Scientists have developed a new artificial intelligence–powered blood test that may be able to detect early signs of serious liver disease long before symptoms appear.
The research was carried out by scientists at the Johns Hopkins Kimmel Cancer Center in the United States and was partly funded by the National Institutes of Health.
The findings were published on March 4 in the journal Science Translational Medicine. This new test is based on a technology known as a liquid biopsy.
Instead of taking tissue directly from an organ, a liquid biopsy studies tiny pieces of DNA that circulate in the bloodstream.
These DNA fragments come from cells throughout the body that naturally break down and release genetic material into the blood.
By analyzing these fragments, scientists can learn important information about what is happening inside the body. Liquid biopsies have been studied for several years, mostly as a way to detect cancer.
But the new study shows that the same technology may also help identify other serious illnesses, including liver disease.
In this research, scientists used artificial intelligence to examine patterns in fragments of cell-free DNA, also known as cfDNA. These fragments float in the bloodstream after cells die and release their genetic material.
The researchers looked at how the DNA pieces were broken apart and where they appeared across the entire human genome.
This approach is known as fragmentome analysis because it studies the complete pattern of DNA fragments found in the blood. The team used advanced machine learning tools to analyze these patterns.
Machine learning is a type of artificial intelligence that can find hidden patterns in extremely large datasets. By training the system to recognize certain DNA fragmentation patterns, the researchers were able to identify signals linked to liver disease. The study focused on conditions such as liver fibrosis and cirrhosis.
Liver fibrosis occurs when repeated damage causes scar tissue to build up in the liver. If the damage continues, the disease can progress to cirrhosis, a severe stage in which the liver becomes heavily scarred and cannot function properly. Cirrhosis greatly increases the risk of liver failure and liver cancer.
One of the biggest challenges in treating liver disease is that it often develops quietly. Many people do not know they have a problem until the damage has become severe. Early detection is very important because liver fibrosis can sometimes be reversed if treated early enough.
In the study, researchers analyzed blood samples from 1,576 individuals who had liver disease or related health conditions. They performed whole genome sequencing on the DNA fragments circulating in the blood. This means they examined DNA signals across the entire genome rather than focusing on only a few genes.
The scientists studied both the size of the DNA fragments and their distribution across thousands of regions of the genome. Each test involved examining around 40 million DNA fragments, creating a massive amount of data. Machine learning algorithms then analyzed this information to identify patterns linked to disease.
The results showed that these fragmentation patterns could detect early stages of liver fibrosis as well as advanced fibrosis and cirrhosis with high sensitivity. Dr. Victor Velculescu, co-director of the cancer genetics and epigenetics program at the Johns Hopkins Kimmel Cancer Center and co-senior author of the study, explained that the work builds on earlier research focused on cancer detection.
He said that by studying genome-wide fragmentation patterns of cell-free DNA, scientists can now apply the same approach to chronic diseases. He noted that early detection of liver disease could make a major difference for patients. If fibrosis is detected early, doctors may be able to treat the underlying causes before the condition progresses to cirrhosis or liver cancer.
What makes this method different from many existing liquid biopsy tests is that it does not search for individual gene mutations. Instead, it analyzes the entire pattern of DNA fragments across the genome. According to the researchers, this broader approach provides a much richer source of information about a person’s overall health.
First author Akshaya Annapragada, an M.D./Ph.D. student in the Velculescu laboratory, explained that the fragmentome contains a huge amount of biological information about how cells behave in the body. When combined with artificial intelligence, these large datasets allow scientists to develop specific detection systems for different diseases.
Another part of the study explored whether fragmentome patterns could reveal information about overall health status. The researchers developed what they called a fragmentation comorbidity index. This measurement helped distinguish people with many serious medical conditions from those with fewer health problems.
The index was compared with the Charlson Comorbidity Index, a well-known tool doctors use to estimate a patient’s risk of death based on existing illnesses. In some cases, the fragmentome-based index was able to predict survival outcomes more precisely than traditional inflammatory markers used in blood tests.
The researchers also noticed signals related to other types of diseases, including cardiovascular, inflammatory, and neurodegenerative disorders. However, there were not enough cases in the study to build separate detection models for each condition. Still, the findings suggest that fragmentome analysis may eventually help detect a wide range of chronic diseases.
The scientists estimate that around 100 million people in the United States may have conditions that increase their risk of liver fibrosis or cirrhosis. Unfortunately, current blood tests often fail to detect early stages of liver disease. Standard markers usually identify cirrhosis only about half the time.
Imaging tests such as special ultrasound scans or MRI scans can help detect liver damage, but these technologies require expensive equipment and may not always be available. This new AI-driven liquid biopsy could potentially offer a simpler and more accessible screening tool.
However, the researchers emphasize that the test is still a prototype and is not yet available for clinical use. The next step will be to refine the system and confirm its accuracy in larger studies. The team also plans to explore how fragmentome technology might be used to detect other chronic diseases in the future.
Reviewing the study’s findings shows why this research is so promising. The approach does not depend on finding specific gene mutations, which limits many existing diagnostic tests. Instead, it studies overall DNA fragmentation patterns that reflect how cells are functioning throughout the body.
This broader biological signal may allow scientists to detect disease earlier and more accurately. If future studies confirm these results, AI-based fragmentome testing could become a powerful new tool for early disease detection. Earlier diagnosis would allow doctors to treat conditions sooner, potentially preventing severe complications such as cirrhosis or cancer.
While more work is needed before the test becomes widely available, the research represents an important step toward using artificial intelligence and genomic data to transform how chronic diseases are diagnosed and monitored.
If you care about liver health, please read studies that refined fiber is link to liver cancer, and the best and worst foods for liver health.
For more health information, please see recent studies about how to boost your liver naturally, and simple ways to detox your liver.
Copyright © 2026 Knowridge Science Report. All rights reserved.
