Mayo Clinic researchers have developed AI tools that predict severe asthma risks in children as young as three. By analyzing health records, these tools aim to catch high-risk cases early, potentially transforming pediatric asthma care.
In a groundbreaking development, researchers at the Mayo Clinic have created artificial intelligence tools capable of identifying children with asthma who are at the highest risk of severe asthma exacerbations and acute respiratory infections.
This innovative approach, detailed in a recent study published in the Journal of Allergy and Clinical Immunology, can detect these risks in children as young as three years old.
Childhood Asthma’s Toll
Asthma affects nearly 6 million children in the United States and is one of the leading causes of missed school days, emergency room visits and hospital stays, according to the Centers for Disease Control and Prevention (CDC).
Respiratory infections are a common trigger for asthma attacks, but symptoms can vary greatly and change over time, making it difficult for clinicians to pinpoint which children are most susceptible.
The Mayo Clinic’s AI tools aim to close this gap.
“This study takes us a step closer to precision medicine in childhood asthma, where care shifts from reactive care for advanced severe asthma to prevention and early detection of high-risk patients,” senior author Young Juhn, a professor of pediatrics at Mayo Clinic, said in a news release.
Juhn oversees several prominent research programs at the Mayo Clinic, including the AI Program of Mayo Clinic Children’s and the Precision Population Science Lab.
Using AI to Detect Asthma Early
For the study, the researchers analyzed electronic health records from over 22,000 children born between 1997 and 2016 in southeastern Minnesota.
To handle this vast array of data, they developed multiple AI tools using machine learning and natural language processing to extract critical information from doctors’ notes.
These tools were able to capture crucial data points such as symptoms and family history, allowing researchers to apply two widely recognized diagnostic checklists for assessing asthma in young children: the Predetermined Asthma Criteria and the Asthma Predictive Index.
Children who met criteria on both checklists were identified as a distinct subgroup with a higher risk for serious complications.
Detecting Asthma Risk by Age 3
By analyzing this high-risk subgroup, the researchers found stark differences compared to other children in the study.
By age three, children in the high-risk subgroup experienced pneumonia more than twice as often and influenza nearly three times as frequently.
They also exhibited the highest rates of asthma attacks requiring steroids, emergency visits or hospitalization, and were more prone to respiratory syncytial virus (RSV) infection during their first three years.
Additionally, these children were more likely to have a family history of asthma, eczema, allergic rhinitis or food allergies.
Laboratory tests indicated signs of allergic inflammation, including elevated eosinophil counts, allergen-specific IgE and periostin levels, which are markers of type 2 inflammation, coupled with impaired lung function.
These findings point to a high-risk asthma subtype that predisposes some children to more severe respiratory issues.
What’s Next?
The research team plans to extend testing of these AI tools in broader clinical settings, encompassing more diverse populations and healthcare systems. Their goal is to integrate biological data to refine the definitions and treatments of asthma subtypes for early intervention.
The team is also preparing to investigate a compound designed to mitigate overactive immune responses associated with asthma. Utilizing lab-grown cell models known as organoids, they aim to advance early detection and prevention strategies for childhood asthma on a larger scale.
Source: Mayo Clinic