Optical coherence tomography angiography (OCTA) is an emerging non-invasive ophthalmic imaging modality. The morphology and contour of the foveal avascular zone (FAZ) are critical biomarkers for the diagnosis of various ophthalmic and systemic diseases; therefore, achieving its accurate segmentation is of substantial clinical significance. To address challenges such as low contrast, indistinct boundaries, and structural confusion with the surrounding retinal vasculature in OCTA images, this paper proposes a multi-scale dilated convolution and dual attention network (MSAD-Net). By integrating a multi-scale dilated convolution module (MSDM) to capture extensive multi-scale contextual information and employing a dual attention module (DAM) to integrate complementary channel and spatial features, thereby synergistically boosting the feature representation of key regions. Experimental results demonstrated that the model achieved superior performance and robust generalization across multiple evaluation metrics, including the Dice similarity coefficient, Jaccard index, precision, and recall—on two public datasets. These findings confirm the robustness of MSAD-Net in the fine segmentation of the FAZ, providing high-precision technical support for the clinical quantitative analysis of ophthalmic diseases.
Endometrial cancer (EC) is one of the most common gynecological malignancies, with an increasing incidence rate worldwide. Accurate segmentation of lesion areas in computed tomography (CT) images is a critical step in assisting clinical diagnosis. In this study, we propose a novel deep learning-based segmentation model, termed spatial choice and weight union network (SCWU-Net), which incorporates two newly designed modules: the spatial selection module (SSM) and the combination weight module (CWM). The SSM enhances the model’s ability to capture contextual information through deep convolutional blocks, while the CWM, based on joint attention mechanisms, is employed within the skip connections to further boost segmentation performance. By integrating the strengths of both modules into a U-shaped multi-scale architecture, the model achieves precise segmentation of EC lesion regions. Experimental results on a public dataset demonstrate that SCWU-Net achieves a Dice similarity coefficient (DSC) of 82.98%, an intersection over union (IoU) of 78.63%, a precision of 92.36%, and a recall of 84.10%. Its overall performance is significantly outperforming other state-of-the-art models. This study enhances the accuracy of lesion segmentation in EC CT images and holds potential clinical value for the auxiliary diagnosis of endometrial cancer.
