European Parliament and the Council on the European Health Data Space COM, 197 final (2022)
Rae, J.W., et al.: Scaling language models: methods, analysis & insights from training gopher. arXiv:2112.11446. arXiv (2022). https://doi.org/10.48550/arXiv.2112.11446
Ramesh, A. et al. (2022) Hierarchical Text-Conditional Image Generation with CLIP Latents. arXiv:2204.06125. arXiv. doi:https://doi.org/10.48550/arXiv.2204.06125
Rasmy, L., et al.: Med-BERT: pretrained contextualized embeddings on large-scale structured electronic health records for disease prediction. NPJ Digit. Med. 4(1), 86 (2021). https://doi.org/10.1038/s41746-021-00455-y
Regulation 2017/745 of the European Parliament and the Council of 5 April 2017 on medical devices amending Directive 2001/83/EC, Regulation (EC) No 178/2002 and Regulation (EC) No 1223/2009 and repealing Council Directives 90/385/EEC and 93/42/EEC OJ L 117, pp. 1–175 (5 May 2017)
Rousseau, A., Baudelaire, C., Riera, K.: Doctor GPT-3: hype or reality?. Nabla. https://www.nabla.com/blog/gpt-3/ (27 Oct 2020). Accessed 20 Mar 2022
Scholz, N., European parliament, and directorate-general for parliamentary research services: Addressing health inequalities in the European Union: concepts, action, state of play: in-depth analysis. https://op.europa.eu/publication/manifestation_identifier/PUB_QA0120125ENN (2020). Accessed 12 May 2022
Tsaban, T., et al.: Harnessing protein folding neural networks for peptide–protein docking. Nat. Commun. 13(1), 176 (2022). https://doi.org/10.1038/s41467-021-27838-9
World Health Organization: Ethics and governance of artificial intelligence for health: WHO guidance. World Health Organization, Geneva. https://apps.who.int/iris/handle/10665/341996 (2021). Accessed 12 May 2022
Zhang, H., et al.: Hurtful words: quantifying biases in clinical contextual word embeddings. arXiv:2003.11515. arXiv (2020).https://doi.org/10.48550/arXiv.2003.11515
Zhang D., et al.: The AI Index 2022 Annual Report’ AI Index Steering Committee. Stanford Institute for Human-Centered AI (2022)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Wójcik, M.A. (2022). Foundation Models in Healthcare: Opportunities, Biases and Regulatory Prospects in Europe. In: Kő, A., Francesconi, E., Kotsis, G., Tjoa, A.M., Khalil, I. (eds) Electronic Government and the Information Systems Perspective. EGOVIS 2022. Lecture Notes in Computer Science, vol 13429. Springer, Cham. https://doi.org/10.1007/978-3-031-12673-4_3
Download citation
DOI: https://doi.org/10.1007/978-3-031-12673-4_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-12672-7
Online ISBN: 978-3-031-12673-4
eBook Packages: Computer ScienceComputer Science (R0)