Search

We propose a stacked convolutional neural network incorporating a novel and efficient pyramid residual attention (PRA) module for the task of automatic segmentation of dermoscopic images. Precise segmentation is a significant and challenging step for computer-aided diagnosis technology in skin lesion diagnosis and treatment. The proposed PRA has the following characteristics: First, we concentrate on three widely used modules in the PRA. The purpose of the pyramid structure is to extract the feature information of the lesion area at different scales, the residual means is aimed to ensure the efficiency of model training, and the attention mechanism is used to screen effective features maps. Thanks to the PRA, our network can still obtain precise boundary information that distinguishes healthy skin from diseased areas for the blurred lesion areas. Secondly, the proposed PRA can increase the segmentation ability of a single module for lesion regions through efficient stacking. The third, we incorporate the idea of encoder-decoder into the architecture of the overall network. Compared with the traditional networks, we divide the segmentation procedure into three levels and construct the pyramid residual attention network (PRAN). The shallow layer mainly processes spatial information, the middle layer refines both spatial and semantic information, and the deep layer intensively learns semantic information. The basic module of PRAN is PRA, which is enough to ensure the efficiency of the three-layer architecture network. We extensively evaluate our method on ISIC2017 and ISIC2018 datasets. The experimental results demonstrate that PRAN can obtain better segmentation performance comparable to state-of-the-art deep learning models under the same experiment environment conditions. [ABSTRACT FROM AUTHOR]

Accurate segmentation of retinal vessels is critical to the mechanism, diagnosis, and treatment of many ocular pathologies. Due to the poor contrast and inhomogeneous background of fundus imaging and the complex structure of retinal fundus images, this makes accurate segmentation of blood vessels from retinal images still challenging. In this paper, we propose an effective framework for retinal vascular segmentation, which is innovative mainly in the retinal image pre-processing stage and segmentation stage. First, we perform image enhancement on three publicly available fundus datasets based on the multiscale retinex with color restoration (MSRCR) method, which effectively suppresses noise and highlights the vessel structure creating a good basis for the segmentation phase. The processed fundus images are then fed into an effective Reverse Fusion Attention Residual Network (RFARN) for training to achieve more accurate retinal vessel segmentation. In the RFARN, we use Reverse Channel Attention Module (RCAM) and Reverse Spatial Attention Module (RSAM) to highlight the shallow details of the channel and spatial dimensions. And RCAM and RSAM are used to achieve effective fusion of deep local features with shallow global features to ensure the continuity and integrity of the segmented vessels. In the experimental results for the DRIVE, STARE and CHASE datasets, the evaluation metrics were 0.9712, 0.9822 and 0.9780 for accuracy (Acc), 0.8788, 0.8874 and 0.8352 for sensitivity (Se), 0.9803, 0.9891 and 0.9890 for specificity (Sp), area under the ROC curve(AUC) was 0.9910, 0.9952 and 0.9904, and the F1-Score was 0.8453, 0.8707 and 0.8185. In comparison with existing retinal image segmentation methods, e.g. UNet, R2UNet, DUNet, HAnet, Sine-Net, FANet, etc., our method in three fundus datasets achieved better vessel segmentation performance and results. [ABSTRACT FROM AUTHOR]

Segmentation of retinal vessels is important for doctors to diagnose some diseases. The segmentation accuracy of retinal vessels can be effectively improved by using deep learning methods. However, most of the existing methods are incomplete for shallow feature extraction, and some superficial features are lost, resulting in blurred vessel boundaries and inaccurate segmentation of capillaries in the segmentation results. At the same time, the "layer-by-layer" information fusion between encoder and decoder makes the feature information extracted from the shallow layer of the network cannot be smoothly transferred to the deep layer of the network, resulting in noise in the segmentation features. In this paper, we propose the MFI-Net (Multi-resolution fusion input network) network model to alleviate the above problem to a certain extent. The multi-resolution input module in MFI-Net avoids the loss of coarse-grained feature information in the shallow layer by extracting local and global feature information in different resolutions. We have reconsidered the information fusion method between the encoder and the decoder, and used the information aggregation method to alleviate the information isolation between the shallow and deep layers of the network. MFI-Net is verified on three datasets, DRIVE, CHASE_DB1 and STARE. The experimental results show that our network is at a high level in several metrics, with F1 higher than U-Net by 2.42%, 2.46% and 1.61%, higher than R2U-Net by 1.47%, 2.22% and 0.08%, respectively. Finally, this paper proves the robustness of MFI-Net through experiments and discussions on the stability and generalization ability of MFI-Net. [ABSTRACT FROM AUTHOR]

