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

2026-03-05

AI Summary

This work addresses the challenge of medical image segmentation under low-quality imaging conditions prevalent in primary care settings. While existing methods rely solely on the final reconstructed image, they overlook the semantically rich intermediate representations generated during the reconstruction process. To bridge this gap, the authors propose IRTTA, a test-time adaptation framework that dynamically modulates the parameters of normalization layers in a frozen segmentation model by leveraging these intermediate features—without altering the reconstruction pipeline or retraining the downstream model. IRTTA is the first method to enable test-time adaptive segmentation based on reconstruction intermediates and introduces an average entropy loss to refine uncertainty estimation. Experiments demonstrate that IRTTA significantly improves segmentation accuracy while providing semantically consistent and reliable uncertainty quantification, all without additional computational overhead.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Uncertainty Representations

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

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.

Problem

• Optical Coherence Tomography
• Medical Image Segmentation
• Intermediate Reconstructions
• Test-Time Adaptation
• Uncertainty Estimation

Innovation

• test-time adaptation
• intermediate reconstructions
• optical coherence tomography
• uncertainty estimation
• modulator network