Checklist for Reproducibility of Deep Learning in Medical Imaging
Journal of Imaging Informatics in Medicine · 14.03.2024 · 37 (4) · 1664-1673
DOI: 10.1007/s10278-024-01065-2 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Imaging Informatics in Medicine, online erschienen am 14.03.2024. Verfasst von Mana Moassefi und 18 weiteren Autor:innen. Beteiligte Einrichtung: Mayo Clinic Artificial Intelligence Laboratory, Department of Radiology, Mayo Clinic, 200 1st St SW, Rochester, MN, 55905, USA. Hauptthemen laut MEDLINE: Deep Learning, Checklist, Delphi Technique, Diagnostic Imaging. Die Arbeit wurde bisher 22-mal zitiert (Crossref) und nennt 17 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The application of deep learning (DL) in medicine introduces transformative tools with the potential to enhance prognosis, diagnosis, and treatment planning. However, ensuring transparent documentation is essential for researchers to enhance reproducibility and refine techniques. Our study addresses the unique challenges presented by DL in medical imaging by developing a comprehensive checklist using the Delphi method to enhance reproducibility and reliability in this dynamic field. We compiled a preliminary checklist based on a comprehensive review of existing checklists and relevant literature. A panel of 11 experts in medical imaging and DL assessed these items using Likert scales, with two survey rounds to refine responses and gauge consensus. We also employed the content validity ratio with a cutoff of 0.59 to determine item face and content validity. Round 1 included a 27-item questionnaire, with 12 items demonstrating high consensus for face and content validity that were then left out of round 2. Round 2 involved refining the checklist, resulting in an additional 17 items. In the last round, 3 items were deemed non-essential or infeasible, while 2 newly suggested items received unanimous agreement for inclusion, resulting in a final 26-item DL model reporting checklist derived from the Delphi process. The 26-item checklist facilitates the reproducible reporting of DL tools and enables scientists to replicate the study’s results.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Mayo Clinic Artificial Intelligence Laboratory, Department of Radiology, Mayo Clinic, 200 1st St SW, Rochester, MN, 55905, USA.
- Department of Orthopedic Surgery, Orthopedic Surgery Artificial Intelligence Laboratory, Mayo Clinic, Rochester, MN, USA.
- Department of Radiology and Imaging Sciences, Emory Healthcare, Emory University, Atlanta, GA, USA.
- Department of Radiology, NYU Langone Health, New York, NY, USA.
- DasaInova, Dasa, Universidade Federal de São Paulo, São Paulo, Brazil.
- San Diego VA Health Care System, San Diego, CA, USA.
ORCID-Profile: Shahriar Faghani
Zitieren
APA
Moassefi, M., Singh, Y., Conte, G. M., Khosravi, B., Rouzrokh, P., Vahdati, S., Safdar, N., Moy, L., Kitamura, F., Gentili, A., Lakhani, P., Kottler, N., Halabi, S. S., Yacoub, J. H., Hou, Y., Younis, K., Erickson, B. J., Krupinski, E., & Faghani, S. (2024). Checklist for Reproducibility of Deep Learning in Medical Imaging. Journal of Imaging Informatics in Medicine, 37(4), 1664-1673. https://doi.org/10.1007/s10278-024-01065-2
Vancouver
Moassefi M, Singh Y, Conte GM, Khosravi B, Rouzrokh P, Vahdati S, et al. Checklist for Reproducibility of Deep Learning in Medical Imaging. Journal of Imaging Informatics in Medicine. 2024;37(4):1664-1673. doi:10.1007/s10278-024-01065-2
BibTeX
@article{Moassefi2024,
author = {Moassefi, Mana and Singh, Yashbir and Conte, Gian Marco and Khosravi, Bardia and Rouzrokh, Pouria and Vahdati, Sanaz and Safdar, Nabile and Moy, Linda and Kitamura, Felipe and Gentili, Amilcare and Lakhani, Paras and Kottler, Nina and Halabi, Safwan S and Yacoub, Joseph H and Hou, Yuankai and Younis, Khaled and Erickson, Bradley J and Krupinski, Elizabeth and Faghani, Shahriar},
title = {Checklist for Reproducibility of Deep Learning in Medical Imaging},
journal = {Journal of Imaging Informatics in Medicine},
year = {2024},
volume = {37},
number = {4},
pages = {1664--1673},
doi = {10.1007/s10278-024-01065-2}
}
EUVIMED