Don't Mind the Gaps: Implicit Neural Representations for Resolution-Agnostic Retinal OCT Analysis

2026-01-05

arXiv.org

AI Summary

This work addresses the challenges posed by anisotropic resolution in retinal OCT volumes—stemming from large inter-B-scan spacing—and the inconsistency and resolution dependency of conventional 2D methods in 3D analysis. To overcome these limitations, the study introduces, for the first time, a general-purpose implicit neural representation (INR) framework for retinal OCT analysis. By leveraging coordinate-driven continuous modeling that integrates en-face multimodal information, the method enables high-fidelity interpolation between B-scans and constructs a population-trained, generalizable neural atlas. This approach facilitates resolution-agnostic, high-quality 3D reconstruction, significantly improving the consistency of retinal layer segmentation and enabling robust, cross-device, and cross-protocol analysis under diverse imaging conditions.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Representation Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

Routine clinical imaging of the retina using optical coherence tomography (OCT) is performed with large slice spacing, resulting in highly anisotropic images and a sparsely scanned retina. Most learning-based methods circumvent the problems arising from the anisotropy by using 2D approaches rather than performing volumetric analyses. These approaches inherently bear the risk of generating inconsistent results for neighboring B-scans. For example, 2D retinal layer segmentations can have irregular surfaces in 3D. Furthermore, the typically used convolutional neural networks are bound to the resolution of the training data, which prevents their usage for images acquired with a different imaging protocol. Implicit neural representations (INRs) have recently emerged as a tool to store voxelized data as a continuous representation. Using coordinates as input, INRs are resolution-agnostic, which allows them to be applied to anisotropic data. In this paper, we propose two frameworks that make use of this characteristic of INRs for dense 3D analyses of retinal OCT volumes. 1) We perform inter-B-scan interpolation by incorporating additional information from en-face modalities, that help retain relevant structures between B-scans. 2) We create a resolution-agnostic retinal atlas that enables general analysis without strict requirements for the data. Both methods leverage generalizable INRs, improving retinal shape representation through population-based training and allowing predictions for unseen cases. Our resolution-independent frameworks facilitate the analysis of OCT images with large B-scan distances, opening up possibilities for the volumetric evaluation of retinal structures and pathologies.

Problem

• optical coherence tomography
• anisotropic imaging
• resolution-agnostic analysis
• retinal segmentation
• volumetric analysis

Innovation

• Implicit Neural Representations
• Resolution-Agnostic
• Retinal OCT
• 3D Interpolation
• Retinal Atlas