Cancer is associated with significant morbimortality globally. Advances in screening, diagnosis, management and survivorship were substantial in the last decades, however, challenges in providing p... Plain language summary Cancer is associated with significant morbimortality globally. Although significant advances occurred in the last decades, challenges in providing personalized care remain. A...

The COVID-19 pandemic has affected all aspects of society. Researchers worldwide have been working to provide new solutions to and better understanding of this coronavirus. In this research, our goal was to perform a Bibliometric Network Analysis (BNA) to investigate the strategic themes, thematic evolution structure and trends of coronavirus during the first eight months of COVID-19 in the Web of Science (WoS) database in 2020. To do this, 14,802 articles were analyzed, with the support of the SciMAT software. This analysis highlights 24 themes, of which 11 of the more important ones were discussed in-depth. The thematic evolution structure shows how the themes are evolving over time, and the most developed and future trends of coronavirus with focus on COVID-19 were visually depicted. The results of the strategic diagram highlight ‘CHLOROQUINE’, ‘ANXIETY’, ‘PREGNANCY’ and ‘ACUTE-RESPIRATORY-SYNDROME’, among others, as the clusters with the highest number of associated citations. The thematic evolution. structure presented two thematic areas: “Damage prevention and containment of COVID-19” and “Comorbidities and diseases caused by COVID-19”, which provides new perspectives and futures trends of the field. These results will form the basis for future research and guide decision-making in coronavirus focused on COVID-19 research and treatments.

This prospective exploratory study conducted from January 2023 through May 2023 evaluated the ability of ChatGPT to answer questions from Brazilian radiology board examinations, exploring how different prompt strategies can influence performance using GPT-3.5 and GPT-4. Three multiple-choice board examinations that did not include image-based questions were evaluated: (a) radiology and diagnostic imaging, (b) mammography, and (c) neuroradiology. Five different styles of zero-shot prompting were tested: (a) raw question, (b) brief instruction, (c) long instruction, (d) chain-of-thought, and (e) question-specific automatic prompt generation (QAPG). The QAPG and brief instruction prompt strategies performed best for all examinations (P < .05), obtaining passing scores (≥60%) on the radiology and diagnostic imaging examination when testing both versions of ChatGPT. The QAPG style achieved a score of 60% for the mammography examination using GPT-3.5 and 76% using GPT-4. GPT-4 achieved a score up to 65% in the neuroradiology examination. The long instruction style consistently underperformed, implying that excessive detail might harm performance. GPT-4's scores were less sensitive to prompt style changes. The QAPG prompt style showed a high volume of the "A" option but no statistical difference, suggesting bias was found. GPT-4 passed all three radiology board examinations, and GPT-3.5 passed two of three examinations when using an optimal prompt style. Keywords: ChatGPT, Artificial Intelligence, Board Examinations, Radiology and Diagnostic Imaging, Mammography, Neuroradiology © RSNA, 2023 See also the commentary by Trivedi and Gichoya in this issue.

The RSNA Abdominal Traumatic Injury CT (ie, RATIC) dataset contains 4274 abdominal CT studies with annotations related to traumatic injuries and is available at https://www.kaggle.com/competitions/rsna-2023-abdominal-trauma-detection and https://imaging.rsna.org/dataset/5 .

The RSNA Lumbar Degenerative Imaging Spine Classification dataset is the largest publicly available adult MRI lumbar spine dataset for degenerative disease. The dataset includes multisequence, multiplanar MRI scans from 2697 patients through contributions from eight institutions across six countries and five continents.

This article explains the inner workings of transformer models, helping radiologists understand their functionality and build trust in artificial intelligence systems as decision-support tools, rather than seeing them as “black boxes.”