GE HealthCare has introduced CareIntellect for Operations, an AI-powered SaaS application built on Amazon Web Services that leverages predictive models to identify hospital operational bottlenecks up to 72 hours in advance.
The American health systems The Queen's Health Systems and Duke Health will be the first organizations to test the application, acting as clinical and operational evaluation centers. Their feedback will be used for future improvements to the software. At Duke Health, the integration of the application comes as a continuation of a three-year collaboration with GE HealthCare, while from the side of The Queen's Medical Center, its use is linked to strategic planning for better utilization of the available infrastructure.
The system operates as a cloud-based subscription software, based on the Amazon Web Services infrastructure, and is available both directly from GE HealthCare and through the AWS Marketplace. It uses two main artificial intelligence models to operate. The first, called Pressure Forecast, combines historical data with real-time information from electronic health records and management systems, predicting potential pressures for the next three days. The second model, Estimated Day of Discharge, processes patient data to estimate the likely day of discharge, updating its forecast every hour. Based on this data, the application takes into account parameters such as bed availability, staffing, waiting times and delays in the movement of patients between different care departments, facilitating the prioritization of operational priorities.
The new application leverages expertise from the company's previous Command Center, which is already used in nearly 500 hospitals and healthcare facilities worldwide. GE HealthCare clarified that the cost savings and capacity gains included in the announcement relate to users of that established system and not performance metrics from the new CareIntellect for Operations.
>Stay connected with Hacks.gr
Set us as a preferred source on Google and find us on social media.
Preferred source
↗
end of article
Top comments (0)