Growing evidence has shown that a large number of miRNAs are abnormally expressed in cervical cancer (CC) tissues and play irreplaceable roles in tumorigenesis, progression, and metastasis. This study aimed to identify new biomarkers and pivotal genes associated with CC prognosis through comprehensive bioinformatics analysis. At first, the data of gene expression microarray (GSE30656) was downloaded from GEO database and differential miRNAs were obtained. Additionally, 4 miRNAs associated with the survival time of patients with CC were screened through TCGA differential data analysis, Kaplan-Meier, and Landmark analysis. Among them, the low expression of miR-188 and high expression of miR-223 correlated with the short survival of CC patients, while the down-regulation of miR-99a and miR-125b was closely related to the 5-year survival rate of patients. Then, based on the correspondence between the differentially expressed genes (DEGs) in CC from the TCGA data and the 4 miRNAs target genes, 58 target genes were screened to perform the analysis of function enrichment and the visualization of protein-protein interaction (PPI) networks. The seven pivotal genes of the PPI network as the target genes of four miRNAs related to prognosis, they were directly or indirectly involved in the development of CC. In this study, based on high-throughput data mining, differentially expressed miRNAs and related target genes were analyzed to provide an effective bioinformatics basis for further understanding of the pathogenesis and prognosis of CC. And the results may be a promising biomarker for the early screening of high-risk populations and early diagnosis of cervical cancer.

Long non‐coding RNAs (lncRNAs), which competitively bind miRNAs to regulate target mRNA expression in the competing endogenous RNAs (ceRNAs) network, have attracted increasing attention in breast cancer research. We aim to find more effective therapeutic targets and prognostic markers for breast cancer. LncRNA, mRNA and miRNA expression profiles of breast cancer were downloaded from TCGA database. We screened the top 5000 lncRNAs, top 5000 mRNAs and all miRNAs to perform weighted gene co‐expression network analysis. The correlation between modules and clinical information of breast cancer was identified by Pearson's correlation coefficient. Based on the most relevant modules, we constructed a ceRNA network of breast cancer. Additionally, the standard Kaplan‐Meier univariate curve analysis was adopted to identify the prognosis of lncRNAs. Ultimately, a total of 23 and 5 modules were generated in the lncRNAs/mRNAs and miRNAs co‐expression network, respectively. According to the Green module of lncRNAs/mRNAs and Blue module of miRNAs, our constructed ceRNA network consisted of 52 lncRNAs, 17miRNAs and 79 mRNAs. Through survival analysis, 5 lncRNAs (AL117190.1, COL4A2‐AS1, LINC00184, MEG3 and MIR22HG) were identified as crucial prognostic factors for patients with breast cancer. Taken together, we have identified five novel lncRNAs related to prognosis of breast cancer. Our study has contributed to the deeper understanding of the molecular mechanism of breast cancer and provided novel insights into the use of breast cancer drugs and prognosis.

No abstract

Segmentation of skin lesions is a challenging task because of the wide range of skin lesion shapes, sizes, colors, and texture types. In the past few years, deep learning networks such as U-Net have been successfully applied to medical image segmentation and exhibited faster and more accurate performance. In this paper, we propose an extended version of U-Net for the segmentation of skin lesions using the concept of the triple attention mechanism. We first selected regions using attention coefficients computed by the attention gate and contextual information. Second, a dual attention decoding module consisting of spatial attention and channel attention was used to capture the spatial correlation between features and improve segmentation performance. The combination of the three attentional mechanisms helped the network to focus on a more relevant field of view of the target. The proposed model was evaluated using three datasets, ISIC-2016, ISIC-2017, and PH2. The experimental results demonstrated the effectiveness of our method with strong robustness to the presence of irregular borders, lesion and skin smooth transitions, noise, and artifacts.