English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Background: General-purpose large language models that utilize both text andimages have not been evaluated on a diverse array of challenging medical cases.  Methods: Using 934 cases from the NEJM Image Challenge published between 2005and 2023, we evaluated the accuracy of the recently released GenerativePre-trained Transformer 4 with Vision model (GPT-4V) compared to humanrespondents overall and stratified by question difficulty, image type, and skintone. We further conducted a physician evaluation of GPT-4V on 69 NEJMclinicopathological conferences (CPCs). Analyses were conducted for modelsutilizing text alone, images alone, and both text and images.  Results: GPT-4V achieved an overall accuracy of 61% (95% CI, 58 to 64%)compared to 49% (95% CI, 49 to 50%) for humans. GPT-4V outperformed humans atall levels of difficulty and disagreement, skin tones, and image types; theexception was radiographic images, where performance was equivalent betweenGPT-4V and human respondents. Longer, more informative captions were associatedwith improved performance for GPT-4V but similar performance for humanrespondents. GPT-4V included the correct diagnosis in its differential for 80%(95% CI, 68 to 88%) of CPCs when using text alone, compared to 58% (95% CI, 45to 70%) of CPCs when using both images and text.  Conclusions: GPT-4V outperformed human respondents on challenging medicalcases and was able to synthesize information from both images and text, butperformance deteriorated when images were added to highly informative text.Overall, our results suggest that multimodal AI models may be useful in medicaldiagnostic reasoning but that their accuracy may depend heavily on context.

Accuracy of a Vision-Language Model on Challenging Medical Cases