CoChat

Resilient Artificial Intelligence in Health: Synthesis and Research Agenda Toward Next-Generation Trustworthy Clinical Decision Support

Open access
Sáez, Carlos, Ferri, Pablo, García-Gómez, Juan M · 2024 · Journal of Medical Internet Research

Published version of a preprint

The DOI matches its Crossref record; 56 of the 56 references it cites were found in a registry or an index.

Abstract

Artificial intelligence (AI)–based clinical decision support systems are gaining momentum by relying on a greater volume and variety of secondary use data. However, the uncertainty, variability, and biases in real-world data environments still pose significant challenges to the development of health AI, its routine clinical use, and its regulatory frameworks. Health AI should be resilient against real-world environments throughout its lifecycle, including the training and prediction phases and maintenance during production, and health AI regulations should evolve accordingly. Data quality issues, variability over time or across sites, information uncertainty, human-computer interaction, and fundamental rights assurance are among the most relevant challenges. If health AI is not designed resiliently with regard to these real-world data effects, potentially biased data-driven medical decisions can risk the safety and fundamental rights of millions of people. In this viewpoint, we review the challenges, requirements, and methods for resilient AI in health and provide a research framework to improve the trustworthiness of next-generation AI-based clinical decision support.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.2196/50295

Cites

56 of 56 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Comparison of the conventional AI process and the proposed resilient AI paradigm. Conventional AI is generally aimed at generalization and stability in controlled environments. That is, typically learning and validating a static model from a standardized, quality-controlled dataset, and applying the inference on equivalently consistent inputs. Resilient AI, however, aims to handle real-world variable environments during all its lifecycle, from adapting and explaining against potential biases and variability in the secondary use of electronic health records, both at model training and during use, to handling uncertainty in decision-making, enabling next-generation, trustworthy clinical decision support systems. AI: artificial intelligence, MD: medical device, I/O: input/output.
Figure 1 Comparison of the conventional AI process and the proposed resilient AI paradigm. Conventional AI is generally aimed at generalization and stability in controlled environments. That is, typically learning and validating a static model from a standardized, quality-controlled dataset, and applying the inference on equivalently consistent inputs. Resilient AI, however, aims to handle real-world variable environments during all its lifecycle, from adapting and explaining against potential biases and variability in the secondary use of electronic health records, both at model training and during use, to handling uncertainty in decision-making, enabling next-generation, trustworthy clinical decision support systems. AI: artificial intelligence, MD: medical device, I/O: input/output.CC BY · PubMed Central
Figure 2 A classification for the expected behavior and functionality of resilient AI in health, regarding the AI training phase, the prediction in new cases, the routine maintenance of AI during production, and related regulatory aspects. We focus on supervised ML as the methodological driver for current well-established medical AI and clinical decision support systems. AI: artificial intelligence; FRA: fundamental rights assurance; GDPR: General Data Protection Regulation; ML: machine learning; MNAR: missing not-at-random.
Figure 2 A classification for the expected behavior and functionality of resilient AI in health, regarding the AI training phase, the prediction in new cases, the routine maintenance of AI during production, and related regulatory aspects. We focus on supervised ML as the methodological driver for current well-established medical AI and clinical decision support systems. AI: artificial intelligence; FRA: fundamental rights assurance; GDPR: General Data Protection Regulation; ML: machine learning; MNAR: missing not-at-random.CC BY · PubMed Central
Figure 3 Summary of available solutions and new proposed research for resilient AI in health and CDSS. AI: artificial intelligence; CDSS: clinical decision support systems; DQ: data quality; EHDS: European Health Data Space; EU: European Union; FDA: Food and Drug Administration; GDPR: General Data Protection Regulation; UX/UI: user experience and interface.
Figure 3 Summary of available solutions and new proposed research for resilient AI in health and CDSS. AI: artificial intelligence; CDSS: clinical decision support systems; DQ: data quality; EHDS: European Health Data Space; EU: European Union; FDA: Food and Drug Administration; GDPR: General Data Protection Regulation; UX/UI: user experience and interface.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 56 references

Research works printed but not linked: 8

  1. 10.1136/bmj.m1328 · doi:10.1136/bmj.m1328 · Paper page
  2. European Union Agency for Fundamental RightsData quality and artificial intelligence: mitigating bias and error to protect fundamental rightsLU: Publications Office20192022-06-01https://data.europa.eu/doi/10.2811/546219 · doi:10.2811/546219
  3. Sáez, C, 2012, Stud Health Technol Inform, 180, 721
  4. Quiñonero-Candela, J, 2009, Mass: MIT Press
  5. Goodfellow, I, 2016, Deep learning
  6. BengioYDucharmeRVincentPA neural probabilistic language modelAdvances in Neural Information Processing Systems20002024-04-16https://papers.nips.cc/paper_files/paper/2000/hash/728f206c2a01bf572b5940d7d9a8fa4c-Abstract.html
  7. European CommissionEthics guidelines for trustworthy AIDirectorate-General for Communications Networks, Content and Technology20192024-06-12https://data.europa.eu/doi/10.2759/346720 · doi:10.2759/346720
  8. 10.48550/arXiv.1602.04938 · arXiv · doi:10.48550/arxiv.1602.04938 · Paper page

Cited by

Metadata

Authors
Sáez, C., Ferri, P., & García-Gómez, J. M.
Affiliations
Universitat Politècnica de València (ES)
Corresponding author
Sáez, Carlos
Venue
Journal of Medical Internet Research
Published
2024-06-28
Volume
26
Pages
e50295
ISSN
1438-8871 (ISSN-L)
OpenAlex
W4400117494
Cited by
36
References
56
Citations by year
2026: 16 · 2025: 17 · 2024: 3
FWCI
0.82
Type
journal article
Language
en
Open access
Licence
cc-by
Locations
  • Journal of Medical Internet Research (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • RiuNet (Universitat Politècnica de València) (publishedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Topics
  • Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
  • Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
  • Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
Keywords
clinical decision support systems, health AI, trustworthy AI, real-world data, data quality, algorithmic bias, data variability, information uncertainty, human-computer interaction, fundamental rights, AI regulation, AI safety
MeSH terms
Artificial Intelligence, Humans, Decision Support Systems, Clinical

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.