Sign In

Communications of the ACM

ACM TechNews

One-Stop Machine Learning Platform Turns Healthcare Data Into Insights

View as: Print Mobile App Share:
Cardea is built to help hospitals plan for events as large as global pandemics and as small as no-show appointments.

The Cardea software system aims to bring the power of prediction to hospitals.

Credit: Arash Akhgari

Software developed by researchers at the Massachusetts Institute of Technology (MIT) uses machine learning to transform healthcare data into insights that can inform decision-making.

The open-source software system, called Cardea, works with Fast Healthcare Interoperability Resources (FHIR), the current industry standard for electronic healthcare records.

MIT's Kalyan Veeramanchaneni said the software can "adapt to different conditions and different datasets seamlessly," which is important because hospital use of FHIR varies.

After gathering information from the user, Cardea asks what they want to learn, allowing them to choose among different models, then uses the dataset and models to learn patterns and make predictions.

It also performs model audits to help users determine whether to accept the results or begin again with a new model.

MIT's Sarah Alnegheimish said the goal is for Cardea to ultimately "solve any prediction problem within the healthcare domain."

From MIT News
View Full Article


Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA


No entries found