New AI Blood Test Detects Silent Liver Disease Before Symptoms Appear

Scientists at the Johns Hopkins Kimmel Cancer Center have created an AI-driven blood test that detects liver fibrosis and cirrhosis years before symptoms emerge. The test examines genome-wide patterns of cell-free DNA (cfDNA) fragments in blood, including repetitive DNA regions, to uncover disease markers. Published in *Science Translational Medicine* and supported by the National Institutes of Health, this is the first large-scale use of fragmentome technology—originally developed for cancer—for non-cancer chronic diseases. The study analyzed whole-genome sequencing data from 1,576 individuals, evaluating 40 million DNA fragments across thousands of genomic locations. Machine learning identified fragmentation patterns linked to early liver disease, advanced fibrosis, and cirrhosis with high accuracy. Unlike traditional liquid biopsies that focus on mutations, this method assesses how DNA breaks, packages, and distributes, providing broader health insights. Co-led by Victor Velculescu, Robert Scharpf, and Jill Phallen, the research suggests early detection could reverse liver fibrosis before it progresses to cirrhosis or liver cancer. Current tests often miss early-stage disease, leaving millions at risk. The team, including first author Akshaya Annapragada, believes this approach could extend beyond liver disease to other chronic conditions. With an estimated 100 million Americans at risk for liver disease, the test offers a scalable, non-invasive solution. Existing methods like ultrasounds or MRIs are costly and less reliable for early detection. The AI classifier’s ability to analyze fragmentomes could revolutionize preventive healthcare by identifying high-risk individuals before irreversible damage occurs.