# Learning Temporally Equivariance for Degenerative Disease Progression in OCT by Predicting Future Representations

arXiv: https://arxiv.org/abs/2405.09404 · HTML: https://mlsanity.com/paper/2405.09404/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 28 references that are in the map.
- First posted on arXiv 15 May 2024.

## Builds on (most cited first, 19 of 19)

- [Barlow Twins: Self-Supervised Learning via Redundancy Reduction](https://mlsanity.com/paper/2103.03230/) (2021)
- [An Empirical Study of Training Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.02057/) (2021)
- [Unsupervised Representation Learning by Predicting Image Rotations](https://mlsanity.com/paper/1803.07728/) (2018)
- [Group Equivariant Convolutional Networks](https://mlsanity.com/paper/1602.07576/) (2016)
- [VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learning](https://mlsanity.com/paper/2105.04906/) (2021)
- [Rotation Equivariant CNNs for Digital Pathology](https://mlsanity.com/paper/1806.03962/) (2018)
- [Understanding Dimensional Collapse in Contrastive Self-supervised Learning](https://mlsanity.com/paper/2110.09348/) (2021)
- [Big Self-Supervised Models Advance Medical Image Classification](https://mlsanity.com/paper/2101.05224/) (2021)
- [What Should Not Be Contrastive in Contrastive Learning](https://mlsanity.com/paper/2008.05659/) (2020)
- [Improving Transferability of Representations via Augmentation-Aware Self-Supervision](https://mlsanity.com/paper/2111.09613/) (2021)
- [Time-Equivariant Contrastive Video Representation Learning](https://mlsanity.com/paper/2112.03624/) (2021)
- [Self-supervised learning of Split Invariant Equivariant representations](https://mlsanity.com/paper/2302.10283/) (2023)
- [Longitudinal Self-Supervised Learning](https://mlsanity.com/paper/2006.06930/) (2020)
- [Slow and steady feature analysis: higher order temporal coherence in video](https://mlsanity.com/paper/1506.04714/) (2015)
- [Modeling Disease Progression In Retinal OCTs With Longitudinal Self-Supervised Learning](https://mlsanity.com/paper/1910.09420/) (2019)
- [EquiMod: An Equivariance Module to Improve Self-Supervised Learning](https://mlsanity.com/paper/2211.01244/) (2022)
- [Metadata-enhanced contrastive learning from retinal optical coherence tomography images](https://mlsanity.com/paper/2208.02529/) (2022)
- [TINC: Temporally Informed Non-Contrastive Learning for Disease Progression Modeling in Retinal OCT Volumes](https://mlsanity.com/paper/2206.15282/) (2022)
- [Learning to Compare Longitudinal Images](https://mlsanity.com/paper/2304.02531/) (2023)

## Built on by (most cited first, 1 of 1)

- [Stochastic Siamese MAE Pretraining for Longitudinal Medical Images](https://mlsanity.com/paper/2512.23441/) (2025)

## Abstract (arXiv)

Contrastive pretraining provides robust representations by ensuring their invariance to different image transformations while simultaneously preventing representational collapse. Equivariant contrastive learning, on the other hand, provides representations sensitive to specific image transformations while remaining invariant to others. By introducing equivariance to time-induced transformations, such as disease-related anatomical changes in longitudinal imaging, the model can effectively capture such changes in the representation space. In this work, we propose a Time-equivariant Contrastive Learning (TC) method. First, an encoder embeds two unlabeled scans from different time points of the same patient into the representation space. Next, a temporal equivariance module is trained to predict the representation of a later visit based on the representation from one of the previous visits and the corresponding time interval with a novel regularization loss term while preserving the invariance property to irrelevant image transformations. On a large longitudinal dataset, our model clearly outperforms existing equivariant contrastive methods in predicting progression from intermediate age-related macular degeneration (AMD) to advanced wet-AMD within a specified time-window.