EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

The Framework That Survives Bad Models: Human-AI Collaboration for Clinical Trials

Yao Chen, David Ohlssen, Aimee Readie, Gregory Ligozio, Ruvie Martin, Alain Marti, Thibaud Coroller

The New England Journal of Statistics in Data Science · 01.01.2026 · 1-8

DOI: 10.51387/26-nejsds111 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The New England Journal of Statistics in Data Science, online erschienen am 10.11.2026. Verfasst von Yao Chen und 6 weiteren Autor:innen. Die Arbeit nennt 23 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Artificial intelligence (AI) holds great promise for supporting clinical trials, from patient recruitment and endpoint assessment to treatment response prediction. However, deploying AI without safeguards poses significant risks, particularly when evaluating patient endpoints that directly impact trial conclusions. We compared two AI frameworks against human-only assessment for medical image-based disease evaluation, measuring cost, accuracy, robustness, and generalization ability. To stress-test these frameworks, we injected bad models—ranging from random guesses to naive predictions—to ensure that observed treatment effects remain valid even under severe model degradation. We evaluated the frameworks using two randomized controlled trials with endpoints derived from spinal X-ray images. Our findings indicate that using AI as a supporting reader (AI-SR) is the most suitable approach for clinical trials, as it meets all criteria across various model types, even with bad models. This method consistently provides reliable disease estimation, preserves clinical trial treatment effect estimates and conclusions, and retains these advantages when applied to different populations.

Zitieren

APA

Chen, Y., Ohlssen, D., Readie, A., Ligozio, G., Martin, R., Marti, A., & Coroller, T. (2026). The Framework That Survives Bad Models: Human-AI Collaboration for Clinical Trials. The New England Journal of Statistics in Data Science, 1-8. https://doi.org/10.51387/26-nejsds111

Vancouver

Chen Y, Ohlssen D, Readie A, Ligozio G, Martin R, Marti A, et al. The Framework That Survives Bad Models: Human-AI Collaboration for Clinical Trials. The New England Journal of Statistics in Data Science. 2026:1-8. doi:10.51387/26-nejsds111

BibTeX
@article{Chen2026,
  author = {Chen, Yao and Ohlssen, David and Readie, Aimee and Ligozio, Gregory and Martin, Ruvie and Marti, Alain and Coroller, Thibaud},
  title = {The Framework That Survives Bad Models: Human-AI Collaboration for Clinical Trials},
  journal = {The New England Journal of Statistics in Data Science},
  year = {2026},
  pages = {1--8},
  doi = {10.51387/26-nejsds111}
}
Kostenlos starten →