A Calibrated Memorization Index (MI) for Detecting Training Data Leakage in Generative MRI Models
This work addresses privacy risks in medical image generation by providing a robust detection tool for data leakage, though it is incremental as it builds on existing memorization detection methods.
The paper tackles the problem of detecting training data duplication in generative MRI models to address privacy concerns, proposing a calibrated metric that achieves near-perfect detection of duplicates across datasets.
Image generative models are known to duplicate images from the training data as part of their outputs, which can lead to privacy concerns when used for medical image generation. We propose a calibrated per-sample metric for detecting memorization and duplication of training data. Our metric uses image features extracted using an MRI foundation model, aggregates multi-layer whitened nearest-neighbor similarities, and maps them to a bounded \emph{Overfit/Novelty Index} (ONI) and \emph{Memorization Index} (MI) scores. Across three MRI datasets with controlled duplication percentages and typical image augmentations, our metric robustly detects duplication and provides more consistent metric values across datasets. At the sample level, our metric achieves near-perfect detection of duplicates.