Skip to content

Artificial intelligence-based digital pathology image analysis for clinical phenotyping and prognostic assessment of steatotic liver disease

Artificial intelligence-based digital pathology image analysis for clinical phenotyping and prognostic assessment of steatotic liver disease - AI-PathSLD

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2600128316
Enrollment
Unknown
Registered
2026-07-17
Start date
2024-12-31
Completion date
Unknown
Last updated
2026-07-20

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

Conditions

Nonalcoholic Fatty Liver Disease (NAFLD) / Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD)

Interventions

AI Model Development and Validation Group:None

Sponsors

Minhang Hospital Affiliated to Fudan University
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to No maximum

Inclusion criteria

Inclusion criteria: 1. Age >= 18 years; diagnosis of non-alcoholic fatty liver disease (NAFLD/MASLD) confirmed via liver biopsy pathology or imaging. 2. Provision of high-quality pathological sections (H&E and Masson staining), complete clinical data, and lifestyle questionnaire information. 3. No history of liver transplantation or systemic anti-fibrotic intervention therapy. 4. Subject (or legal representative) has signed the informed consent form and voluntarily participates in the study.

Exclusion criteria

Exclusion criteria: 1. Substandard quality of pathological sections: blurred images, tissue damage, uneven staining, or evaluable tissue area < 1 cm^2. 2. Concomitant active liver disease: chronic hepatitis B (HBsAg-positive), chronic hepatitis C (HCV-RNA-positive), primary biliary cholangitis, autoimmune hepatitis, etc. 3. Confirmed diagnosis of hepatocellular carcinoma or other hepatic malignancies prior to study enrollment. 4. Pregnant or lactating women. 5. Receipt of immunosuppressants, targeted therapy, or other experimental interventions within the past 3 months. 6. Critical clinical data missing to the extent that core covariate analysis cannot be performed.

Design outcomes

Primary

MeasureTime frame
Incidence of HCC and cirrhosis; Predictive accuracy (AUC) of the AI model;

Secondary

MeasureTime frame
Prediction accuracy for metabolic syndrome-related complications (e.g., liver failure, portal hypertension);Clinical utility metrics (PPV, NPV, calibration curve, and decision curve analysis);Interpretability analysis of key pathological imaging features (SHAP values, Class Activation Mapping).;

Countries

China

Contacts

Public ContactChen Wei

Minhang Hospital Affiliated to Fudan University

wei_chen@fudan.edu.cn+86 180 4982 5885

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Jul 23, 2026