Bloodstream infections (BSIs) can quickly become fatal, especially for immunocompromised patients. A new study from Houston Methodist Research Institute finds that artificial intelligence can help clinicians identify infection patterns in patients they’ve never seen before.
The study, led by Masayuki Nigo, MD, Ph.D., associate professor of medicine at Houston Methodist University, was published in the American Journal of Transplantation and used an unsupervised machine learning model to identify three major clusters of BSI patients. Researchers analyzed data from more than 15,000 patients and formed cluster features based on clinical data obtained within the first 48 hours after BSI diagnosis.
“This study is important because it shows that patients with bloodstream infections, including solid organ transplant recipients, are not clinically homogeneous despite sharing the same diagnosis,” Nigo said. “Using data routinely collected within the first 48 hours of infection, we identified three distinct clinical patterns based on patient characteristics, disease severity, and need for organ support through a machine learning-based clustering approach.”
The highest-risk groups included older, predominantly male patients who required more ventilator and vasopressor support, and transplant patients. These people are especially susceptible to infections because their immune systems are weakened, and one in 10 will experience an infection within a year after transplantation. Mortality rates can be as high as 60%.
In contrast, the other two patient groups shared similar clinical profiles with only minor differences in initial features. However, the outcomes were very different, with one group showing patients with milder symptoms, and the other group with more severe symptoms and higher patient mortality.
“Our model transforms routine initial data into risk maps that clinicians can immediately use,” said Dr. Stefano Casarin, assistant professor in the Center for Precision Surgery at Houston Methodist Research Institute. “This gives us new ways to understand and predict how sick patients will get. The sooner we can identify high-risk patients, the sooner we can act.”
Nigo said next steps include further research to validate the findings in external health systems to ensure reproducibility and determine how the methodology can be improved to promote better clinical decision-making and patient outcomes.
Other collaborators on the study are Houston Methodist’s Max Adelman, James Kurian, Jiakyung Xu, David Su, Arjab Sanghvi, Stephen Jones, Ashton Connor, Ahmed Gaber, Mark Ghobrial, and Cesar Arias.
This research was supported by funding from the National Institutes of Health’s National Institute of Allergy and Infectious Diseases and Houston Methodist Academic Research Institute.
/Open to the public. This material from the original organization/author may be of a contemporary nature and has been edited for clarity, style, and length. Mirage.News does not take any institutional position or position, and all views, positions, and conclusions expressed herein are those of the authors alone. Read the full text here.
