Predictive models are finding success in healthcare as more and more hospitals and professionals use them to diagnose and treat cancer and other diseases. But these machine learning tools are still not as accurate or powerful as they could be – and that often results in not having enough quality clinical data to train with.
Steve Irvine, founder and CEO of integration.ai, says one way to help address the fact that many sample sizes are too small is to aggregate data from other sources. That can be done, while protecting patient privacy, with federated learning techniques, which could open up new troves of data for researchers. He explains more in this HIMSSCast episode.
Like what you hear? Subscribe to podcasts on Apple Podcasts, Spotify or Google Play app market!
Communication skill:
-
How machine learning models for cancer have evolved in recent years
-
The key to building a good prediction algorithm
-
Why is it so difficult to find enough quality data to train models?
-
What is federated learning and how can it help?
-
Opportunities and challenges of applying associative learning
-
Irvine sees how cancer models predict how to develop in the coming years
More info about this episode:
Top 10 AI and machine learning stories in 2022
HIMSSCast: 2023 – 5G forecast, AI command center, hybrid working model, etc.
Unraveling AI’s role in healthcare to reassure new providers – and old professionals
AI is rapidly solving data requests and enhancing interoperability, says an expert
Oncology practice uses AI to dramatically improve end-of-life care
How a medical system developed AI to solve patient safety problems
Orlando Health rolls out AI-driven home hospital services
