claim
Underrepresentation of specific pathologic scenarios in training data can cause generative AI models to produce synthesized artifacts that do not correspond to actual medical conditions when processing out-of-distribution samples.

Authors

Sources

Referenced by nodes (1)