Unsupervised Anomaly Detection for Image Dataset Quality Assurance in Multi-Center Breast MRI
This work builds a controlled AD benchmark of 17 realistic QA-relevant anomaly types from six public datasets and proposes a taxonomy of radiological image anomalies based on human visual perception, enabling fine-grained analysis of AD failure modes, establishing a foundation and practical guidance on scalable unsuper...
C. Tappermann, S. Renisch, L. O. Schwen et al.
· 0 citations