Big Data 4 Small Babies

A Data-Driven Predictive Algorithm Reduces the Unnecessary Antibiotics Given to Preterm Babies at NICU

Finaps designed and built a data-driven algorithm to help doctors decide whether the administration of antibiotics to preterm infants is necessary or not. The algorithm significantly reduces the unnecessary administration of antibiotics by identifying unjustified cases of sepsis. As a result, these infants have a better chance of naturally fighting off sepsis because the ‘good’ bacteria in their bodies are not killed by the antibiotics. This results in babies returning to their parents sooner and greatly improves their long-term outcome.

The Challenge

Babies with the age less than 32 weeks have a high probability of obtaining sepsis during their stay at the Neonatal Intensive Care Unit (NICU). UMC Utrecht launched the ‘Applied Data Analytics in Medicine’ (ADAM) to explore the opportunities for data analytics in healthcare. The NICU of the Wilhelmina Kinder Ziekenhuis needs an algorithm that validates suspicions of sepsis and predicts the type of bacteria that will most likely be found in the blood. Big Data has been used as a solution to proactively treat and even prevent sepsis.

Big Data Solution

Multiple databases, such as the Research Data Platform, MetaVision (real-time data), UPOD (cell data) and HiX (patient information), are combined to construct a predictive algorithm. SAS Enterprise Guide and SAS Enterprise Miner is used to connect the databases and to construct the algorithm. This algorithm helps to predict whether the diagnosis of sepsis is justified and if so, which type of bacteria is likely to be present. In addition, a detailed overview of the variables, values and predictions is provided to the Doctors. This gives Doctors solid decision support for the treatment of sepsis.

Team

2 Data Scientists

Technology

SAS, Python

Time

3 - 4 months

Interested? Come have a cup of coffee with us!

 
Tim Pijl

Tim Pijl

Data Engineer at Finaps