There’s a Fatty Liver Epidemic. AI Could Help Get Ahead of It
| Source: Wired AI
Tags: AI in healthcare, fatty liver disease, EHR, semaglutide, medical AI, Fib-4, drug discovery
With fatty liver disease affecting 30% of adults globally and three-quarters of cirrhosis cases diagnosed only at life-threatening stages, researchers are using AI to scan electronic health records and prioritize patients for early detection — opening the window when lifestyle changes and GLP-1 medications can fully reverse liver scarring.
Details
Fatty liver disease — where fat exceeds 5-10% of liver weight — now affects approximately 30% of adults worldwide and typically develops without symptoms. The result is that three-quarters of cirrhosis cases are only discovered when the condition is life-threatening, well past the point where intervention is most effective.\n\nThe reversibility window is the key clinical insight: caught early, lifestyle changes (reduced alcohol, weight loss, exercise, more coffee), along with GLP-1 medication semaglutide and a drug called resmetirom, can fully reverse scarring and inflammation. The liver is uniquely capable of self-repair if intervention happens in time.\n\nResearchers including CUNY professor Jeffrey Lazarus are investigating AI tools that comb through electronic health records to identify high-risk patients using data already available — no new tests required. The Fib-4 index, which computes a 0-6 liver risk score from age, two enzyme levels, and blood-clotting ability, is simple, noninvasive, and validated, but rarely applied even to high-risk patients with obesity or type 2 diabetes. AI-driven application of Fib-4 at scale could substantially expand the diagnosed population to reach the treatable stage.