Axial super-resolution optical coherence tomography via complex-valued network

Abstract
Objectives. Optical coherence tomography (OCT) is a fast and non-invasive optical interferometric imaging technique that can provide high-resolution cross-sectional images of biological tissues. OCT’s key strength is its depth resolving capability which remains invariant along the imaging depth and is determined by the axial resolution. The axial resolution is inversely proportional to the bandwidth of the OCT light source. Thus, the use of broadband light sources can effectively improve the axial resolution and however leads to an increased cost. In recent years, real-valued deep learning technique has been introduced to obtain super-resolution optical imaging. This study is aimed to achieve axial super-resolution OCT imaging with complex-valued deep learning technique.
Approach. A complex-valued super-resolution network (CVSR-Net) was designed to fully utilize the amplitude and phase of OCT signal to reconstruct an OCT image with an improved axial resolution.
Main results. The method was evaluated on three OCT datasets. The results show that the CVSR-Net outperforms its real-valued counterpart with a better depth resolving capability. Furthermore, comparisons were made between our network, six prevailing real-valued networks and their complex-valued counterparts. The results demonstrate that the complex-valued network exhibited a better super-resolution performance than its real-valued counterpart and our proposed CVSR-Net achieved the best performance. In addition, the CVSR-Net was tested on out-of-distribution domain datasets and its super-resolution performance was well maintained as compared to that on source domain datasets, indicating a good generalization capability.
Significance. The good performance of the CVSR-Net in both axial super-resolution and generalization indicate that this method has a good potential to reduce OCT system cost by utilizing narrow-band light source while retaining the axial resolution.

Figure 1. The network architecture of the CVSR-Net.

Figure 2. Generalization performance of the CVSR-Net: (a) and (e) are the GT images of the in vivo swine cornea dataset and the in vivo human retina dataset respectively; (b) and (f) are the LR images; (c) and (g) are the SR images generated by the model trained with their respective source domain datasets; (d) and (h) are the SR images generated by the model trained with the ex vivo swine esophagus dataset. The LR images are created by the 37.5% spectral truncation.

Download paper here