CTEPH Differential Diagnosis Assisted by Enhanced Large Language Model Support
Key facts
- Where
- Beijing, ChinaSee the site
Where in China
- China-Japan Friendship HospitalBeijing, ChinaNo longer taking part
Questions to ask the study team
- How often do I need to come in, and for how long each time?
- Are travel costs paid?
- Can I keep my own doctor?
- What happens if I want to stop?
Trial details
- Who runs it
- China-Japan Friendship Hospital
- Trial code
- NCT07902011
- Record updated
- 29 October 2026
- Original description
This randomized controlled trial aims to evaluate whether a locked large language model (LLM)-based clinical decision support system improves the differential diagnosis of pulmonary artery filling defects among physicians who do not specialize in pulmonary vascular diseases. Participants will evaluate 30 de-identified clinical cases involving chronic thromboembolic pulmonary hypertension (CTEPH), acute pulmonary embolism (PE), and pulmonary artery sarcoma (PAS), with or without LLM assistance. The primary outcome is sensitivity for identifying CTEPH, while secondary outcomes include overall diagnostic accuracy, time to diagnosis, appropriateness of referral recommendations, and AI-related diagnostic errors.Read full original descriptionShow less
Large Language Model-Assisted Differential Diagnosis of Pulmonary Artery Filling Defects: A Stratified Randomized Controlled Trial
This randomized controlled trial aims to evaluate whether a locked large language model (LLM)-based clinical decision support system improves the differential diagnosis of pulmonary artery filling defects among physicians who do not specialize in pulmonary vascular diseases. Participants will evaluate 30 de-identified clinical cases involving chronic thromboembolic pulmonary hypertension (CTEPH), acute pulmonary embolism (PE), and pulmonary artery sarcoma (PAS), with or without LLM assistance. The primary outcome is sensitivity for identifying CTEPH, while secondary outcomes include overall diagnostic accuracy, time to diagnosis, appropriateness of referral recommendations, and AI-related diagnostic errors.
Inclusion: Holds a valid physician qualification, or is enrolled in a standardized residency training program and is permitted to participate in clinical care under supervision; Has not previously reviewed the validation cases or their final diagnoses; Is willing to participate and provides informed consent; Is able to use the standardized study platform and complete two formal case-reading sessions covering all 30 cases.
Exclusion: Has a specialist professional background in pulmonary vascular disease, pulmonary hypertension, or thoracic imaging; Participated in validation-case selection, reference-standard confirmation, model development, prompt optimization, or review of locked model outputs; Is unable to use the standardized platform or to comply with independent case interpretation and the prohibition on external generative AI tools; Has another circumstance that, in the investigator's judgment, could materially compromise independent case interpretation.
The study participants are clinicians; patients represented in the de-identified case materials are not enrolled as study participants.