Qventus Uses AI and ML to Accelerate Patient Flow by Automating Healthcare Operations on AWS

During the COVID-19 pandemic, hospital leaders need a way to plan for related demand and resource usage (e.g., intensive care unit beds and ventilators), create virtual capacity, and manage elective surgery volumes at a hospital level. Qventus Inc. (Qventus) has used Amazon Web Services (AWS) tools to provide models and planning solutions to address these needs.

Qventus’s patient flow automation solution uses real-time data, artificial intelligence (AI), and machine learning (ML) to help health systems and hospitals improve their operations. In the context of the COVID-19 pandemic, Qventus made planning models available for free. “We have created a set of 450 localized epidemiological models that enable health system leaders to predict hospitalizations—as well as the impact of COVID-19 on medical-surgical, intensive care unit, ventilator, and personal protective equipment resources—and to tailor assumptions and create scenarios around social distancing policies, mask compliance, doubling times, asymptomatic rates, and more,” says Mudit Garg, cofounder and CEO of Qventus. By applying Qventus’s platform during the COVID-19 pandemic, health systems and hospitals can gain visibility into and better manage critical resources, balance COVID-19 patients with elective surgery patients, and create new effective capacity through the use of artificial intelligence for earlier identification of patients ready to step down from the intensive care unit or discharge from medical-surgical units.

By using AWS services, Qventus is able to develop and maintain mature systems more efficiently. For example, by using Amazon Relational Database Service (Amazon RDS), “the team has spent less time on database maintenance, which frees up resources to focus on building new products for end users,” says Garg. Similarly, Amazon SageMaker has enabled Qventus to dynamically scale its ML training, which would otherwise be significantly costlier and more time intensive.