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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

In this work, we present an approach, which we call Embeddings forLanguage/Image-aligned X-Rays, or ELIXR, that leverages a language-alignedimage encoder combined or grafted onto a fixed LLM, PaLM 2, to perform a broadrange of chest X-ray tasks. We train this lightweight adapter architectureusing images paired with corresponding free-text radiology reports from theMIMIC-CXR dataset. ELIXR achieved state-of-the-art performance on zero-shotchest X-ray (CXR) classification (mean AUC of 0.850 across 13 findings),data-efficient CXR classification (mean AUCs of 0.893 and 0.898 across fivefindings (atelectasis, cardiomegaly, consolidation, pleural effusion, andpulmonary edema) for 1% (~2,200 images) and 10% (~22,000 images) trainingdata), and semantic search (0.76 normalized discounted cumulative gain (NDCG)across nineteen queries, including perfect retrieval on twelve of them).Compared to existing data-efficient methods including supervised contrastivelearning (SupCon), ELIXR required two orders of magnitude less data to reachsimilar performance. ELIXR also showed promise on CXR vision-language tasks,demonstrating overall accuracies of 58.7% and 62.5% on visual questionanswering and report quality assurance tasks, respectively. These resultssuggest that ELIXR is a robust and versatile approach to CXR AI.

ELIXR: Towards a general purpose X-ray artificial intelligence system through alignment of large language models and radiology vision encoders