Machine learning can help avert infants’ discomfort, overtreatment
An AI algorithm could have kept almost 850 of 1,240 fevered infants—close to 70%—from having to undergo a painful lumbar puncture, aka “spinal tap,” for finding out if their high temps owed to a serious bacterial infection.
The babies in the study behind the numbers were 60 days old or younger and treated at more than one hospital. Senior author Elizabeth Alpern, MD, of Northwestern University and Lurie Children’s Hospital in Chicago and colleagues had their work published in Pediatrics and covered by the hospital’s news division.
The team developed risk stratification models in four flavors of supervised machine learning—logistic regression, random forest, support vector machine and a single-hidden-layer neural network.
Internal validation using lab results for ground truth showed the random forest technique achieved the best specificity (75%) as well as the highest sensitivity (99%).
The random forest model misclassified just one case of bacteremia. (Other types of serious bacterial infection include bacterial meningitis and urinary tract infection.)
“Although computationally complex, lacking parameter cutoffs and in need of external validation, this strategy may allow for reductions in unnecessary procedures, hospitalizations and antibiotics while maintaining excellent sensitivity,” the authors conclude.
In the hospital news item, lead author Sriram Ramgopal, MD, says quickly figuring out which febrile infants are at high risk for a serious bacterial infection is especially critical in his clinical domain, the emergency department.
The random forest risk-assessment model impressed the researchers by “surpassing the predictive capability of the current decision rules we use,” he adds.
“Our results are very promising and may pave the way to an eventual use of this type of artificial intelligence clinically,” Ramgopal says.
Dave P. has worked in journalism, marketing and public relations for more than 30 years, frequently concentrating on hospitals, healthcare technology and Catholic communications. He has also specialized in fundraising communications, ghostwriting for CEOs of local, national and global charities, nonprofits and foundations.