COVID-19 Deep Learning Model Validated on ACR AI-LAB

An article in the April issue of the Journal of the American College of Radiology (JACR®) documents how AI-LAB™ was successfully used to test an external COVID-19 Chest Radiograph (CXR) Artificial Intelligence (AI) algorithm on local data by a clinical radiology department with no dedicated data science infrastructure. The experiment shows generalizability of the Pulmonary X-ray Severity (PXS) score model from one practice to another. More importantly, the authors explain how a modest investment of resources and effort can empower a real-world clinical department to participate in data science. 

For AI models to scale and be clinically useful, software tools that facilitate the local testing process, like the freely available AI-LAB provided by the American College of Radiology®(ACR®) to members, will be important to cross the AI implementation gap in health care systems.

This successful test demonstrates AI-LAB’s value for local testing and model validation. For more information, contact Alison Loughlin.