Anterior segment optical coherence tomography (AS-OCT) is a popular imaging technique that can directly visualize the anterior segment structures while inherent speckle noise severely impairs visual readability and subsequent clinical analysis. Though unpaired OCT image denoising algorithms have been developed to improve visual quality considering the limited supervised clinical data, preserving the edge structures while denoising remains challenging, especially in AS-OCT images with little hierarchy and low contrast. This work proposes an edge enhancement generative adversarial network ($E^{2}GAN$) based contrast-aware, particularly for unpaired AS-OCT image denoising. Specifically, to improve edge-structure consistency, we design a contrast attention mechanism for exploiting diverse hierarchical knowledge from multiple contrast images and adopt particular gradient-guided speckle filtering modules with an edge preservation loss for stabilizing the network. Additionally, considering that bi-directional GANs often focus on global appearance rather than essential features,$E^{2}GAN$adds a perceptual quality constraint into the cycle consistency. Extensive experiments validate the superiority of$E^{2}GAN$for AS-OCT image denoising and the benefits for downstream clinical analysis. Further experiments on the synthetic retinal OCT images prove the generalization of$E^{2}GAN$.

Optical coherence tomography (OCT) is a widely used non-invasive imaging modality for ophthalmic diagnosis. However, the inherent speckle noise becomes the leading cause of OCT image quality, and efficient speckle removal algorithms can improve image readability and benefit automated clinical analysis. As an ill-posed inverse problem, it is of utmost importance for speckle removal to learn suitable priors. In this work, we develop a score prior guided iterative solver (SPIS) with logarithmic space to remove speckles in OCT images. Specifically, we model the posterior distribution of raw OCT images as a data consistency term and transform the speckle removal from a nonlinear into a linear inverse problem in the logarithmic domain. Subsequently, the learned prior distribution through the score function from the diffusion model is utilized as a constraint for the data consistency term into the linear inverse optimization, resulting in an iterative speckle removal procedure that alternates between the score prior predictor and the subsequent non-expansive data consistency corrector. Experimental results on the private and public OCT datasets demonstrate that the proposed SPIS has an excellent performance in speckle removal and out-of-distribution (OOD) generalization. Further downstream automatic analysis on the OCT images verifies that the proposed SPIS can benefit clinical applications.

Optical coherence tomography (OCT) imaging technique has been widely used for ocular disease diagnosis. However, speckles occur in OCT images due to the property of coherent imaging, inevitably affecting the visual quality and clinical analysis. To alleviate this problem, we propose a novel gradient-guided speckle image filtering method (GGSF) with structure enhancement for directly removing speckles in OCT images. Specifically, the multiplicative characteristic of speckle noise is incorporated into the guided filtering processing for modeling raw OCT images. To avoid getting trapped in image distortions, we further employ gradient regularization to integrate the structure prior information into the guided speckle image filtering procedure. Additionally, we introduce the statistical property of speckle noise obeying a gamma distribution into the least square method solver for the resulting non-convex GGSF model. Experimental results on the AS-OCT dataset demonstrate the effectiveness of GGSF for OCT image despeckling compared with competitive methods. Furthermore, we validate the benefits of GGSF for subsequent clinical analysis with the CM-OCT dataset.

Ciliary muscle segmentation in Anterior Segment Optical Coherence Tomography (AS-OCT) images is critical significance, yet challenging due to ambiguous boundaries. In this paper, we propose a hierarchical attention multi-task network, HA-Net, based on U-Net for ciliary muscle segmentation using AS-OCT images. The network comprises a primary task for ciliary muscle segmentation and two auxiliary tasks for signed distance map regression and key point localization. The signed distance map is employed to incorporate shape priors into the model and delineate the ciliary muscle boundary, while key point localization guides the model to focus on ambiguous regions. Notably, in contrast to the widely-used multi-task model that generates results in parallel, we introduce a hierarchical attention module to exploit the affiliation prior of three tasks for generating outputs serially. Experimental results on CM544 dataset demonstrate that HA-Net outperforms state-of-the-art methods in ciliary muscle segmentation, with 0.9178 Dice score and 7.11 pixels HD95. Additionally, as a by-product of the multi-task model, key point localization facilitates the measurement of ciliary muscle thickness in clinical analysis.

Image restoration (IR) is a long-standing challenging problem in low-level image processing. It is of utmost importance to learn good image priors for pursuing visually pleasing results. In this paper, we develop a multi-channel and multi-model-based denoising autoencoder network as image prior for solving IR problem. Specifically, the network that trained on RGB-channel images is used to construct a prior at first, and then the learned prior is incorporated into single-channel grayscale IR tasks. To achieve the goal, we employ the auxiliary variable technique to integrate the higher-dimensional network-driven prior information into the iterative restoration procedure. In addition, according to the weighted aggregation idea, a multi-model strategy is put forward to enhance the network stability that favors to avoid getting trapped in local optima. Extensive experiments on image deblurring and deblocking tasks show that the proposed algorithm is efficient, robust, and yields state-of-the-art restoration quality on grayscale images.

Most low-rank tensor approximations are NP-hard problems. In this paper, we introduce a novel concept: field-of-experts (FoE) filters guided tensor completion, which aims to integrate the strengths of the emerging tensor completion method and the conventional FoE filters. Specifically, the target image is convolved by FoE filters to produce multiview features as a high-order tensor, which captures complementary information from multiple views. In order to impose the concept, we employ two strategies to model the new tensor, one is called FoE filters guided low-rank tensor completion, and another is called FoE filters guided simultaneous tensor decomposition and completion (FoE-STDC). The resulting objectives are solved efficiently by alternating minimization. Extensive experimental results validate the superior performance and robustness of the proposed methods over their corresponding counterparts in all cases. Particularly, the proposed FoE-STDC is superior to the state-of-the-art tensor completion methods.

Image prior information is a determinative factor to tackle with the ill-posed problem. In this paper, we present a multi-filters guided low-rank tensor coding (MF-LRTC) model for image restoration. The appeal of constructing a low-rank tensor is obvious in many cases for data that naturally comes from different scales and directions. The MF-LRTC takes advantages of the low-rank tensor coding to capture the sparse convolutional features generated by multi-filters representation. Using such a low-rank tensor coding would reduce the redundancy between feature vectors at neighboring locations and improve the efficiency of the overall sparse representation. In this work, we are committed to achieving this goal by convoluting the target image with filters to formulate multi-features images. Then similarity-grouped cube set extracted from the multi-features images is regarded as a low-rank tensor. The potential effectiveness of this tensor construction strategy is demonstrated in image restoration including image deblurring and compressed sensing (CS) applications.

Most of low-rank tensor approximation problems are NP-hard. Hence a great number of synthesis tensor decomposition approximation have been proposed. In this paper, we instead present an analysis-operator guided tensor decomposition. The proposed method first employs the classical Field-of-Experts (FoE) filters to produce multi-view features such that forming a higher-order tensor, and then do simultaneous tensor decomposition and completion (STDC). The multi-view features are obtained by convolving the target image with high-frequency FoE filters along different directions and scales. The proposed method is solved efficiently by alternating direction of multipliers method (ADMM). Experiments are conducted to demonstrate the superior performance of our method to state-of-the-art tensor completion methods.