Raffaele Marchesi
Titles as the sources gave them. In 1 entry, with 0 checked quotations.
In the Base
Every entry they appear in.
When uncertainty quantification fails: structural vulnerabilities and corrections for selective prediction under clinical class imbalance
According to the abstract, Endrizzi, Ragni, Marchesi and co-authors say uncertainty scores used by clinical AI to defer cases to human experts can become a proxy for class membership under severe class imbalance, penalizing minority patients. They test their corrections on synthetic conditions and three neurodegenerative cohorts.