Search

ObjectiveThe study aimed to summarize the morphological characteristics of low-grade gastric intraepithelial neoplasia (LGIN) and explore its outcomes and risk factors. Additionally, it aimed to screen the core different expression genes (DEGs) of high-grade gastric intraepithelial neoplasia (HGIN) using bioinformatics methods to identify biomarkers for early gastric cancer outcomes.MethodsThe clinical and pathological data of 449 patients with LGIN in the endoscopy center of the Second Hospital of Hebei Medical University from June 2013 to September 2018 were collected for retrospective analysis. The GSE130823 and GSE55696 data sets were selected from the Gene Expression Omnibus database, and the GEO2R tool was used to screen DEGs in HGIN and chronic gastritis tissue types. A DEG functional enrichment analysis was conducted using the Database for Annotation, Visualization, and Integrated Discovery. The STRING database was utilized to create a protein–protein interaction network, and the CytoHubba plug-in was used to screen the key genes of HGIN.ResultsThe incidence of LGIN increased with age, and most of the patients were aged between 45–59 years (P = 0.048). Lesions were found mainly in the cardia, mostly in people aged 60 (P < 0.05). Progression occurred in 42 of 449 patients, with a 9.4% rate of cancer development. Foci larger than 10 mm, ulcerative lesions, and an Helicobacter pylori-positive result were factors affecting the outcome of LGIN (P < 0.05). Seven core genes of HGIN were screened, including MYC, SOX2, CDX2, TBX3, KRT7, CDKN2A, and MUC5AC.ConclusionThe patients with LGIN reflected the potential for developing cancer. A magnifying gastroscope can contribute to the detection of early gastric cancer. Additionally, the MYC, CDX2, and TBX3 genes may act as specific biomarkers of HGIN.

Pansharpening is an important yet challenging remote sensing image processing task, which aims to reconstruct a high-resolution (HR) multispectral (MS) image by fusing a HR panchromatic (PAN) image and a low-resolution (LR) MS image. Though deep learning (DL)-based pansharpening methods have achieved encouraging performance, they are infeasible to fully utilize the deep semantic features and shallow contextual features in the process of feature fusion for a HR-PAN image and LR-MS image. In this paper, we propose an efficient full-depth feature fusion network (FDFNet) for remote sensing pansharpening. Specifically, we design three distinctive branches called PAN-branch, MS-branch, and fusion-branch, respectively. The features extracted from the PAN and MS branches will be progressively injected into the fusion branch at every different depth to make the information fusion more broad and comprehensive. With this structure, the low-level contextual features and high-level semantic features can be characterized and integrated adequately. Extensive experiments on reduced- and full-resolution datasets acquired from WorldView-3, QuickBird, and GaoFen-2 sensors demonstrate that the proposed FDFNet only with less than 100,000 parameters performs better than other detail injection-based proposals and several state-of-the-art approaches, both visually and quantitatively.