A model-based iterative learning approach for diffuse optical tomography

TL;DR

This paper introduces a model-based iterative learning method for diffuse optical tomography that improves image reconstruction accuracy and computational efficiency by combining learned components with physical models, validated on simulations and experiments.

Contribution

It presents a novel hybrid approach that intertwines deep learning with the physical DOT model, enhancing reconstruction quality and speed over traditional methods.

Findings

Improved absorption and scattering estimates for complex targets.

Effective compensation for modeling errors due to coarse discretization.

Faster computation times compared to standard Gauss-Newton methods.

Abstract

Diffuse optical tomography (DOT) utilises near-infrared light for imaging spatially distributed optical parameters, typically the absorption and scattering coefficients. The image reconstruction problem of DOT is an ill-posed inverse problem, due to the non-linear light propagation in tissues and limited boundary measurements. The ill-posedness means that the image reconstruction is sensitive to measurement and modelling errors. The Bayesian approach for the inverse problem of DOT offers the possibility of incorporating prior information about the unknowns, rendering the problem less ill-posed. It also allows marginalisation of modelling errors utilising the so-called Bayesian approximation error method. A more recent trend in image reconstruction techniques is the use of deep learning techniques, which have shown promising results in various applications from image processing to tomographic reconstructions. In this work, we study the non-linear DOT inverse problem of estimating the absorption and scattering coefficients utilising a `model-based' learning approach, essentially intertwining learned components with the model equations of DOT. The proposed approach was validated with 2D simulations and 3D experimental data. We demonstrated improved absorption and scattering estimates for targets with a mix of smooth and sharp image features, implying that the proposed approach could learn image features that are difficult to model using standard Gaussian priors. Furthermore, it was shown that the approach can be utilised in compensating for modelling errors due to coarse discretisation enabling computationally efficient solutions. Overall, the approach provided improved computation times compared to a standard Gauss-Newton iteration.