Predictive models are gaining ground in healthcare as more and more hospitals and specialists use them for the diagnosis and treatment of cancer and other diseases. But these machine learning tools are still not as accurate or powerful as they could be, and that often comes down to not having enough quality clinical data to train on.
One way to help address the fact that many sample sizes are too small is to aggregate data from other sources, says Steve Irvine, founder and CEO of integra.ai. That can be done, while protecting patient privacy, with federated learning techniques, which can open up vast new treasure troves of data for researchers. Explain more in this episode of HIMSSCast.
Do you like what you hear? Subscribe to the podcast at Apple Podcasts, Spotify either Google Play!
Talking Points:
-
How machine learning models for oncology have evolved in recent years
-
The keys to building a good predictive algorithm
-
Why is it so hard to find enough quality data to train models?
-
What is federated learning and how can it help?
-
Opportunities and challenges of adopting a federated learning approach
-
How Irvine sees the evolution of predictive oncology models in the coming years
More about this episode:
Top 10 Artificial Intelligence and Machine Learning Stories of 2022
HIMSSCast: 2023 forecast: 5G, AI command centers, hybrid work models and more
Demystifying the role of AI in healthcare to reassure new providers and former practitioners
AI is rapidly addressing data requirements and advancing interoperability, says expert
Oncology practice uses AI to significantly improve end-of-life care
How a healthcare system developed AI to address patient safety
Orlando Health to launch AI-powered hospital-at-home services
