AI-Assisted Chest CT Report Drafting to Enhance Radiologist Workflow and Maintain Quality A Two-Phase Multicenter Controlled Trial Evaluating Diagnostic Reporting

What this Trial Is About

Researchers are evaluating an artificial intelligence AI system designed to assist experienced radiologists in drafting chest CT scan reports more quickly without reducing report quality. This two-phase study involves board-certified radiologists interpreting complex chest CT cases, comparing report writing with AI-generated drafts to their usual method of writing reports from scratch. The study aims to determine if AI can improve efficiency while maintaining high standards of patient care. The study consists of two phases. In Phase 1, a controlled crossover design uses historical chest CT cases where radiologists interpret the same cases with and without AI draft assistance, with randomized order and a washout period. In Phase 2, the AI system is integrated into the clinical workflow for real-time use on new chest CT scans, allowing radiologists to use AI drafts during routine practice. The AI generates preliminary report drafts that radiologists review and edit as needed. Participants are board-certified radiologists with at least three years of thoracic imaging experience. During the study, the time taken to complete reports and the quality of the final reportsassessed by blinded senior cliniciansare measured. The study also evaluates the significance of edits made to AI drafts and monitors system usability and workflow integration. The trial spans from mid-2026 to early 2027, with data analysis planned shortly after completion.

Evaluating a Text-Prompt AI Assistant for Chest CT Scans (AI-REPORT Study)