The optic disc(OD) and the optic cup(OC) segmentation is an key step in fundus medical image analysis. Previously, FCN-based methods have been proposed for medical image segmentation tasks. However, the consecutive convolution and pooling operations usually hinder dense prediction tasks which require detailed spatial information, such as image segmentation. In this paper, we propose a network called Recurrent Fully Convolution Network(RFC-Net) for automatic joint segmentation of the OD and the OC, which can captures more high-level information and subtle edge information. The RFC-Net can minimize the loss of spatial information. It is mainly composed of multi-scale input layer, recurrent fully convolutional network, multiple output layer and polar transformation. In RFC-Net, the multi-scale input layer constructs an image pyramid. We propose four recurrent units, which are respectively applied to RFC-Net. Recurrent convolution layer effectively ensures feature representation for OD and OC segmentation tasks through feature accumulation. For each multiple output image, the multiple output cross entropy loss function is applied. To better balance the cup ratio of the segmented image, the polar transformation is used to transform the fundus image from the cartesian coordinate system to the polar coordinate system. We evaluate the effectiveness and generalization of the proposed method on the DRISHTI-GS1 dataset. Compared with the original FCN method and other state-of-the-art methods, the proposed method achieves better segmentation performance. [ABSTRACT FROM AUTHOR]

Although successful detection of malignant tumors from histopathological images largely depends on the long-term experience of radiologists, experts sometimes disagree with their decisions. Computer-aided diagnosis provides a second option for image diagnosis, which can improve the reliability of experts’ decision-making. Automatic and precision classification for breast cancer histopathological image is of great importance in clinical application for identifying malignant tumors from histopathological images. Advanced convolution neural network technology has achieved great success in natural image classification, and it has been used widely in biomedical image processing. In this paper, we design a novel convolutional neural network, which includes a convolutional layer, small SE-ResNet module, and fully connected layer. We propose a small SE-ResNet module which is an improvement on the combination of residual module and Squeeze-and-Excitation block, and achieves the similar performance with fewer parameters. In addition, we propose a new learning rate scheduler which can get excellent performance without complicatedly fine-tuning the learning rate. We use our model for the automatic classification of breast cancer histology images (BreakHis dataset) into benign and malignant and eight subtypes. The results show that our model achieves the accuracy between 98.87% and 99.34% for the binary classification and achieve the accuracy between 90.66% and 93.81% for the multi-class classification. [ABSTRACT FROM AUTHOR]

Dasypyrum villosum has been used as a valuable gene resource for disease resistances, yield increase and quality improvement in wheat. A novel wheat-D. villosum alien introgression line CD-3 was generated through hybridization between the common wheat Chinese Spring (CS) and a CS- D. villosum 3V addition line having considerably high stripe rust resistance, which enable the characterization of a potential new stripe rust resistance gene (s) derived from D. villosum. The results of non-denaturing fluorescent in situ hybridization (ND-FISH) showed that CD-3 contained 42 chromosomes, including a 3V chromosome pair, and the absence of both of the 3D chromosomes. PCR-based Landmark Unique Gene (PLUG) molecular marker analysis supported results from the FISH analysis, revealing CD-3 was a wheat-D. villosum 3V (3D) disomic substitution line. Resistant test of stripe rust on 52 plants of F2 generation (CD-3/CS), CD-3, CS and D.villosum have been conducted at seedling stage. 7 plants of F2 generation possessing two 3V chromosomes exhibited high resistance to stripe rust as CD-3 and D.villosum, 10 plants carrying one 3V chromosome and 35 plants without 3V chromosome were susceptive to stripe rust as CS. The result implied the high stripe rust resistance of CD-3 should be controlled by recessive gene(s) originating from D.villosum. To rapidly detect chromosome 3V in the genetic background of wheat, we developed a novel Sequence Characterized Amplified Region (SCAR) marker specific for 3V chromosome based on the sequence of a grain size-related gene DvGS5 in D. villosum, an orthologue of TaGS5 from wheat. The SCAR marker was designated DvGS5-1443, which could successfully amplify a unique 3V-specific fragment in CD-3 and D. villosum, suggesting that this SCAR marker could facilitate targeting the chromosome 3V in the genetic background of wheat for wheat improvement. [ABSTRACT FROM AUTHOR]

Introduction: Hybrid ablation, an emerging therapy that combines surgical intervention and catheter ablation, has become a viable option for the treatment of persistent atrial fibrillation. In this analysis, we aimed to evaluate the safety and efficacy of hybrid ablation, as well as compare the outcomes of one-step and staged approaches. Methods: We conducted a search in major online databases and selected the studies that met the inclusion criteria. The primary endpoint was defined as no episode of atrial fibrillation or atrial tachycardia lasting longer than 30 seconds without administration of antiarrhythmic drugs. Results: Sixteen studies including 785 patients (paroxysmal atrial fibrillation, n = 83; persistent atrial fibrillation, n = 214; long-standing persistent atrial fibrillation, n = 488) were selected. Average history of atrial fibrillation was (5.0±1.6) years. The pooled proportion of patients who were arrhythmia-free at the primary endpoint was 73% (95% CI, 64%–81%, Cochran’s Q, P<0.001; I2 = 81%). The pooled rate of severe short-term complications was 4% (95% CI, 2%–7%, Cochran’s Q, P = 0.01; I2 = 51%). The success rate after one-step procedures (69%) was lower than that after staged procedures (78%). The staged approach could ultimately prove to be safer, although complication rates were relatively low for both approaches (2% and 5%, respectively). Conclusions: Hybrid ablation is an effective and generally safe procedure. The current data suggest that staged hybrid ablation could be the optimal approach, as it is associated with a higher success rate and a seemingly lower complication rate. Additional randomized controlled trials are necessary to confirm these results. [ABSTRACT FROM AUTHOR]

