AI Could Help Catch Fatty Liver Disease Before It's Too Late

Fatty liver disease—in which fat exceeds 5 percent or even 10 percent of the organ's total weight—now affects approximately 30 percent of adults worldwide. The condition causes inflammation, cell damage, and scarring known as fibrosis. If left untreated, it can lead to liver failure and is linked to increased risk of cardiovascular disease and various cancers. The central problem is that it typically develops without noticeable symptoms, and three-quarters of people are diagnosed only when their condition has become life-threatening. Even in cases of advanced cirrhosis, detection often comes too late for effective intervention.

Yet the disease is not hopeless if caught early. Lifestyle changes—reducing alcohol intake, losing weight through diet and exercise, and even drinking more coffee—have all been shown capable of reversing scarring and inflammation. For people with moderate to advanced scarring, novel medications such as the GLP-1 drug semaglutide and resmetirom have proven highly effective. As Jeffrey Lazarus, a professor at the CUNY Graduate School of Public Health and Health Policy, notes, "The liver is a very versatile and interesting organ because it can regenerate, fibrosis can be reversed, and you can be completely healthy again."

The frustration among specialists is that simple, noninvasive tools already exist but are rarely used. The Fib-4 index, for example, rates a person's risk of advanced liver fibrosis through a score of between 0 and 6, based on age, liver enzyme levels, and blood-clotting ability. Doctors also have access to the enhanced liver fibrosis test, which measures levels of two proteins involved in scar tissue formation and an enzyme that inhibits scar clearance. Using both tests together improves diagnosis of advanced fibrosis by four-fold in patients with worrying amounts of liver fat. Yet for physicians facing heavy workloads and administrative burden, adding more testing is not sustainable. Jonathan Dranoff, a professor of medicine at Yale University, observes: "You have to have something that can run in the background or it's easy to just hit a button and do it."

This is where AI enters. Lazarus and Dranoff both see a role for AI in automating the calculation of Fib-4 scores from routine blood testing data, making it easier for primary care physicians to identify patients for referral to liver specialists. AI can also work with x-ray images. Last year, scientists at Osaka Metropolitan University in Japan published a study using an AI model to analyze routine chest x-ray scans. While primarily intended to examine lungs and heart, these images also capture parts of the liver. The researchers found their model identified people with fatty liver disease with an accuracy of 82 percent. Lazarus envisions AI algorithms incorporated into all routine x-ray analyses where the liver is scanned. "The AI could pick up cases of excess liver fat, check for other risk factors such as if the person is overweight, has high cholesterol or type 2 diabetes, and then make a recommendation to the doctor," he explains. "It could tell them, 'This might not have been what you were looking for, but this is what was picked up, and you should refer to hepatology or endocrinology who can do further tests.'"

Several AI tools are already being commercialized. The Danish health tech startup Evido has developed LiverPRO, an AI-powered algorithm that assesses a patient's risk of liver fibrosis based on age and nine routine blood-based biomarkers. Now being commercialized in partnership with pharmaceutical company Roche, it has been shown to outperform Fib-4 in predicting risk of serious liver problems in more than 470,000 middle-aged people. Earlier this year, an international collective of hepatologists published results of a study evaluating an AI model called ALADDIN—another machine learning algorithm based on routine blood tests—that showed it performed better than Fib-4 and other risk scores in identifying patients who could benefit most from resmetirom treatment.

Paul Brennan, a specialty registrar in gastroenterology, hepatology, and internal medicine at the University of Dundee, notes that these tools "won't completely replace biopsies or imaging" but "could fix the bottlenecks in primary care where most fibrosis goes undetected." He expects them to be adopted "as a smarter first pass, catching the moderate-risk patients that blunter tools may miss, and reducing unnecessary referrals to hepatologists." For now, AI use in liver care has largely remained confined to research, but Lazarus is optimistic the trend will shift. Research carried out in Denmark found that informing people they have liver fibrosis makes them more likely to commit to dietary and exercise regimes. For health care systems grappling with rising chronic disease burden, identifying and treating patients in earlier stages would save vast amounts of money. Lazarus states: "I would love to say, let's go back through the electronic medical records across the various US health systems and find people before they have cirrhosis. Liver transplants are extraordinarily expensive in any country, but especially the US. So there's a lot of good humane and economic reasons to find people earlier on."