RAPIER: Radiology-Pathology Imaging Exchange Resource

Metabolic dysfunction-associated steatotic liver disease (MASLD) is now the most common chronic liver disease worldwide, affecting an estimated 30% of adults. Singapore expects significant case growth in the years ahead, raising the risk burden of decompensated cirrhosis, liver cancer and liver-related death. Severity grading currently depends on histopathologists examining liver biopsies under high magnification to assess steatosis, fibrosis, inflammation and hepatocyte damage — a process with notable inter-observer variability.

AISG, A*STAR, Singapore General Hospital and the National Cancer Centre Singapore built STEATstat, an AI tool that quantifies liver fat precisely and reduces inter-observer variability. The result is more objective, accurate disease grading, which sharpens predictions of which patients will progress toward cirrhosis or liver cancer. STEATstat is now in production at SGH's anatomical pathology lab.

Caption language: en · Fetched: 2026-05-02

Metabolic dysfunction associated stootic liver disease is now the most common chronic liver disease worldwide with an estimated 30% of the global adult population affected. It is also common in Singapore and projections suggest significant increase in the cases in the years to come. Of more concern is the potential associated clinical burden of decompensated cerosis, liver cancer and liver mortality cases. Imaging and hystopathology tools play a critical role in the diagnosis and risk stratification of patients, guiding clinicians to better manage patients and improving outcomes in liver disease. Assessment of the severity of metabolic dysfunction associated stic liver disease or muscle D is performed by a hisystopathologist who will examine the liver biopsy tissue under high magnification with a microscope.

The liver tissue is evaluated for the presence of fat or hippatic sttosis among other features such as fibrosis, inflammation and liver cell damage. Special hisystochemical stains may be utilized to highlight such features. Aside from identifying these features, the hystopathologist will quantify and grade the liver fat into categories of severity. Our assistive AI fat estimation tool atly named STO stat will allow for precise quantification of liver fat and reduces interobserver favorability. This increases the objectivity of hystopathologists allowing for accurate grading of disease which predicts if the patient will be at risk of progressive liver disease such as cerosis or liver cancer ultimately resulting in better patient care.

STEAT is now deployed at the Singapore General Hospital anatomical pathology laboratory assisting pathologist to make faster and more precise assessments of fatty liver disease. [Music]