Using our mobile app? Be sure to check for any new app updates to receive any enhancements.
Logo

Get Healthy!

AI Might Spot Rare Diseases in Patients Years Earlier
  • Posted May 3, 2024

AI Might Spot Rare Diseases in Patients Years Earlier

Artificial intelligence might be able to identify patients who have rare diseases years earlier than they would typically be diagnosed, a new study says.

A newly developed AI program was able to successfully identify people at risk of developing a rare immune disorder, researchers report in Science Translational Medicine.

Out of a group of 100 people judged at highest risk by the AI program, 74 very likely have the disorder, researchers found.

This shows that AI could potentially improve the outcomes of these folks by getting them earlier treatment, they said.

"Patients who have rare diseases may face prolonged delays in diagnosis and treatment, resulting in unnecessary testing, progressive illness, psychological stresses, and financial burdens,"senior researcher Dr. Manish Butte said in a news release.

"Using (AI) tools, we developed an approach to speed the diagnosis of undiagnosed patients by identifying patterns in their electronic health records that resemble those of patients who are known to have the disorders," said Butte, a professor in pediatrics, human genetics and microbiology/immunology at the University of California, Los Angeles.

Researchers focused on a collection of disorders called common variable immunodeficiency, or CVID. These disorders often elude diagnosis for years or decades.

CVID disorders are estimated to affect about 1 in 25,000 people, and typically cause antibody deficiencies and impaired immune responses, researchers said.

"For every year a diagnosis is delayed, there is an increase in infections, antibiotic use, emergency room visits, hospitalizations, and missed days of work and school,"lead researcher Ruth Johnson, a fellow in biomedical informatics at Harvard Medical School, said in a news release.

Not only are CVID disorders rare, but symptoms can vary greatly between patients and often overlap with those of more common illnesses, researchers said.

"The clinical presentation of rare immune phenotypes such as CVID intersects with many medical specialties,"Butte said. "Patients may be seen in ear, nose and throat clinics for sinus infections. They may be treated in pulmonology clinics for pneumonias. This fragmentation of care across multiple specialists leads to long delays in diagnosis and treatment."

Further, CVID disorders are often driven by changes in only one gene out of more than 60 genes that have been linked to them. This rules out the possibility of a genetic test to provide a definitive diagnosis, researchers said.

For this study, researchers developed an AI called PheNet. The name refers to the word "phenotypes,"which is the medical term for the traits of a disease that can be seen in patients.

PheNet learns the phenotype patterns from verified CVID cases, and then uses this knowledge to rank an individual's risk of having the disorder, researchers said.

PheNet reviewed millions of UCLA electronic patient records, and ranked all of the patients for their risk of CVID based on what it had learned.

About 74% of the 100 patients that PheNet ranked as highest risk for CVID were deemed probable to have one of the disorders, based on doctors' follow-up review, results show.

Based on those results, the research team has received $4 million in funding from the National Institutes of Health to further study the AI program in real-world settings.

"We show that artificial intelligence algorithms such as PheNet can offer clinical benefits by expediting the diagnosis of CVID, and we expect this to apply to other rare diseases, as well,"said senior researcher Bogdan Pasaniuc. He's a professor of computational medicine, human genetics, and pathology and laboratory medicine at UCLA David Geffen School of Medicine.

"Our implementation across all five University of California medical centers is already making an impact,"he said in a news release. "We are now improving the precision of our approach to better identify CVID while expanding to other diseases. We will also plan to teach the system to read medical notes to glean even more information about patients and their illnesses."

More information

The Mayo Clinic has more about AI in health care.

SOURCE: UCLA, news release, May 1, 2024

HealthDay
Health News is provided as a service to The Medicine Shoppe #503 site users by HealthDay. The Medicine Shoppe #503 nor its employees, agents, or contractors, review, control, or take responsibility for the content of these articles. Please seek medical advice directly from your pharmacist or physician.
Copyright © 2024 HealthDay All Rights Reserved.