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

Locating pathologies automatically from medical images aids the understandingof the emergence and progression of diseases, and such an ability cansignificantly benefit clinical diagnostics. However, existing deep learningmodels heavily rely on expert annotations and lack generalization capabilitiesin open clinical environments. In this study, we present a generalizablevision-language pre-training model for Annotation-Free pathology Localization(AFLoc). The core strength of AFLoc lies in its image annotation-freemulti-level semantic structure-based contrastive learning, whichcomprehensively aligns multi-granularity medical concepts from reports withabundant image features, to adapt to the diverse expressions of observed andemerging unseen pathologies. We conducted extensive experimental validationacross 4 distinct external datasets, encompassing 11 types of chestpathologies, to verify its generalization ability. The results demonstrate thatAFLoc surpasses 6 state-of-the-art methods and even outperforms the humanbenchmark in locating 5 different pathologies, underscoring its suitability forcomplex clinical environments.

Generalizable vision-language pre-training for annotation-free pathology localization