Machine learning helps physicians see which patients will respond to asthma treatment

With help from clinical variables identified using machine learning techniques, physicians can predict which of their severe asthma patients may benefit from treatment with systemic corticosteroids, a study published in the American Journal of Respiratory and Critical Care Medicine found.

"Systemic corticosteroids are the most effective therapy we have for asthma, but not all patients respond in the same way," said researcher Wei Wu, PhD. "Unfortunately, when clinicians don't see a big improvement after initial treatment, they might give patients even higher doses. If a patient is one of those who can't be helped by corticosteroids, the higher dose just means worse side effects."

To better understand how different subgroups of patients respond to systemic corticosteroid therapy, the researchers, from Pittsburgh-based Carnegie Mellon University, used a machine learning algorithm to sort through 100 variables for 346 adult patients in the Severe Asthma Research Program.

The algorithm, developed by Dr. Wu and Seojin Bang, a PhD student in Carnegie Mellon's computational biology department, recognizes patterns in massive volumes of complex clinical data. The algorithm put patients into four subgroups, including two for severe asthmatics — one that responded to systemic corticosteroids and one that did not.

Of the original 100 variables, the researchers found 12 — including age of onset, weight, race and scores on a quality-of-life questionnaire — that could correctly categorize patients with high confidence for responding to the therapy if processed by a computer app.

They then used the 12 variables to categorize a group of 182 participants not included in the original analysis and found the variables proved effective in successfully categorizing these patients.

"We believe we've made progress toward making precision medicine a reality," Dr. Wu said. "Five years ago, we were only able to categorize patients clinically. Now, using incredibly complex data, we're able to predict how these subgroups will respond to a critical drug treatment."

More articles on clinical leadership and infection control:
Why CRISPR may work when vaccines don't
Controversial study on genetically modified mosquitoes for malaria enters next phase
Text message system may improve patient outcomes after surgery, study finds

© Copyright ASC COMMUNICATIONS 2020. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.

 

Featured Webinars

Featured Whitepapers