Search papers, labs, and topics across Lattice.
This paper introduces DeepSSIM++, a self-supervised similarity metric designed to detect memorization in medical generative models while preserving anatomical sensitivity. Traditional pixel-based and generic embedding metrics fall short in either accuracy or sensitivity, but DeepSSIM++ leverages multi-scale feature aggregation and anatomy-preserving augmentations to create an efficient embedding space. The method demonstrates significant improvements over state-of-the-art baselines, achieving up to a 46 percentage point increase in Macro F1 score and drastically speeding up similarity computations, making it a valuable tool for ensuring patient privacy in medical AI applications.
DeepSSIM++ reveals that you can achieve a 46-point boost in memorization detection accuracy without sacrificing computational efficiency in medical generative models.
While deep generative models offer new opportunities for medical image synthesis and data sharing, their ability to memorize and reproduce training samples raises serious concerns about patient confidentiality. Detecting such memorization at scale remains challenging: traditional pixel-based metrics are sensitive to generation artifacts, whereas generic embedding-based metrics often lack the anatomical sensitivity required for medical data. To address this challenge, we introduce DeepSSIM++, a self-supervised similarity metric for scalable memorization auditing in medical generative models. By leveraging multi-scale feature aggregation and anatomy-preserving augmentations, DeepSSIM++ learns an embedding space where cosine similarity approximates the Structural Similarity Index (SSIM), eliminating the need for exact pixel-level registration. Compared with state-of-the-art baselines, DeepSSIM++ achieves an average Macro F1 improvement of 33 percentage points under ideal alignment and 46 percentage points under realistic spatial and intensity perturbations. Furthermore, it accelerates large-scale similarity computation by several orders of magnitude compared with analytical SSIM. By combining anatomical sensitivity and computational efficiency, DeepSSIM++ provides an open-source tool for scalable memorization auditing in medical generative AI. Code and data are publicly available at: https://github.com/brAIn-science/DeepSSIM.
