Search papers, labs, and topics across Lattice.
This paper introduces a hybrid anomaly detection framework for medical images that combines self-supervised representation learning with a novel Mean Shift Density Enhancement (MSDE) technique. MSDE iteratively refines latent space embeddings by shifting samples towards higher density regions, thereby improving the separation between normal and anomalous samples. Experiments across seven medical imaging datasets demonstrate state-of-the-art anomaly detection performance, with MSDE achieving superior AUC and Average Precision compared to existing methods.
Shifting latent space embeddings towards higher density regions with Mean Shift Density Enhancement unlocks near-perfect anomaly detection in medical images, even with limited training data.
Anomaly detection in medical imaging is essential for identifying rare pathological conditions, particularly when annotated abnormal samples are limited. We propose a hybrid anomaly detection framework that integrates self-supervised representation learning with manifold-based density estimation, a combination that remains largely unexplored in this domain. Medical images are first embedded into a latent feature space using pretrained, potentially domain-specific, backbones. These representations are then refined via Mean Shift Density Enhancement (MSDE), an iterative manifold-shifting procedure that moves samples toward regions of higher likelihood. Anomaly scores are subsequently computed using Gaussian density estimation in a PCA-reduced latent space, where Mahalanobis distance measures deviation from the learned normal distribution. The framework follows a one-class learning paradigm and requires only normal samples for training. Extensive experiments on seven medical imaging datasets demonstrate state-of-the-art performance. MSDE achieves the highest AUC on four datasets and the highest Average Precision on five datasets, including near-perfect performance on brain tumor detection (0.981 AUC/AP). These results underscore the potential of the proposed framework as a scalable clinical decision-support tool for early disease detection, screening in low-label settings, and robust deployment across diverse imaging modalities.
