Artificial intelligence–based image analysis in clinical testing: lessons from cervical cancer screening
JNCI: Journal of the National Cancer Institute · 2024
DOI: 10.1093/jnci/djad202 ↗Kurz gefasst
Fachartikel in der Zeitschrift JNCI: Journal of the National Cancer Institute, 2024. Verfasst von Didem Egemen und 13 weiteren Autor:innen. Beteiligte Einrichtung: Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Rockville, MD, USA. Hauptthemen laut MEDLINE: Artificial Intelligence, Uterine Cervical Neoplasms. Die Arbeit wurde bisher 51-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: NIH HHS, National Institutes of Health.
Abstract (Originalsprache)
Novel screening and diagnostic tests based on artificial intelligence (AI) image recognition algorithms are proliferating. Some initial reports claim outstanding accuracy followed by disappointing lack of confirmation, including our own early work on cervical screening. This is a presentation of lessons learned, organized as a conceptual step-by-step approach to bridge the gap between the creation of an AI algorithm and clinical efficacy. The first fundamental principle is specifying rigorously what the algorithm is designed to identify and what the test is intended to measure (eg, screening, diagnostic, or prognostic). Second, designing the AI algorithm to minimize the most clinically important errors. For example, many equivocal cervical images cannot yet be labeled because the borderline between cases and controls is blurred. To avoid a misclassified case-control dichotomy, we have isolated the equivocal cases and formally included an intermediate, indeterminate class (severity order of classes: case>indeterminate>control). The third principle is evaluating AI algorithms like any other test, using clinical epidemiologic criteria. Repeatability of the algorithm at the borderline, for indeterminate images, has proven extremely informative. Distinguishing between internal and external validation is also essential. Linking the AI algorithm results to clinical risk estimation is the fourth principle. Absolute risk (not relative) is the critical metric for translating a test result into clinical use. Finally, generating risk-based guidelines for clinical use that match local resources and priorities is the last principle in our approach. We are particularly interested in applications to lower-resource settings to address health disparities. We note that similar principles apply to other domains of AI-based image analysis for medical diagnostic testing.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Rockville, MD, USA.
- Department of Obstetrics and Gynecology, Boston Medical Center/Boston University School of Medicine, Boston, MA, USA.
- Information Management Services Inc, Calverton, MD, USA.
- Department of Epidemiology, School of Public Health, University of Washington, Seattle, WA, USA.
- Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Boston, MA, USA.
- Harvard Graduate Program in Biophysics, Harvard Medical School, Harvard University, Cambridge, MA, USA.
ORCID-Profile: Didem Egemen, Brian Befano, Sameer Antani, Nicolas Wentzensen, Mark Schiffman
Förderung
- NIH HHS
- National Institutes of Health
Zitieren
APA
Egemen, D., Perkins, R. B., Cheung, L. C., Befano, B., Rodriguez, A. C., Desai, K., Lemay, A., Ahmed, S. R., Antani, S., Jeronimo, J., Wentzensen, N., Kalpathy-Cramer, J., De Sanjose, S., & Schiffman, M. (2024). Artificial intelligence–based image analysis in clinical testing: lessons from cervical cancer screening. JNCI: Journal of the National Cancer Institute. https://doi.org/10.1093/jnci/djad202
Vancouver
Egemen D, Perkins RB, Cheung LC, Befano B, Rodriguez AC, Desai K, et al. Artificial intelligence–based image analysis in clinical testing: lessons from cervical cancer screening. JNCI: Journal of the National Cancer Institute. 2024. doi:10.1093/jnci/djad202
BibTeX
@article{Egemen2024,
author = {Egemen, Didem and Perkins, Rebecca B and Cheung, Li C and Befano, Brian and Rodriguez, Ana Cecilia and Desai, Kanan and Lemay, Andreanne and Ahmed, Syed Rakin and Antani, Sameer and Jeronimo, Jose and Wentzensen, Nicolas and Kalpathy-Cramer, Jayashree and De Sanjose, Silvia and Schiffman, Mark},
title = {Artificial intelligence–based image analysis in clinical testing: lessons from cervical cancer screening},
journal = {JNCI: Journal of the National Cancer Institute},
year = {2024},
doi = {10.1093/jnci/djad202}
}
EUVIMED