# Exploiting Intermediate Reconstructions in Optical Coherence Tomography for Test-Time Adaption of Medical Image Segmentation

arXiv: https://arxiv.org/abs/2603.05041 · HTML: https://mlsanity.com/paper/2603.05041/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 34 references that are in the map.
- First posted on arXiv 5 Mar 2026.

## Builds on (most cited first, 21 of 21)

- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [Elucidating the Design Space of Diffusion-Based Generative Models](https://mlsanity.com/paper/2206.00364/) (2022)
- [Diffusion Posterior Sampling for General Noisy Inverse Problems](https://mlsanity.com/paper/2209.14687/) (2022)
- [Albumentations: fast and flexible image augmentations](https://mlsanity.com/paper/1809.06839/) (2018)
- [Continual Test-Time Domain Adaptation](https://mlsanity.com/paper/2203.13591/) (2022)
- [fastMRI: An Open Dataset and Benchmarks for Accelerated MRI](https://mlsanity.com/paper/1811.08839/) (2018)
- [Learning a Variational Network for Reconstruction of Accelerated MRI Data](https://mlsanity.com/paper/1704.00447/) (2017)
- [Practical Blind Image Denoising via Swin-Conv-UNet and Data Synthesis](https://mlsanity.com/paper/2203.13278/) (2022)
- [Back to the Source: Diffusion-Driven Test-Time Adaptation](https://mlsanity.com/paper/2207.03442/) (2022)
- [Denoising Diffusion Gamma Models](https://mlsanity.com/paper/2110.05948/) (2021)
- [Flow Matching for Medical Image Synthesis: Bridging the Gap Between Speed and Quality](https://mlsanity.com/paper/2503.00266/) (2025)
- [Stable Deep MRI Reconstruction using Generative Priors](https://mlsanity.com/paper/2210.13834/) (2022)
- [CUNSB-RFIE: Context-aware Unpaired Neural Schr"odinger Bridge in Retinal Fundus Image Enhancement](https://mlsanity.com/paper/2409.10966/) (2024)
- [SD-LayerNet: Semi-supervised retinal layer segmentation in OCT using disentangled representation with anatomical priors](https://mlsanity.com/paper/2207.00458/) (2022)
- [Noise transfer for unsupervised domain adaptation of retinal OCT images](https://mlsanity.com/paper/2209.08097/) (2022)
- [Shared Prior Learning of Energy-Based Models for Image Reconstruction](https://mlsanity.com/paper/2011.06539/) (2020)
- [Joint semi-supervised and contrastive learning enables domain generalization and multi-domain segmentation](https://mlsanity.com/paper/2405.05336/) (2024)
- [Simulation of Arbitrary Level Contrast Dose in MRI Using an Iterative Global Transformer Model](https://mlsanity.com/paper/2307.11980/) (2023)
- [MOST: MR reconstruction Optimization for multiple downStream Tasks via continual learning](https://mlsanity.com/paper/2409.10394/) (2024)
- [GARD: Gamma-based Anatomical Restoration and Denoising for Retinal OCT](https://mlsanity.com/paper/2509.10341/) (2025)

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

- [Test-Time Adaptation in Optical Coherence Tomography Using Trajectory-Aligned Time-Independent Flow](https://mlsanity.com/paper/2606.18876/) (2026)

## Abstract (arXiv)

Primary health care frequently relies on low-cost imaging devices, which are commonly used for screening purposes. To ensure accurate diagnosis, these systems depend on advanced reconstruction algorithms designed to approximate the performance of high-quality counterparts. Such algorithms typically employ iterative reconstruction methods that incorporate domain-specific prior knowledge. However, downstream task performance is generally assessed using only the final reconstructed image, thereby disregarding the informative intermediate representations generated throughout the reconstruction process. In this work, we propose IRTTA to exploit these intermediate representations at test-time by adapting the normalization-layer parameters of a frozen downstream network via a modulator network that conditions on the current reconstruction timescale. The modulator network is learned during test-time using an averaged entropy loss across all individual timesteps. Variation among the timestep-wise segmentations additionally provides uncertainty estimates at no extra cost. This approach enhances segmentation performance and enables semantically meaningful uncertainty estimation, all without modifying either the reconstruction process or the downstream model.