Search

Additional File 1: VSS score. Additional File 2: Digital photographs of scars. Additional File 3: Ultrasound of scars. Additional File 4: Patient satisfaction of scars.

The aim of this study is to investigate the efficacy of treatment-combined electrocauterization after removal of apocrine sweat glands with ultrasound-guided (BOTOX) injection for adolescents with axillary bromhidrosis. From January 2015 to January 2018, 90 adolescents with axillary bromhidrosis were recruited and randomly divided into three groups (group A, B, and C). Patients in group A underwent electrocauterization after removal of apocrine sweat glands, patients in group B received ultrasound-guided BOTOX injection, and patients in group C had electrocauterization after the removal of apocrine sweat glands followed by ultrasound-guided BOTOX administration after two weeks. All patients were followed up for one year. The percentages of axillary malodor score at twelve months after treatments compared with pre-operation for group A, group B, and group C were 20.2%, 27.5%, and 12.5%, respectively. Significant statistical differences were observed in changes of axillary malodor score among the three groups at three, six, nine and twelve months postoperatively. Satisfaction rates of group A, group B, and group C were 90.0%, 93.3%, and 96.7%, respectively. The side effects after surgery in group A were similar to group C, and no side effect was observed in group B. One-year postoperative scar formation rate of group A was 16.7%, which was significantly higher than that of group B (0) and group C (6.66%). Combination of electrocauterization after removal of apocrine sweat glands with ultrasound-guided BOTOX injection showed better treatment effects.

University of Technology Sydney. Faculty of Engineering and Information Technology. Fine-grained recognition is challenging in computer vision and artificial intelligence. It aims to identify under subcategories of given images but suffers from small inter-class variance and large intra-class variance along with multiple object scales and complex background, leading to a more complex problem space. Recently, deep neural networks have extensively promoted the development of fine-grained recognition. However, the existing methods still suffer from several issues, including data limitation, model interpretation, and performance. In this thesis, we propose several data-transformation models to address these challenges. First, we develop a unified framework (MGN-CNN) based on a mixture of experts to promote diversity among experts by combing a gradually-enhanced learning strategy and a KullbackLeibler divergence based constraint. The strategy learns new experts on the dataset with prior knowledge from former experts and adds them to the model sequentially. At the same time, the introduced constraint forces the experts to produce diverse prediction distributions. These drive the experts to learn the task from different aspects, making them specialized in various subspace problems. Second, we propose Intra-class Part Swapping (InPS) that produces new data by performing attention-guided content swapping on input pairs from the same class. Compared with previous approaches, InPS avoids introducing noisy labels and ensures a likely holistic structure of objects in generated images. We demonstrate InPS outperforms the most recent augmentation approaches in both fine-grained recognition and weakly object localization. Finally, we explore fine-grained zero-shot learning and introduce a novel structure-aware feature generation scheme, termed SA-GAN, to explicitly account for the topological structure in learning both the latent space and the generative networks. This topology-preserving mechanism enables our method to significantly enhance the generalization capability on unseen-classes and consequently improve the classification performance.

Background A keloid is a benign human skin tumor that resulted by fibrous overgrowths during wound healing. Long noncoding RNAs (lncRNAs) are indicated to involve in the development of keloid. However, the role of lncRNA LINC01615 in regulating keloid development and the underlying mechanism are still unknown. Methods The expression levels of LINC01615, miR-590-3p and fibroblast growth factor 2 (FGF2) mRNA were detected by quantitative real-time polymerase chain reaction (qRT-PCR). The protein levels of apoptosis-related proteins, α-smooth muscle actin (α-SMA), collagens and FGF2 were measured by western blot. The effect of LINC01615 on keloid development was assessed by cell proliferation, apoptosis and collagen deposition of keloid fibroblasts which were determined by Cell Counting Kit-8 (CCK-8), flow cytometry assay and the protein levels of collagens, respectively. The relationships between LINC01615 and miR-590-3p, miR-590-3p and FGF2 were predicated by online software and confirmed by dual-luciferase reporter assay and RNA pull-down assay. Results We first found that LINC01615 was upregulated in keloid tissues and fibroblasts, and LINC01615 promoted cell proliferation and collagen deposition and suppressed apoptosis in keloid fibroblasts. LINC01615 targeted miR-590-3p and downregulated miR-590-3p expression, and overexpression of miR-590-3p inhibited the development of keloid. Then, FGF2 was identified as a target of miR-590-3p. LINC01615 facilitated keloid development via regulating FGF2 expression through miR-590-3p. Conclusion Our study demonstrated that LINC01615 contributed to keloid development via the miR-590-3p/FGF2 axis.

