Pragmatic RCT of AI-Assisted Reading for Malignant Hepatic Lesions on CE-CT

NCT07768085 | Not Yet Recruiting DMC

• 1 other identifier: SH-Lion2RCT--Intervention
• Study Type: interventional
• Target: 40,000
• Locations: 1 country
• Sites: 1

Brief Summary

The purpose of this multicenter pragmatic randomized controlled trial is to determine whether AI-assisted interpretation of multiphasic liver contrast-enhanced computed tomography (CE-CT) is non-inferior to standard radiology reporting with respect to missed clinically significant malignant focal liver lesions, \*\*with the non-inferiority margin prespecified in the statistical analysis plan before enrollment\*\*, and whether it improves lesion detection, diagnostic characterization, downstream clinical management, and reporting efficiency. On AI-assisted center-days, AI results will be revealed only after the first-line radiologist has saved an unaided initial assessment and may be used to revise the final report. On control center-days, examinations will be interpreted using the standard radiology workflow without access to AI tools.

Conditions

Hepatocellular Carcinoma (HCC); Intrahepatic Cholangiocarcinoma (Icc); Hepatic Metastasis; Hepatic Hemangioma; Cyst; Focal Nodular Hyperplasia

Keywords

Dynamic Contrast-Enhanced CT; Focal Liver Lesions; Artificial intelligence; AI-assisted interpretation

Outcome Measures

Primary Outcomes (1)

• Per-participant rate of missed clinically significant liver malignancy

  Proportion of all enrolled participants with at least one reference-standard-confirmed clinically significant liver malignancy, including hepatocellular carcinoma, liver metastasis, or another primary hepatic malignancy, that was not identified or appropriately characterized in the final index CE-CT report. The reference standard will comprise histopathology when available; otherwise, 12-month imaging and clinical follow-up and relevant record linkage, adjudicated by an independent committee masked to allocation. Non-inferiority will be assessed against a margin prespecified before enrollment.

  From the index CE-CT examination through 12 months

Secondary Outcomes (5)

• Senior-review workload

  Up to 24 hours after the eligible CE-CT examination becomes available for interpretation

• Final report turnaround time

  Up to 24 hours after the complete CE-CT examination becomes available for interpretation

• Detection rate of clinically significant malignant liver lesions

  From the index CE-CT examination through 12 months

• Clinically appropriate management-change rate

  From the index CE-CT examination through 12 months

• Time to initiation of treatment for liver malignancy

  From the index CE-CT examination through 12 months

Study Arms (2)

AI-on: AI-assisted abdominal CE-CT reporting

EXPERIMENTAL

Intervention: AI-assisted interpretation of abdominal CE-CT

Other: AI-assisted abdominal CE-CT reporting

AI-off: Standard abdominal CE-CT reporting without AI assistance

ACTIVE COMPARATOR

Routine interpretation of abdominal CE-CT without access to AI system (standard of care)

Other: Standard abdominal CE-CT reporting without AI assistance

Interventions

AI-assisted abdominal CE-CT reporting OTHER

After the unaided initial assessment is saved, the locked AI system will display lesion localization, patient-level and lesion-level malignancy probabilities, prespecified lesion-class suggestions, and uncertainty or technical-failure warnings. The first-line radiologist may revise or retain the initial assessment and must record acceptance, rejection, or uncertainty for clinically important AI suggestions. The senior reviewer may view the initial assessment, AI output, and revision history. AI may not autonomously authorize reports, prescribe management, or communicate diagnoses directly to participants.

AI-on: AI-assisted abdominal CE-CT reporting

Standard abdominal CE-CT reporting without AI assistance OTHER

Eligible multiphasic abdominal CE-CT examinations will be interpreted and reported according to the site's standard radiology workflow, including routine senior review where applicable. AI system outputs will not be available to radiologists or clinical staff, and no AI-triggered safety-net review will be performed. Routine report addenda, urgent communication, multidisciplinary review, additional imaging, and subsequent clinical management will remain available according to standard clinical practice.

AI-off: Standard abdominal CE-CT reporting without AI assistance

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Adult (18-64), Older Adult (65+)

You may qualify if:

• Age range 18 years and above
• Underwent dynamic contrast-enhanced abdominal CT examination with liver coverage
• Imaging must include at least three required phases: non-contrast, arterial phase, and venous phase; a delayed phase is optional
• Complete imaging data that meet AI system and radiologist interpretation requirements.

