Fatty liver disease is becoming a quiet global health problem because it can progress for years before a person knows anything is wrong. The core issue is simple but serious: fat builds up in the liver, causing inflammation, cell damage, and fibrosis, the scar tissue that can mark worsening disease.
The opportunity for AI is not to invent a new diagnosis from nowhere. It is to make better use of information hospitals and doctors already collect, then help clinicians decide who needs closer attention before the disease reaches a dangerous stage.
Why early detection matters
In a normal, healthy liver, fat is negligible. In fatty liver disease, many adults and even children have livers where fat exceeds 5 percent or even 10 percent of the organ's total weight. That excess fat can trigger inflammation, cell damage, and scarring.
The condition now impacts approximately 30 percent of adults worldwide. It can also move silently. According to the source, fatty liver disease typically develops without noticeable symptoms, which means it is rarely detected while it is still early and more treatable.
The consequences can be severe if the disease is allowed to advance. Progressive fat accumulation can ultimately lead to liver failure, and it has been linked to an increased risk of cardiovascular disease and various cancers. Even in cases of cirrhosis or advanced scarring of the liver, three-quarters of people are only diagnosed once their condition has become life-threatening.
This is the central gap AI researchers are trying to address: the health system often has useful signals, but those signals may not be gathered, calculated, or acted on early enough.
How AI could search existing medical data
Jeffrey Lazarus, a professor at the CUNY Graduate School of Public Health and Health Policy, sees a role for AI in reviewing large volumes of electronic health records. The goal would be to identify people who may already have worrying amounts of liver fat and should be prioritized for follow-up.
AI can retrospectively go through massive numbers of hospital visits and lab reports
Lazarus says that this kind of review could help prioritize people at the highest risk. That matters because clinicians already face growing workloads and administrative pressure. A useful tool would need to fit into existing workflows rather than add a new manual burden.
Jonathan Dranoff, a professor of medicine at Yale University, described the practical requirement clearly: clinicians need something that can run in the background or be triggered easily. In other words, AI would be most useful if it turned routine data into a prompt for action without requiring doctors to rebuild the whole screening process around it.
Blood tests and Fib-4 could become easier to use
Doctors already have simple, noninvasive ways to assess liver health, but the source notes that they are rarely used, even among people at higher risk of the consequences of fatty liver disease, including those with obesity and type 2 diabetes.
One example is the Fib-4 index. It rates a person's risk of advanced liver fibrosis by computing a score of between 0 and 6, based on age, levels of two liver enzymes, and blood-clotting ability. The calculation requires a liver blood test, which is often carried out as part of an annual medical checkup in the US.
Doctors can also use a second-line blood test called the enhanced liver fibrosis test. It measures the levels of two proteins involved in creating scar tissue and an enzyme that inhibits the clearance of scars in the liver.
Using both tests in patients with worrying amounts of liver fat has been shown to improve the diagnosis of advanced fibrosis by four-fold. But simply asking physicians to do more testing is not considered sustainable when workloads are already high.
This is where AI could be useful in a basic, practical way. Dranoff and Lazarus both foresee systems that take existing data from routine blood testing and automate Fib-4 score calculations. That could help primary care physicians identify which patients should be referred to a liver specialist.
X-rays and newer algorithms may widen the net
AI could also find signs of fatty liver disease in images collected for other reasons. Last year, scientists at Osaka Metropolitan University in Japan published a study using an AI model to analyze routine chest x-ray scans. Although those scans are mainly intended to examine the lungs and heart, they also capture parts of the liver.
The researchers found that the model could identify people with fatty liver disease with an accuracy of 82 percent. Lazarus suggests that AI-powered algorithms could be built into routine x-ray analysis when the liver appears alongside other organs.
In that kind of workflow, AI could detect excess liver fat, check for risk factors such as whether a person is overweight, has high cholesterol or type 2 diabetes, and then recommend further evaluation by the right doctor. The point is not that every x-ray becomes a liver test. It is that images already being taken might reveal findings that otherwise go unnoticed.
Blood-test-based AI is also advancing. The Danish health tech startup Evido has developed an AI-powered algorithm called LiverPRO, which assesses liver fibrosis risk based on age and nine routine blood-based biomarkers. It is now being commercialized in partnership with Roche and has been shown to outperform Fib-4 in predicting risk of serious liver problems in more than 470,000 middle-aged people.
Another model, ALADDIN, is based on routine blood tests and was evaluated in a study published earlier this year by an international collective of hepatologists. The study found that ALADDIN performed better than Fib-4 and other risk scores in identifying patients who could benefit most from resmetirom treatment.
AI is a first pass, not a replacement
The source is clear that AI tools are not being presented as a complete substitute for existing liver care. Paul Brennan, a specialty registrar in gastroenterology, hepatology, and internal medicine at the University of Dundee, said these tools will not completely replace biopsies or imaging.
The more realistic promise is triage. AI could help address bottlenecks in primary care, where much fibrosis goes undetected, while also reducing unnecessary referrals to hepatologists. That would make the technology a smarter first pass rather than the final word.
Early detection also matters because fatty liver damage can be reversible. In initial stages, lifestyle changes such as reducing alcohol intake, losing weight through dietary improvements and exercise, and even drinking more coffee have been shown to reverse scarring and inflammation. For people with moderate to advanced liver scarring, treatments including the GLP-1 medication semaglutide and a drug called resmetirom have been shown to be highly effective therapeutics.
So far, the use of AI in liver care has largely stayed within research. Lazarus is optimistic that this will begin to change, and he points to research carried out in Denmark showing that informing people they have liver fibrosis makes them more likely to commit to dietary and exercise regimes. If AI can help identify those people sooner, it could shift fatty liver disease care toward prevention instead of late-stage crisis management.