LIM kinases modulate multiple aspects of cancer development, including cell proliferation and survival. As the mechanisms of LIMK‐associated tumorigenesis are still unclear, we analyzed the tumorigenic functions of LIM kinase 2 (LIMK2) in human bladder cancer (BC) and explored whether the newly identified LIMK2 3´‐UTR SNP rs2073859 (G‐to‐A allele) is correlated with clinical features. Expression levels of LIMK2 in 38 human BC tissues and eight cell lines were examined using quantitative real‐time PCR and immunohistochemistry. LIMK2 was overexpressed in most BC tissues (27/38, 71%) and BC‐derived cell lines (6/8), and was more frequently overexpessed in high‐grade than low‐grade BC (80% vs. 47%). The effects of LIMK2 on BC cell proliferation, survival and migration, were studied by overexpression and RNA interference approaches in vitro and in vivo. LIMK2 overexpression promoted proliferation, migration and invasion of BC cells, while LIMK2 depletion inhibited cell invasion and viability and induced growth arrest in vitro and in vivo. PCR‐Restriction Fragment Length Polymorphism (RFLP) was used to genotype LIMK2 SNP rs2073859 and multivariate logistic regression applied to assess the relationship between allele frequency and clinical features in 139 BC patients. Functional analyses localized SNP rs2073859 within the microRNA‐135a seed‐binding region and revealed significantly lower LIMK2 G allele expression. The frequency of A genotypes (AG + AA) was higher in the BC group than normal controls and correlated with risks of high‐grade and high‐stage BC. In conclusion, LIMK2 may function as an oncogene in human BC, while allele‐specific regulation by microRNA‐135a may influence disease risk., What's new? Genetic variations in microRNA (miRNA)‐binding regions are suspected to influence tumorigenesis. In this study, in tissues from bladder cancer (BC) patients, the single nucleotide polymorphism rs2073859 (G‐to‐A allele) in the 3’‐UTR miRNA‐135a‐binding site of the oncogene LIM kinase 2 (LIMK2) was associated with gene overexpression and higher clinical grade. A high frequency of rs2073859 A genotypes was detected in BC tissues. In BC cells and mice, LIMK2 overexpression promoted cell proliferation, migration, and invasion and accelerated BC tumor growth. The findings suggest that LIMK2 normally is downregulated by miRNA‐135a and becomes tumorigenic following miRNA‐135a disruption by rs2073859 variant A.

Background Colorectal cancer (CRC) is a malignant tumor, and the overall prognosis of patients with advanced CRC is still unsatisfactory. Circular RNAs (circRNAs) vesicle-associated membrane protein-associated protein A (circVAPA) could act as an underlying biomarker in CRC. This study aimed to explore the mechanism of circVAPA in the regulation of CRC growth. Methods CircVAPA level was measured in CRC tumor tissues. The expression levels of circVAPA, VAPA mRNA, microRNA-125a (miR-125a), and cAMP response element binding 5 (CREB5) in CRC cells were detected by RT-qPCR. Cell cycle progression, migration and invasion, extracellular acidification rate (ECAR) and oxygen consumption rate (OCR) were measured by flow cytometry, transwell assays and Seahorse XF96 Glycolysis Analyzer, severally. The levels of glucose uptake, lactate and ATP production were examined by Glucose Uptake Colorimetric Assay kit, Lactate Assay kit and ATP Colorimetric Assay kit, respectively. The interaction between miR-125a and circVAPA or CREB5 was predicted by Starbase or DIANA TOOL, and verified by the dual-luciferase reporter and RNA Immunoprecipitation (RIP) assays. Results CircVAPA level was up-regulated in CRC tumor tissues. Expression levels of circVAPA and CREB5 were increased, and miR-125a was decreased in CRC cells. CircVAPA knockdown repressed CRC cells cycle progression, migration, invasion and glycolysis. CircVAPA acted as a miR-125a sponge to regulate CREB5 expression. Rescue assay confirmed that miR-125a deletion or CREB5 overexpression weakened the inhibitory effect of circVAPA knockdown on CRC growth. Conclusion Our studies disclosed that circVAPA knockdown suppressed CRC cells cycle progression, migration, invasion and glycolysis partly by modulating miR-125a/CREB5 axis, suggesting a potential therapeutic strategy for CRC treatment.