Approved Research

Predictive quantification of nonalcoholic steatohepatitis clinical parameters liver steatosis and fibrosis using non-invasive medical imaging

Lay summary

Aims: this study is to validate non-invasive  Artificial Intelligence (AI)-based imaging signatures, namely iBiopsy®, for nonalcoholic steatohepatitis (NASH) diagnosis, liver fibrosis assessment, and steatosis quantification.

Scientific rationale: non-alcoholic fatty liver disease (NAFLD)/NASH is one main cause of chronic liver disease worldwide and will probably emerge as the leading cause of end-stage liver disease in the coming decades. The differentiation between simple steatosis and fibrosis is paramount in identifying patients at high risks of complications. Liver biopsy, as the gold standard for diagnosing progressive NASH, is an invasive procedure with sampling error and risk of complications. Non-invasive serum biomarkers show limited accuracy in diagnosing NASH and staging NASH-related fibrosis. Recently, the imaging biomarker Magnetic Resonance Imaging Proton Density Fat Fraction (MRI-PDFF) is used to quantify liver steatosis and shows a high correlation with histologically determined steatosis grades. However, information regarding the presence of liver inflammation and fibrosis is not provided. Yet, the development of noninvasive methods to accurately diagnose NASH and measure related fibrosis, thereby supplanting the need for liver biopsy, remains an active area of research.

Project duration: a duration of 36 months is estimated:

* 1st - 10th month: data collection, image revisions, and quality control, image preprocessing if necessary.

* 10th - 24th month: imaging features extraction, model utilization, creation of signatures of disease (novel imaging biomarkers), accuracy assessment, reproducibility and repeatability assessment  

* 24th - 36th month: additional analysis, final report generation, manuscript redaction, and submission.

Public health impact: Median Technologies is a well-known player in medical imaging and has recently developed AI-based quantitative assays to diagnose and stage liver fibrosis in NASH patients with large potential for clinical routine use upon validation.