Semantic segmentation of surgery scenarios is a fundamental task for computer-aided surgery systems. Precise segmentation of surgical instruments and anatomies contributes to capturing accurate spatial information for tracking. However, uneven reflection and class imbalance lead the segmentation in cataract surgery to a challenging task. To desirably conduct segmentation, a network with multi-view decoders (MVD-Net) is proposed to present a generalizable segmentation for cataract surgery. Two discrepant decoders are implemented to achieve multi-view learning with the backbone of U-Net. The experiment is carried out on the Cataract Dataset for Image Segmentation (CaDIS). The ablation study verifies the effectiveness of the proposed modules in MVD-Net, and superior performance is provided by MVD-Net in the comparison with the state-of-the-art methods. The source code will be publicly released.

Cataracts are the most common blinding disease, and also impact the observation of the fundus. To boost the fundus examination of cataract patients, restoration algorithms have been proposed to address the degradation of fundus images caused by cataracts. However, it is impractical in clinics to collect paired or annotated fundus images for developing restoration models. In this paper, a restoration algorithm is designed for cataractous images without paired or annotated data. Domain generalization (DG) is applied to learn domain-invariant features (DIFs) from synthesized data, and the high-frequency components (HFCs) are extracted to conduct domain alignment. The proposed algorithm is used on unseen target data in the experiments. The effectiveness of the algorithm is demonstrated in the ablation study and compared with state-of-the-art methods. The code of this paper will be released at https://github.com/HeverLaw/Restoration-of-Cataract-Images-via-Domain-Generalization.

Quality degradation (QD) is common in the fundus images collected from the clinical environment. Although diagnosis models based on convolutional neural networks (CNN) have been extensively used to interpret retinal fundus images, their performances under QD have not been assessed. To understand the effects of QD on the performance of CNN-based diagnosis model, a systematical study is proposed in this paper. In our study, the QD of fundus images is controlled by independently or simultaneously importing quantified interferences (e.g., image blurring, retinal artifacts, and light transmission disturbance). And the effects of diabetic retinopathy (DR) grading systems are thus analyzed according to the diagnosis performances on the degraded images. With images degraded by quantified interferences, several CNN-based DR grading models (e.g., AlexNet, SqueezeNet, VGG, DenseNet, and ResNet) are evaluated. The experiments demonstrate that image blurring causes a significant decrease in performance, while the impacts from light transmission disturbance and retinal artifacts are relatively slight. Superior performances are achieved by VGG, DenseNet, and ResNet in the absence of image degradation, and their robustness is presented under the controlled degradation.