The National Institutes of Health awarded the Applied Pharmacokinetics and Bioinformatics Laboratory at Children’s Hospital Los Angeles $2.8 million over four years to use artificial intelligence to predict dosage and target Targeting the status of each critically ill patient over time and improving clinical treatment.
The lab will build a series of neural networks to predict changes in kidney function in children over time and how that affects their response to drugs.
According to the announcement, by mining the hospital’s huge database of Virtual Pediatric Intensive Care Units (VPICUs), machine learning can unlock patterns in clinical measurements from 20,000 children with seriously ill have been treated at the hospital since 2009.
Variables such as drug amount and clearance in the child’s body can vary from day to day or from time to time.
“Doctors can estimate the dose of medication needed, but that may not necessarily be the right dose for a particular patient. … We create drug-system models in patients to try to. understand how drugs work,” Dr. Michael Neely, professor of pediatricians and clinical scholar at the University of Southern California’s Keck School of Medicine, said in a statement.
Computer modeling of how drugs work in patients can account for individual dosing differences to some extent but is limited when using current or past measurements to predict future dosage.
The researchers will test these algorithms using 5,000 VPICU plasma measurements of the antibiotic vancomycin to measure patient exposure over time.
“We’re trying to predict what the kidney function of an unstable, critically ill child will be like tomorrow, based on what happened today, so we can predict,” Neely said. the child’s medication needs”.
Andrea Fox is the senior editor of Healthcare IT News.
Email: [email protected]
Healthcare IT News is a publication of HIMSS.
