Automatic segmentation and differentiation of retinal arteriole and venule (AV), defined as small blood vessels directly before and after the capillary plexus, are of great importance for the diagnosis of various eye diseases and systemic diseases, such as diabetic retinopathy, hypertension, and cardiovascular diseases. Optical coherence tomography angiography (OCTA) is a recent imaging modality that provides capillary-level blood flow information. However, OCTA does not have the colorimetric and geometric differences between AV as the fundus photography does. Various methods have been proposed to differentiate AV in OCTA, which typically needs the guidance of other imaging modalities. In this study, we propose a cascaded neural network to automatically segment and differentiate AV solely based on OCTA. A convolutional neural network (CNN) module is first applied to generate an initial segmentation, followed by a graph neural network (GNN) to improve the connectivity of the initial segmentation. Various CNN and GNN architectures are employed and compared. The proposed method is evaluated on multi-center clinical datasets, including 3 ×3 mm2 and 6 ×6 mm2 OCTA. The proposed method holds the potential to enrich OCTA image information for the diagnosis of various diseases.

The segmentation of pathological fluid lesions in optical coherence tomography (OCT), including intraretinal fluid, subretinal fluid, and pigment epithelial detachment, is of great importance for the diagnosis and treatment of various eye diseases such as neovascular age-related macular degeneration and diabetic macular edema. Although significant progress has been achieved with the rapid development of fully convolutional neural networks (FCN) in recent years, some important issues remain unsolved. First, pathological fluid lesions in OCT show large variations in location, size, and shape, imposing challenges on the design of FCN architecture. Second, fluid lesions should be continuous regions without holes inside. But the current architectures lack the capability to preserve the shape prior information. In this study, we introduce an FCN architecture for the simultaneous segmentation of three types of pathological fluid lesions in OCT. First, attention gate and spatial pyramid pooling modules are employed to improve the ability of the network to extract multi-scale objects. Then, we introduce a novel curvature regularization term in the loss function to incorporate shape prior information. The proposed method was extensively evaluated on public and clinical datasets with significantly improved performance compared with the state-of-the-art methods.

BACKGROUND: There are ever increasing researches implying that noncoded RNAs (ncRNAs) specifically circular RNAs (circRNAs) and microRNAs (miRNAs) in exosomes play vital roles in respiratory disease. However, the detailed mechanisms persist to be unclear in mycobacterial infection. METHODS: In order to detect circRNAs and miRNAs expression pattern and potential biological function in tuberculosis, we performed immense parallel sequencing for exosomal ncRNAs from THP-1-derived macrophages infected by Mycobacterium tuberculosis H37Ra, Mycobacterium bovis BCG and control Streptococcus pneumonia, respectively and uninfected normal cells. Besides, THP-1-derived macrophages were used to verify the validation of differential miRNAs, and monocytes from PBMCs and clinical plasma samples were used to further validate differentially expressed miR-185-5p. RESULTS: Many exosomal circRNAs and miRNAs associated with tuberculosis infection were recognized. Extensive enrichment analyses were performed to illustrate the major effects of altered ncRNAs expression. Moreover, the miRNA-mRNA and circRNA-miRNA networks were created and expected to reveal their interrelationship. Further, significant differentially expressed miRNAs based on Exo-BCG, Exo-Ra and Exo-Control, were evaluated, and the potential target mRNAs and function were analyzed. Eventually, miR-185-5p was collected as a promising potential biomarker for tuberculosis. CONCLUSION: Our findings provide a new vision for exploring biological functions of ncRNAs in mycobacterial infection and screening novel potential biomarkers. To sum up, exosomal ncRNAs might represent useful functional biomarkers in tuberculosis pathogenesis and diagnosis.