AWS Startups Blog

AWS Startups Healthcare Response Part 1: Predictive Analytics on the Front Lines

Nothing accelerates innovation more than challenges that need to be overcome. Over the last few months, the global healthcare industry has stepped up to the occasion: health systems have been deploying innovative solutions to keep their staff safe and improve patient outcomes, startups have been launching or scaling life-saving technologies, and regulatory bodies such as the Food and Drug Administration (FDA) have been issuing emergency use authorizations to permit adoption of these technology solutions.

Responding to requests from multiple healthcare organizations around the world, we introduced a series of webinars featuring vetted enterprise-ready startup solutions built on top of AWS and focused on helping healthcare providers navigate the challenges surrounding COVID-19. Below is the first in a series of blog posts that shines a spotlight on featured innovative startups. Dive deeper into their solutions, catch up on their presentations via webinar recordings, and get in touch if your healthcare organization needs help. Bringing awareness to these ready-to-deploy technologies may help arm the global healthcare community with the right tools to withstand the challenge.


By Jelena Joffe, AWS Connections, Global Startup Business Development & Corporate Innovation

What if you could predict patient deterioration hours or days before it happened? Or ask a clinical question and get actionable insights in minutes instead of months or years? How many lives could you save?

AWS startup partners CLEW, ClosedLoop.ai, MDMetrix, and AgileMD are leveraging AI to do exactly that. While these companies have been working on their AI solutions for years, the pandemic brought with it the urgency of making more data-driven, timely and proactive decisions, encouraging a wider adoption of these AI/ML tools among healthcare providers, gradually earning the trust of clinicians, and ultimately saving lives.

If you missed the AWS Healthcare Response webinar, check out the highlights of the startup solutions and recordings of each presentation below.

CLEW: Set up a complete command and control room in two leading Israeli hospitals in 2.5 weeks. Converted garage and parking lots to serve as COVID ICU control room bases to increase capacity. Real-time AI-based predictive analytics allowed clinical staff to identify risks early and intervene proactively.

Use case: ICU control room. CLEW presentation.

ClosedLoop.ai: Touched over 10 million lives by deploying a predictive AI model able to identify individuals who have high vulnerability to the complications associated with COVID-19. The free open-source model is also 10 times more precise than the Center for Disease Control’s (CDC) risk assessment.

“We spent 3 years building our data science platform to create a pandemic solution over a weekend,” said Carol McCall, Chief Health Analytics Officer at ClosedLoop.ai.

Use case: at-risk patient outreach. ClosedLoop.ai presentation.

MDMetrix: Instead of waiting 9 to 12 months for a report that provides answers to questions like “is this drug effective for this type of patients?” or “what are the characteristics of patients who require a ventilator?” physicians got actionable insights in a self-service intuitive AI-based platform in minutes, improving patient outcomes and saving lives.

“Physicians are compelled to change behavior and treatment protocols based on data and evidence-based best practices,” according to Dr. Dan Low, Chief Medical Officer at MDMetrix. 

Use case: rapid clinical learning. MDMetrix presentation.

AgileMD: Proved that clinical deterioration is highly predictable and often preventable with the right tools. Their eCART ML tool is backed by a decade of ongoing research at the University of Chicago and captures 99 variables versus the industry MEWS (Modified Early Warning System) benchmark that only considers 5. The ML tool is able to predict sudden patient deterioration hours before patients would need critical care (e.g. cardiac arrest or respiratory failure, a troubling pattern among COVID-19 patients). AgileMD developed a number of free COVID-19 clinical pathways already in use by healthcare providers.

Use case: risk-based clinical decision engine. AgileMD presentation.

 

Let our AWS Healthcare Startup team know if we can help your organization in any way at hcls-startups@amazon.com.