VINIF.2020.DA06 – Prediction for the progression of liver fibrosis and the development of hepatocellular carcinoma (HCC) and end-stage liver diseases: A novel approach using artificial intelligence technique

Principle Investigator
Prof.Dr. Tran Binh Giang
Host Organization
Viet Duc University Hospital

In Vietnam, primary liver cancer is the leading cancer in terms of both incidence and mortality, accounting for 15.4% and 22.1% of total cancer cases, respectively. Hepatocellular carcinoma (HCC) is the most common type of liver cancer, accounting for 90-95% of all primary liver cancer cases in Vietnam. Cirrhosis is the most important risk predictor for the incidence of HCC and other end-stage liver diseases in patients with non-alcoholic fatty liver disease (NAFLD). Therefore, it is paramount to identify determinants that predict the progression of cirrhosis to advanced fibrosis and the risk of developing HCC and/or other end-stage liver disease.

Project goals:

  1. Identify determinants, including imaging methods (eg, VCTE and 2D-SWE) and noninvasive biomarkers (aspartate aminotransferase (AST) platelet ratio (APRI), AST/ALT ratio, FIB-4, NAFLD index and BARD index), which are associated with the progression of cirrhosis and end-stage liver diseases, including hepatocellular carcinoma.
  2. Develop an integrated model to predict the progression of cirrhosis and end-stage liver diseases using state-of-the-art artificial intelligence methods.

Impact:

Short term:

– Direct impact: For the first time, we will identify the determinants of progression from cirrhosis to HCC and other end-stage liver diseases. These new determinants will be used to develop an integrated prediction model for the progression of liver fibrosis and other end-stage liver diseases including hepatocellular carcinoma.

– Indirect impact: comprehensive standardized protocols and databases from clinic, histopathology, non-invasive biomarkers and imaging data have all been developed from this research and is ready to be applied for further research.

Mid-term:

– Models predicting progression from cirrhosis to HCC and other end-stage liver diseases developed from the current study will be used both at Viet Duc Hospital and widely used in other clinics and hospitals nationwide.

– The results obtained from the current study will be used for the first time as solid evidence, to contribute to recommendations on prevention and treatment of cirrhosis patients in Vietnam.

Long-term:

Currently, the incidence and mortality rate of primary liver cancer in Vietnam is ranked number one among all types of cancer. While the incidence of primary liver cancer in Vietnam due to chronic hepatitis B and C has decreased, primary liver cancer associated with obesity, metabolic diseases and type 2 diabetes has increased in the past decade. This study identifies/classifies cirrhosis patients at high risk of progression to HCC and other end-stage liver diseases. The main outcome of the present study, therefore, will contribute to reducing the incidence of primary liver cancer and liver-related mortality. The ultimate long-term impact is to reduce the cancer burden in general and the burden due to liver cancer in particular in Vietnam.

Principle Investigator
Prof.Dr. Tran Binh Giang
Host Organization
Viet Duc University Hospital

Tags

Expect Progress
01/10/2020
30/10/2021
Phase 1

– Develop procedures for collecting, extracting, diagnosing and verifying clinical, imaging, pathology/cytology, and biochemical data
– Develop procedures for collecting, transporting, processing and separating, and storing biological samples
– Standardized process and system for integrated model to develop comprehensive data from clinical, histopathological, biochemical and diagnostic imaging, using modern machine learning techniques
– Bank of specimens, biological samples, including blood, stool, urine, throat swab samples, biopsy/histology), image diagnostic data (from FibroScan and 2D-SWE

31/12/2022
Phase 2

– Comprehensive RedCap standardized database, including clinical, anatomical/histopathological, non-invasive biomarker data and imaging
– Machine learning model ready for use in clinical settings to predict progression from cirrhosis to primary hepatocellular carcinoma and other end-stage liver diseases

31/10/2023
Phase 3

– 2-3 Q1 journal articles
– 01 patent application accepted
– 01 Utility Solution Patent for which applications are accepted

Tags