New AI Model for COVID-19 Patients in the ED Using a Federated Learning Approach

We’ve all known that data and AI was going to come together to make healthcare better.  We’ve seen it in fits and starts for a long time.  We were just waiting for people to really connect all the dots and make this a reality.  In my most recent interview, I saw the best example of all those dots coming together in a really exciting way.

In this interview, I sat down with the following experts:

  • Anthony Costa, Assistant Professor of Neurosurgery, Director, Sinai BioDesign at Mount Sinai
  • Fiona Gilbert, Professor of Radiology at University of Cambridge
  • Mona Flores, Global Head of Medical AI at NVIDIA

We discuss the work NVIDIA has been doing to address triaging COVID-19 patients using AI models that predict a patient’s oxygen needs.  The AI model can better help a hospital ensure a patient is moved to the right location in the hospital and avoid later transfers to the ICU which we know can cost healthcare organizations a lot of time and hassle.  Plus, it can help ensure a patient is where they need to be to get the best care possible.

What makes this interview and AI model even more interesting is that NVIDIA brought together 20 healthcare organizations from across the world to be able to ensure the learning happened across a wide variety of patients.  This is particularly interesting because many of the hospitals that participated wouldn’t have been able to do this work on their own because they didn’t have enough COVID patients to train the AI model.  Plus, as Anthony Costa comments in the interview, generalizable AI is something we’ve wanted for a long time.

While it’s still early in evaluating this new AI model, the fact that they include learnings from 20 healthcare organizations from across the world, it means that there’s a better chance that the model is something that will work generally across a larger set of patient populations.  This is something that wouldn’t have been possible with just one institutions involvement.

Plus, NVIDIA used a federated model to train the AI engine.  That means that healthcare organizations kept their data and only the learnings were passed back to NVIDIA to be shared with all of the participating organizations.  This made it easy to get approved by the various healthcare organizations that needed to approve the effort.

If you want to learn more about this new AI Model, how they worked to make it a generalizable AI model, and how effective the federated approach was for these healthcare organizations, you’ll enjoy this interview:

Learn more about NVIDIA: https://www.nvidia.com/en-us/industries/healthcare-life-sciences/

Listen and subscribe to the Healthcare IT Today Interviews Podcast to hear all the latest insights from experts in healthcare IT.

And for an exclusive look at our top storiessubscribe to our newsletter.

Tell us what you think. Contact us here or on Twitter at @hcitoday. And if you’re interested in advertising with us, check out our various advertising packages and request our media kit.

About the author

John Lynn

John Lynn

John Lynn is the Founder of HealthcareScene.com, a network of leading Healthcare IT resources. The flagship blog, Healthcare IT Today, contains over 13,000 articles with over half of the articles written by John. These EMR and Healthcare IT related articles have been viewed over 20 million times.

John manages Healthcare IT Central, the leading career Health IT job board. He also organizes the first of its kind conference and community focused on healthcare marketing, Healthcare and IT Marketing Conference, and a healthcare IT conference, EXPO.health, focused on practical healthcare IT innovation. John is an advisor to multiple healthcare IT companies. John is highly involved in social media, and in addition to his blogs can be found on Twitter: @techguy.

Add Comment

Click here to post a comment

   

Categories