RESUMEN
Purpose: To establish reporting adherence to the Checklist for Artificial Intelligence in Medical Imaging (CLAIM) in diagnostic accuracy AI studies with the highest Altmetric Attention Scores (AAS), and to compare completeness of reporting between peer-reviewed manuscripts and preprints. Methods: MEDLINE, EMBASE, arXiv, bioRxiv, and medRxiv were retrospectively searched for 100 diagnostic accuracy medical imaging AI studies in peer-reviewed journals and preprint platforms with the highest AAS since the release of CLAIM to June 24, 2021. Studies were evaluated for adherence to the 42-item CLAIM checklist with comparison between peer-reviewed manuscripts and preprints. The impact of additional factors was explored including body region, models on COVID-19 diagnosis and journal impact factor. Results: Median CLAIM adherence was 48% (20/42). The median CLAIM score of manuscripts published in peer-reviewed journals was higher than preprints, 57% (24/42) vs 40% (16/42), P < .0001. Chest radiology was the body region with the least complete reporting (P = .0352), with manuscripts on COVID-19 less complete than others (43% vs 54%, P = .0002). For studies published in peer-reviewed journals with an impact factor, the CLAIM score correlated with impact factor, rho = 0.43, P = .0040. Completeness of reporting based on CLAIM score had a positive correlation with a study's AAS, rho = 0.68, P < .0001. Conclusions: Overall reporting adherence to CLAIM is low in imaging diagnostic accuracy AI studies with the highest AAS, with preprints reporting fewer study details than peer-reviewed manuscripts. Improved CLAIM adherence could promote adoption of AI into clinical practice and facilitate investigators building upon prior works.
RESUMEN
Objective The novel coronavirus disease 2019 (Covid-19) has infected millions worldwide and impacted the lives of many folds more. Many clinicians share new Covid-19-related resources, research, and ideas within the online Free Open Access to Medical Education (FOAM) community of practice. This study provides a detailed content and contributor analysis of Covid-19-related tweets among the FOAM community during the first months of the pandemic. Design, Setting, and Participants In this social media content analysis study, Twitter was searched from November 1, 2019, to March 21, 2020, for English tweets discussing Covid-19 in the FOAM community. Tweets were classified into one of 13 pre-specified content categories: original research, editorials, FOAM resource, public health, podcast or video, learned experience, refuting false information, policy discussion, emotional impact, blatantly false information, other Covid-19, and non-Covid-19. Further analysis of linked original research and FOAM resources was performed. One-thousand (1000) randomly selected contributor profiles and those deemed to have contributed false information were analyzed. Results The search yielded 8541 original tweets from 4104 contributors. The number of tweets in each content category were: 1557 other Covid-19 (18.2%), 1190 emotional impact (13.9%), 1122 FOAM resources (13.1%), 1111 policy discussion (13.0%), 928 advice (10.9%), 873 learned experience (10.2%), 424 non-Covid-19 (5.0%), 410 podcast or video (4.8%), 304 editorials (3.6%), 275 original research (3.2%), 245 public health (2.9%), 83 refuting false information (1.0%), and 19 blatantly false (0.2%). Conclusions Early in the Covid-19 pandemic, the FOAM community used Twitter to share Covid-19 learned experiences, online resources, crowd-sourced advice, and research and to discuss the emotional impact of Covid-19. Twitter also provided a forum for post-publication peer review of new research. Sharing blatantly false information within this community was infrequent. This study highlights several potential benefits from engaging with the FOAM community on Twitter.