Skip to content
DataLookout
Menu

NCT07899931

Vision-Language-Based Large-Scale Model for Emergency Screening

StatusRecruiting
PhaseNot applicable
Primary completion (estimated)
Track this trial

Free. We email you which fields changed when ClinicalTrials.gov updates this record.

Record History

All 1 version: 1 with changes below.

  1. Study First Submitted

Endpoints

Primary

Per-case CT diagnostic decision time
Timeframe: Day 1 (at completion of CT image reading)

Secondary (1)

Sensitivity for detecting critical emergency lesions on emergency thoraco-abdominal CT

Arms and interventions

powER vision-language large-model assisted CT readingExperimental

Radiologists interpret emergency thoraco-abdominal CT scans with auxiliary outputs generated by the powER vision-language large model. This AI system identifies suspected critical lesions and presents these findings as references to support radiological review. Radiologists retain the final diagnostic decision. All participants receive routine standard clinical management.

Conventional CT reading without AI assistanceActive comparator

Radiologists perform independent reading of emergency thoraco-abdominal CT on standard PACS workstation, without any AI auxiliary tools. Diagnosis relies solely on radiologists' conventional clinical experience.

Diagnostic tests: powER vision-language large-model assisted CT reading, Conventional CT reading without AI assistance

Sites

1 site in 1 country

  • The First Hospital of Jilin UniversityChangchun, Jilin, China