Skip to content

Construction of Multi-modal Intelligent Diagnosis Model for Liver Fibrosis Based on Ultrasound Images

Construction of Multi-modal Intelligent Diagnosis Model for Liver Fibrosis Based on Ultrasound Images

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2600116070
Enrollment
Unknown
Registered
2026-01-05
Start date
2026-01-05
Completion date
Unknown
Last updated
2026-01-12

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Liver fibrosis

Interventions

Gold Standard:Ultrasound-guided percutaneous liver biopsy and its pathological evaluation.
Index test:Intelligent fusion model based on deep learning, whose input is multimodal data and output is liver fibrosis stage.

Sponsors

The Seventh Affiliated Hospital Sun Yat-sen University
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to No maximum

Inclusion criteria

Inclusion criteria: 1.Adult patients aged 18 or above. 2. Due to the clinical diagnosis and treatment needs of chronic liver diseases, based on the judgment of the patient's attending physician, it has been independently decided and planned to undergo percutaneous liver biopsy, and a clear pathological stage report of liver fibrosis (F0-F4) can be obtained. 3. Before and after liver biopsy (usually within 3 months), the full set of multimodal ultrasound examinations required for the project was completed in our institution, including: two-dimensional gray-scale ultrasound of the liver (static images and dynamic videos), shear wave elastography, attenuation imaging, viscoelastic imaging, and color Doppler ultrasound of the splenic vein. 4. During the same period, it has complete records of clinical laboratory test indicators, including but not limited to: AST, ALT, platelet count, etc., which can be used to calculate the FIB-4 index.

Exclusion criteria

Exclusion criteria: 1. Insufficient length of liver biopsy tissue specimens or poor quality of pathological reports make it impossible to conduct reliable liver fibrosis staging. 2. There are factors that seriously affect the quality of ultrasound images, such as severe obesity and intestinal gas interference, which make it impossible to obtain or measure key parameters (such as elasticity value, attenuation coefficient, viscoelasticity value) reliably. 3. Combined with other types of liver space-occupying lesions (such as liver cancer, huge hemangioma) or extrahepatic biliary obstruction. 4. Has undergone liver surgery or interventional treatment (such as liver lobectomy, TACE). 5. There is a serious lack of clinical data, making it impossible to complete data matching and integration.

Design outcomes

Primary

MeasureTime frame
Diagnostic Accuracy Assessment;

Secondary

MeasureTime frame
Analysis of Variability in Diagnostic Accuracy (Subgroup Analysis);

Countries

China

Contacts

Public ContactZuofeng Xu

The Seventh Affiliated Hospital of Sun Yat-sen University

xuzuofeng77@aliyun.com+86 755 8120 6580

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Feb 4, 2026