Robust Fair Disease Diagnosis in CT Images

ArXi:2604.09710v1 Announce Type: cross Automated diagnosis from chest CT has improved considerably with deep learning, but models trained on skewed datasets tend to perform unevenly across patient graphics. However, the situation is worse than simple graphic bias. In clinical data, class imbalance and group underrepresentation often coincide, creating compound failure modes that neither standard rebalancing nor fairness corrections can fix alone. We