End-to-end reconstruction of OCT optical properties and speckle-reduced structural intensity via physics-based learning

2026-02-02

Computational Optical Imaging and Artificial Intelligence in Biomedical Sciences III

AI Summary

This work addresses the challenge of jointly reconstructing structural images and intrinsic optical properties—such as refractive index, scattering coefficient, and anisotropy—from optical coherence tomography (OCT) data, which is hindered by signal attenuation, speckle noise, and strong parameter coupling. To this end, the authors propose an end-to-end regularized deep learning framework that, for the first time, embeds a physically consistent OCT forward model directly into the network architecture. By generating predicted OCT signals from estimated optical parameters, the method enables joint optimization of quantitative parameter maps and despeckled structural images. Trained on Monte Carlo–simulated data, the model demonstrates robust noise resilience, high resolution, and excellent structural fidelity on synthetic corneal OCT datasets, thereby validating the efficacy of physics-informed deep learning for quantitative OCT imaging.

Technology Category

• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search
• Machine Learning: Structured Learning

Application Category

• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Inverse scattering in optical coherence tomography (OCT) seeks to recover both structural images and intrinsic tissue optical properties, including refractive index, scattering coefficient, and anisotropy. This inverse problem is challenging due to attenuation, speckle noise, and strong coupling among parameters. We propose a regularized end-to-end deep learning framework that jointly reconstructs optical parameter maps and speckle-reduced OCT structural intensity for layer visualization. Trained with Monte Carlo-simulated ground truth, our network incorporates a physics-based OCT forward model that generates predicted signals from the estimated parameters, providing physics-consistent supervision for parameter recovery and artifact suppression. Experiments on the synthetic corneal OCT dataset demonstrate robust optical map recovery under noise, improved resolution, and enhanced structural fidelity. This approach enables quantitative multi-parameter tissue characterization and highlights the benefit of combining physics-informed modeling with deep learning for computational OCT.

Problem

• optical coherence tomography
• inverse scattering
• optical properties
• speckle noise
• parameter coupling

Innovation

• physics-based learning
• optical coherence tomography
• inverse scattering
• speckle reduction
• end-to-end reconstruction