Semantic Iterative Reconstruction: One-Shot Universal Anomaly Detection
This addresses the inefficiency and lack of generalization in existing methods for multi-domain clinical anomaly detection, offering a scalable solution.
The paper tackles the problem of unsupervised medical anomaly detection with scarce normal training samples by proposing Semantic Iterative Reconstruction (SIR), a framework that enables a single universal model to detect anomalies across nine diverse medical domains using extremely few normal samples, achieving state-of-the-art results in all tested settings.
Unsupervised medical anomaly detection is severely limited by the scarcity of normal training samples. Existing methods typically train dedicated models for each dataset or disease, requiring hundreds of normal images per task and lacking cross-modality generalization. We propose Semantic Iterative Reconstruction (SIR), a framework that enables a single universal model to detect anomalies across diverse medical domains using extremely few normal samples. SIR leverages a pretrained teacher encoder to extract multi-scale deep features and employs a compact up-then-down decoder with multi-loop iterative refinement to enforce robust normality priors in deep feature space. The framework adopts a one-shot universal design: a single model is trained by mixing exactly one normal sample from each of nine heterogeneous datasets, enabling effective anomaly detection on all corresponding test sets without task-specific retraining. Extensive experiments on nine medical benchmarks demonstrate that SIR achieves state-of-the-art under all four settings -- one-shot universal, full-shot universal, one-shot specialized, and full-shot specialized -- consistently outperforming previous methods. SIR offers an efficient and scalable solution for multi-domain clinical anomaly detection. Code is available at https://github.com/jusufzn212427/sir4ad.