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/

Zendy Plus

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

Zendy Tools

Zendy Open

Defining pathologies automatically from medical images aids the understandingof the emergence and progression of diseases, and such an ability is crucial inclinical diagnostics. However, existing deep learning models heavily rely onexpert annotations and lack generalization capabilities in open clinicalenvironments. In this study, we present a generalizable vision-language modelfor Annotation-Free pathology Localization (AFLoc). The core strength of AFLoclies in its extensive multi-level semantic structure-based contrastivelearning, which comprehensively aligns multi-granularity medical concepts fromreports with abundant image features, to adapt to the diverse expressions ofpathologies and unseen pathologies without the reliance on image annotationsfrom experts. We conducted primary experiments on a dataset of 220K pairs ofimage-report chest X-ray images, and performed extensive validation across sixexternal datasets encompassing 20 types of chest pathologies. The resultsdemonstrate that AFLoc outperforms state-of-the-art methods in bothannotation-free localization and classification tasks. Additionally, weassessed the generalizability of AFLoc on other modalities, includinghistopathology and retinal fundus images. Extensive experiments show that AFLocexhibits robust generalization capabilities, even surpassing human benchmarksin localizing five different types of pathological images. These resultshighlight the potential of AFLoc in reducing annotation requirements and itsapplicability in complex clinical environments.

Multi-modal vision-language model for generalizable annotation-free pathology localization and clinical diagnosis