RARE Daily

AI Tapped to Identify People Who May Have Rare Liver Disease

July 10, 2024

Rare Daily Staff

Acute hepatic porphyria (AHP) is a group of rare genetic diseases that mostly affect women with and can be easily dismissed as menstrual cramps but can involve life-threating attacks with symptoms that include abdominal pain, nausea and vomiting, limb weakness, and anxiety.

Doctors at the University of California, Los Angeles’ David Geffen School of Medicine and the UC Health Network have devised a predictive algorithm that uses artificial intelligence to scour electronic health records to identify patients who may be at risk of having the condition and should undergo testing.

It can take 10 to 15 years to get correctly diagnosed with AHP because physicians rarely, if ever, see a patient with the condition. During that time, the disease can progress and do irreparable damage, said Kat Schmolly, a physician who was a medical student at UCLA.

Schmolly worked with Vivek Rudrapatna, director of The Real-World Evidence Lab, which applies data science techniques to electronic health records. The two co-invented Project Zebra’s predictive algorithm to analyze healthcare records and identify suspected porphyria patients. The work started at the behest of Alnylam Pharmaceutical, which won U.S. Food and Drug Administration approval in 2019 for givosiran as a prophylactic treatment for recurrent AHP attacks.

In a study published in the Journal of the American Medical Informatics Association, the researchers found the algorithm predicted patients would be referred for AHP testing by a range of 89 to 93 percent accuracy. And when it came to predicting who tested positive for the disease, the algorithm recognized 71 percent of patients earlier than their actual diagnosis, corresponding to an average time saved of 1.2 years.

The AHP case is the latest example of the application of AI to electronic health records to identify people who might be suffering from a rare disease. Unlike physicians who are unable to maintain a working knowledge of the more than 10,000 rare diseases known to exists and who are trained to think in terms of common ailments, AI offers the promise of shortening the diagnostic odyssey by eliminating the long delays many patients face to get critical tests performed.

Physicians will use the app in Winter 2024 when zebraMD will test its algorithms in the real world by embedding it into electronic health records systems at Ronald Reagan Medical Center at UCLA, Olive-View UCLA Medical Center, UCSF and Dartmouth Health.

Read the UC Health article here or the study here.

Stay Connected

Sign up for updates straight to your inbox.