AI Screening for Biliary Tract Cancer on Routine CT Scans Without Contrast
Tests an AI system that looks for biliary tract cancer on routine CT scans.
Plain-language title generated from the registry record
Registry title: Prospective AI-Assisted Screening of Biliary Tract Cancer Using Non-Contrast CT
Key facts
- Where
- Beijing, ChinaSee the site
Where in China
- Beijing Tsinghua Changgung 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
- Beijing Tsinghua Chang Gung Hospital
- Trial code
- NCT07911397
- Record updated
- 2 November 2026
- Original description
Biliary tract cancer is a group of malignant tumors involving the intrahepatic bile ducts, extrahepatic bile ducts, and gallbladder. Early detection can be challenging because non-contrast computed tomography (NCCT), which is commonly performed in routine clinical practice, may provide limited information for identifying biliary tract cancer. This prospective study aims to clinically validate an artificial intelligence (AI) system for screening and detecting biliary tract cancer on NCCT images. The AI system will independently analyze eligible NCCT examinations and generate a patient-level risk score for biliary tract cancer. The study will evaluate the diagnostic performance of the AI system against a predefined clinical reference standard. The primary objective is to assess the ability of the AI system to distinguish patients with biliary tract cancer from those without biliary tract cancer using NCCT. Secondary objectives include evaluating sensitivity, specificity, positive predictive value, and negative predictive value. The study is designed to evaluate the clinical performance and potential screening value of AI-assisted analysis of NCCT. Routine clinical care and clinical decision-making will not be replaced by the AI system.Read full original descriptionShow less
Prospective Clinical Validation of an Artificial Intelligence System for Biliary Tract Cancer Screening Using Non-Contrast Computed Tomography
Biliary tract cancer is a group of malignant tumors involving the intrahepatic bile ducts, extrahepatic bile ducts, and gallbladder. Early detection can be challenging because non-contrast computed tomography (NCCT), which is commonly performed in routine clinical practice, may provide limited information for identifying biliary tract cancer. This prospective study aims to clinically validate an artificial intelligence (AI) system for screening and detecting biliary tract cancer on NCCT images. The AI system will independently analyze eligible NCCT examinations and generate a patient-level risk score for biliary tract cancer. The study will evaluate the diagnostic performance of the AI system against a predefined clinical reference standard. The primary objective is to assess the ability of the AI system to distinguish patients with biliary tract cancer from those without biliary tract cancer using NCCT. Secondary objectives include evaluating sensitivity, specificity, positive predictive value, and negative predictive value. The study is designed to evaluate the clinical performance and potential screening value of AI-assisted analysis of NCCT. Routine clinical care and clinical decision-making will not be replaced by the AI system.
Inclusion: Adults aged 18 years or older; Undergoing abdominal non-contrast computed tomography (NCCT) as part of routine clinical evaluation or health examination; NCCT examination provides adequate coverage of the liver and biliary tract; Complete NCCT imaging data with sufficient image quality for analysis by the artificial intelligence system; Sufficient clinical information and/or follow-up data are available to establish the predefined clinical reference standard.
Exclusion: Previous major hepatobiliary or pancreatic surgery that substantially alters the anatomy of the liver or biliary tract and may affect AI analysis; Recent major upper-abdominal surgery or hepatic trauma that may substantially affect evaluation of the liver or biliary tract; Poor image quality or severe image artifacts that preclude reliable AI analysis, including severe motion, metal, or noise artifacts; Inadequate scan coverage that does not include the liver and biliary tract; Duplicate examinations or repeated enrollment of the same participant; Insufficient clinical information or follow-up to establish the predefined clinical reference standard.