AI Blood Test Enhances Monitoring of Liver Cirrhosis Progression
Posted on 25 Apr 2026
Monitoring chronic liver disease remains difficult because clinicians rely on tools that can be inconsistent and may miss early progression. Standard approaches often combine ultrasound imaging with blood-based protein markers, which can delay recognition of worsening disease. A new artificial intelligence system now analyzes epigenetic patterns in circulating DNA to deliver organ-specific insights for chronic disease monitoring.
Curve Biosciences has announced computational and clinical advances from its Whole-Body Intelligence platform, an AI system designed to uncover organ-specific biology from blood using the company’s Whole-Body Atlas. The platform is delivered through Whole-Body Blood Tests that analyze epigenetic patterns in circulating DNA to generate detailed insights into organ health and chronic disease progression. The Whole-Body Atlas, which the company notes is the largest manually curated collection of tissues characterized by organ and disease state, enables the system to interpret epigenetic signatures with high specificity.
In new computational work accepted for presentation at the International Conference on Learning Representations (ICLR) on April 27, 2026, Curve reports that generalist genomic foundation models can be continually pre-trained on Whole-Body Atlas data to understand methylation patterns from sequence alone. This enables epigenetic signals to emerge directly within model representations. The presentation, titled “Methylation-Aware Embedding Geometry Emerges from Bisulfite Pretraining in DNA Language Models,” will be delivered in the Learning Meaningful Representations of Life session.
The company also completed a large, multicenter study evaluating its test for monitoring liver cirrhosis. The study enrolled 1,482 patients across 23 clinical sites, with all blood samples analyzed by the Whole-Body Blood Test. Models were first pre-trained on the Whole-Body Atlas, then trained on blood data from 885 patients, and finally evaluated in a fully blinded, independent cohort of 597 patients, where the test showed strong performance.
Curve is preparing a manuscript for peer-reviewed publication and presentation of the study findings as part of its path toward regulatory and reimbursement approval for a liver cirrhosis monitoring test, while continuing investment in Whole-Body Intelligence research.
“This study shows how Whole-Body Intelligence can transform how chronic liver disease is monitored, enabling earlier detection of disease progression and more precise clinical decision-making,” said Ritish Patnaik, Ph.D., CEO & Co-Founder, Curve Biosciences.
“These findings validate our approach of combining high-resolution biological data with advanced AI to unlock clinically meaningful signals from blood. A more accurate, non-invasive monitoring tool has the potential to improve outcomes by identifying disease progression earlier, when disease interventions are most effective,” said Amit Singal, M.D., Chief Medical Officer, Curve Biosciences and Chief of Hepatology at the University of Texas Southwestern Medical Center.
Related Links
Curve Biosciences
.jpg)
