Dr. Paul Yi was featured in ACR Article about Bias in AI

Dr. Paul Yi, the director of the University of Maryland Medical Intelligent Imaging Center, suggests practical steps to mitigate AI bias in healthcare. These include addressing bias at the data level, leveraging novel techniques at the model level, and incorporating real-time surveillance during clinical deployment to monitor for unexpected model performance.

Radiology professionals must recognize the risks posed by data and model drift, and active surveillance can allow early intervention to prevent premature clinical decisions based on erroneous AI results. A deliberate and thoughtful approach is necessary to establish best practices for ethical and effective use of imaging AI in practice.

To read more, refer to the article features in ACR here.