# Stochastic Siamese MAE Pretraining for Longitudinal Medical Images Source: [Stochastic Siamese MAE Pretraining for Longitudinal Medical Images]() Collection index: [llms.txt]() Data updated: 2025-12-31T05:00:09.929Z Authors: Taha Emre, Arunava Chakravarty, Thomas Pinetz, Dmitrii Lachinov, Martin J. Menten, Hendrik Scholl, Sobha Sivaprasad, Daniel Rueckert, Andrew Lotery, Stefan Sacu, Ursula Schmidt-Erfurth, Hrvoje Bogunovi\\'c Published: 2025-12-29T13:00:12.000Z ## Abstract Temporally aware image representations are crucial for capturing disease progression in 3D volumes of longitudinal medical datasets. However, recent state-of-the-art self-supervised learning approaches like Masked Autoencoding (MAE), despite their strong representation learning capabilities, lack temporal awareness. In this paper, we propose STAMP (Stochastic Temporal Autoencoder with Masked Pretraining), a Siamese MAE framework that encodes temporal information through a stochastic process by conditioning on the time difference between the 2 input volumes. Unlike deterministic Siamese approaches, which compare scans from different time points but fail to account for the inherent uncertainty in disease evolution, STAMP learns temporal dynamics stochastically by reframing the MAE reconstruction loss as a conditional variational inference objective. We evaluated STAMP on two OCT and one MRI datasets with multiple visits per patient. STAMP pretrained ViT models outperformed both existing temporal MAE methods and foundation models on different late stage Age-Related Macular Degeneration and Alzheimer's Disease progression prediction which require models to learn the underlying non-deterministic temporal dynamics of the diseases. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [AD conversion prediction on ADNI (1-year window)](): AD conversion prediction; ADNI (1-year window); AUROC: 0.812. - [Geographic Atrophy (GA) conversion prediction on PINNACLE](): Geographic Atrophy (GA) conversion prediction; PINNACLE; AUROC: 0.848. - [AD Conversion Prediction on ADNI 3-years window](): AD conversion prediction; ADNI 3-years window; AUROC: 77.3. - [Wet-AMD Conversion Prediction on HARBOR 6-month window (test)](): wet-AMD conversion prediction; HARBOR 6-month window (test); AUROC: 0.7. - [wet-AMD conversion prediction on HARBOR 12-month window (test)](): wet-AMD conversion prediction; HARBOR 12-month window (test); AUROC: 0.671. ## Original sources and implementation [Original paper]() [PDF]()