You may not qualify if:

• History of recent upper-abdominal surgery (within 30 days) or major hepatobiliary-pancreatic surgery affecting liver evaluation (e.g., liver transplantation or Whipple procedure); patients with prior simple cholecystectomy or single-lesion interventional procedures are not excluded
• History of recent hepatic trauma (within 30 days)
• Poor image quality or severe noise artifacts (e.g., metal or motion artifacts)
• Missing required imaging phases (required at least non-contrast, arterial, and venous phases) or inadequate scan range (e.g., lower-abdomen CT such as pelvic or rectal scans not covering the liver)

Sponsors & Collaborators

• Shengjing Hospital: lead

Study Sites (1)

Shengjing Hospital of China Medical University

Shenyang, Liaoning, 110004, China

Location

Related Publications (4)

• Elias-Cabot E, Romero-Martin S, Raya-Povedano JL, Rodriguez-Ruiz A, Alvarez-Benito M. AI-based triage and decision support in mammography and digital tomosynthesis for breast cancer screening: a paired, noninferiority trial. Nat Med. 2026 Apr;32(4):1296-1305. doi: 10.1038/s41591-026-04277-x. Epub 2026 Mar 19.

  PMID: 41857202 BACKGROUND

• Woznitza N, Smith L, Rawlinson J, Au-Yong I, George B, Djearaman MG, Nair A, Lee RW, Navani N, Ndwandwe S, Clarke CS, Creeden A, Newsome J, Das I, Abaokporo S, Tucker R, Hathorn J, Baldwin DR. AI-based chest X-ray prioritization in the lung cancer diagnostic pathway: the LungIMPACT randomized controlled trial. Nat Med. 2026 May;32(5):1737-1744. doi: 10.1038/s41591-026-04253-5. Epub 2026 Mar 24.

  PMID: 41876649 BACKGROUND

• Ding W, Meng Y, Ma J, Pang C, Wu J, Tian J, Yu J, Liang P, Wang K. Contrast-enhanced ultrasound-based AI model for multi-classification of focal liver lesions. J Hepatol. 2025 Aug;83(2):426-439. doi: 10.1016/j.jhep.2025.01.011. Epub 2025 Jan 21.

  PMID: 39848548 BACKGROUND

• Ying H, Liu X, Zhang M, Ren Y, Zhen S, Wang X, Liu B, Hu P, Duan L, Cai M, Jiang M, Cheng X, Gong X, Jiang H, Jiang J, Zheng J, Zhu K, Zhou W, Lu B, Zhou H, Shen Y, Du J, Ying M, Hong Q, Mo J, Li J, Ye G, Zhang S, Hu H, Sun J, Liu H, Li Y, Xu X, Bai H, Wang S, Cheng X, Xu X, Jiao L, Yu R, Lau WY, Yu Y, Cai X. A multicenter clinical AI system study for detection and diagnosis of focal liver lesions. Nat Commun. 2024 Feb 7;15(1):1131. doi: 10.1038/s41467-024-45325-9.

  PMID: 38326351 BACKGROUND

MeSH Terms

Conditions

Carcinoma, Hepatocellular; Cholangiocarcinoma; Cirrhosis, Familial, with Pulmonary Hypertension; Cysts; Focal Nodular Hyperplasia

Condition Hierarchy (Ancestors)

Adenocarcinoma; Carcinoma; Neoplasms, Glandular and Epithelial; Neoplasms by Histologic Type; Neoplasms; Liver Neoplasms; Digestive System Neoplasms; Neoplasms by Site; Digestive System Diseases; Liver Diseases; Pathological Conditions, Anatomical; Pathological Conditions, Signs and Symptoms

Study Officials

• Yu Shi, MD

  Shengjing Hospital

  PRINCIPAL INVESTIGATOR

Central Study Contacts

Yu MD PhD Shi

CONTACT

+86 189 4025 9980; 18940259980@163.com

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: SINGLE
• Who Masked: PARTICIPANT
• Purpose: DIAGNOSTIC
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Professor of Radiology

Model Details: A cluster-randomized, center-day crossover trial with participant-level parallel assignment.

Study Record Dates

• First Submitted: August 11, 2026
• First Posted: August 17, 2026
• Study Start: August 17, 2026
• Primary Completion (Estimated): December 31, 2027
• Study Completion (Estimated): December 31, 2027
• Last Updated: August 17, 2026

Record last verified: 2026-08

Data Sharing

• IPD Sharing: Will share
• Shared Documents: STUDY PROTOCOL

Locations

CN (1)