CoChat

Explainability in the age of large language models for healthcare

Mesinovic, Munib, Watkinson, Peter, Zhu, Tingting · 2025 · Communications Engineering
Cited by 54

The DOI matches its Crossref record; 20 of the 20 references it cites were found in a registry or an index.

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 · 10.1038/s44172-025-00453-y

Cites

20 of 20 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Balancing explainability requirements with model capabilities for LLMs in healthcare. The diagram shows a blue balance scale representing the trade-off between model development factors (left side) and explainability requirements (right side). The left side displays predictive power curves (blue and black lines), complexity (blue network), and multi-modality data types (purple EHR, teal medical imaging, purple genomics). The right side shows explainability components in a black dashed box: interpretability (blue bar chart), fairness (balance scales), and trust (handshake icon). Teal boxes on the right list potential solutions: Advanced Reasoning Capacity, Increased Focus on XAI for LLMs, and Cross-disciplinary Collaboration. The foundation shows LLMs in Healthcare (blue box) supported by Regulation and Responsible Innovation (teal boxes).
Fig. 1 Balancing explainability requirements with model capabilities for LLMs in healthcare. The diagram shows a blue balance scale representing the trade-off between model development factors (left side) and explainability requirements (right side). The left side displays predictive power curves (blue and black lines), complexity (blue network), and multi-modality data types (purple EHR, teal medical imaging, purple genomics). The right side shows explainability components in a black dashed box: interpretability (blue bar chart), fairness (balance scales), and trust (handshake icon). Teal boxes on the right list potential solutions: Advanced Reasoning Capacity, Increased Focus on XAI for LLMs, and Cross-disciplinary Collaboration. The foundation shows LLMs in Healthcare (blue box) supported by Regulation and Responsible Innovation (teal boxes).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 20 references

Research works printed but not linked: 17

  1. Thirunavukarasu, A. J. et al. Large language models in medicine. Nat. Med. 29, 1930–1940 (2023). · DOI · Paper page
  2. Meskó, B. & Topol, E. J. The imperative for regulatory oversight of large language models (or generative AI) in healthcare. NPJ Digit. Med. 6, 120 (2023). · DOI · Paper page
  3. Mesinovic, M., Watkinson, P. & Zhu, T. “Explainable AI for clinical risk prediction: a survey of concepts, methods, and modalities.” arXiv https://doi.org/10.48550/arXiv.2308.08407 (2023). · DOI
  4. Ahdritz, G. et al. Distinguishing the knowable from the unknowable with language models. arXiv https://doi.org/10.48550/arXiv.2402.03563 (2024). · DOI · Paper page
  5. Sun, Z. et al. Redeep: Detecting hallucination in retrieval-augmented generation via mechanistic interpretability. International Conference on Learning Representations (ICLR) https://doi.org/10.48550/arXiv.2410.11414 (2025). · DOI
  6. Ibrahim, L. et al. Explainable prediction of acute myocardial infarction using machine learning and Shapley values. IEEE Access 8, 210410–210417 (2020). · DOI · Paper page
  7. Luo, Y., Peng, J. & Ma, J. When causal inference meets deep learning. Nat. Mach. Intell. 2, 426–427 (2020). · DOI
  8. Kaminski, M. E. Regulating the risks of AI. BUL Rev. 103, 1347 (2023).
  9. Zhang, K. et al. FDA review of radiologic AI algorithms: process and challenges. Radiology 310, e230242 (2024). · DOI · Paper page
  10. Shah, S. A. et al. Use of artificial intelligence–based detection of diabetic retinopathy in the US. JAMA Ophthalmol. 142, 1171–1173 (2024). · DOI
  11. Mahler, M. et al. “Regulatory aspects of artificial intelligence and machine learning-enabled software as medical devices (SaMD).” Precision Med. Artif. Intell. 2021, 237−265 (2021). · DOI
  12. Imrie, F., Davis, R. & Van Der Schaar, M. Multiple stakeholders drive diverse interpretability requirements for machine learning in healthcare. Nat. Mach. Intell. 5, 824–829 (2023). · DOI
  13. Panigutti, C.et al. “The role of explainable AI in the context of the AI act.” In Proceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency, 1139−1150 (ACM, 2023). · DOI
  14. Mahase, E. NICE recommends AI technology for planning radiotherapy. BMJ. Br. Med. J. 382, p1861 (2023). · doi:10.1136/bmj.p1861
  15. Qiu, J. et al. LLM-based agentic systems in medicine and healthcare. Nat. Mach. Intell. 6, 1418–1420 (2024). · DOI · Paper page
  16. Kumar, B. et al. “Selective fine-tuning on llm-labeled data may reduce reliance on human annotation: a case study using schedule-of-event table detection.” arXiv https://doi.org/10.48550/arXiv.2405.06093 (2024). · DOI
  17. Azevedo, D., Legay, A. & Kieffer, S. “User Reception of Babylon Health’s Chatbot.” https://www.scitepress.org/Papers/2022/108030/108030.pdf (2022).

Cited by

See all 54

Metadata

Authors
Mesinovic, M., Watkinson, P., & Zhu, T.
Venue
Communications Engineering
Published
2025-07-17
Volume
4
Issue
1
ISSN
2731-3395
Cited by
54
References
20
Language
en
Licence
Funders
  • Rhodes Trust; EPSRC CDT Health Data Science
  • DH | National Institute for Health Research
  • Royal Academy of Engineering

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

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Explainability in the age of large language models for healthcare. https://app.cochat.ai/papers/10.1038/s44172-025-00453-y
The paper
Mesinovic, M., Watkinson, P., & Zhu, T. (2025). Explainability in the age of large language models for healthcare. Communications Engineering. https://doi.org/10.1038/s44172-025-00453-y

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.