Tibetan semi-wild wheat (Triticum aestivum ssp. tibetanum Shao) is a semi-wild hexaploid wheat resource that is only naturally distributed in the Qinghai-Tibet Plateau. Brittle rachis and hard threshing are two important characters of Tibetan semi-wild wheat. A whole-genome linkage map of T. aestivum ssp. tibetanum was constructed using a recombinant inbred line population (Q1028×ZM9023) with 186 lines, 564 diversity array technology markers, and 117 simple sequence repeat markers. Phenotypic data on brittle rachis and threshability, as two quantitative traits, were evaluated on the basis of the number of average spike rachis fragments per spike and percent threshability in 2012 and 2013, respectively. Quantitative trait locus (QTL) mapping performed using inclusive composite interval mapping analysis clearly identified four QTLs for brittle rachis and three QTLs for threshability. However, three loci on 2DS, 2DL, and 5AL showed pleiotropism for brittle rachis and threshability; they respectively explained 5.3%, 18.6%, and 18.6% of phenotypic variation for brittle rachis and 17.4%, 13.2%, and 35.2% of phenotypic variation for threshability. A locus on 3DS showed an independent effect on brittle rachis, which explained 38.7% of the phenotypic variation. The loci on 2DS and 3DS probably represented the effect of Tg and Br1, respectively. The locus on 5AL was in very close proximity to the Q gene, but was different from the predicted q in Tibetan semi-wild wheat. To our knowledge, the locus on 2DL has never been reported in common wheat but was prominent in T. aestivum ssp. tibetanum accession Q1028. It remarkably interacted with the locus on 5AL to affect brittle rachis. Several major loci for brittle rachis and threshability were identified in Tibetan semi-wild wheat, improving the understanding of these two characters and suggesting the occurrence of special evolution in Tibetan semi-wild wheat. [ABSTRACT FROM AUTHOR]

Bacterial expansin-like proteins have synergistically increased cellulose hydrolysis by cellulolytic enzymes during the initial stages of biofuel production, but they have not been tested on livestock feeds. The objectives of this study were to: isolate and express an expansin-like protein (BsEXLX1), to verify its disruptive activity (expansion) on cotton fibers by immunodetection (Experiment 1), and to determine the effect of dose, pH and temperature for BsEXLX1 and cellulase to synergistically hydrolyze filter paper (FP) and carboxymethyl cellulose (CMC) under laboratory (Experiment 2) and simulated ruminal (Experiment 3) conditions. In addition, we determined the ability of BsEXLX1 to synergistically increase hydrolysis of corn and bermudagrass silages by an exogenous fibrolytic enzyme (EFE) (Experiment 4) and how different doses of BsEXLX1 and EFE affect the gas production (GP), in vitro digestibility and fermentation of a diet for dairy cows (Experiment 5). In Experiment 1, immunofluorescence-based examination of cotton microfiber treated without or with recombinant expansin-like protein expressed from Bacillus subtilis (BsEXLX1) increased the surface area by > 100% compared to the untreated control. In Experiment 2, adding BsEXLX1 (100 μg/g FP) to cellulase (0.0148 FPU) increased release of reducing sugars compared to cellulase alone by more than 40% (P < 0.01) at optimal pH (4.0) and temperature (50°C) after 24 h. In Experiment 3 and 4, adding BsEXLX1 to cellulase or EFE, synergistically increased release of reducing sugars from FP, corn and bermudagrass silages under simulated ruminal conditions (pH 6.0, 39°C). In Experiment 5, increasing the concentration of BsEXLX1 linearly increased (P < 0.01) GP from fermentation of a diet for dairy cows by up to 17.8%. Synergistic effects between BsEXLX1 and EFE increased in vitro NDF digestibility of the diet by 23.3% compared to the control. In vitro digestibility of hemicellulose and butyrate concentration were linearly increased by BsEXLX1 compared to the control. This study demonstrated that BsEXLX1 can improve the efficacy of cellulase and EFE at hydrolyzing pure substrates and dairy cow feeds, respectively. [ABSTRACT FROM AUTHOR]

The changes in the microbial community structure during acute exacerbations of severe chronic obstructive pulmonary disease (COPD) in hospitalized patients remain largely uncharacterized. Therefore, further studies focused on the temporal dynamics and structure of sputum microbial communities during acute exacerbation of COPD (AECOPD) would still be necessary. In our study, the use of molecular microbiological techniques provided insight into both fungal and bacterial diversities in AECOPD patients during hospitalization. In particular, we examined the structure and varieties of lung microbial community in 6 patients with severe AECOPD by amplifying 16S rRNA V4 hyper-variable and internal transcribed spacer (ITS) DNA regions using barcoded primers and the Illumina sequencing platform. Sequence analysis showed 261 bacterial genera representing 20 distinct phyla, with an average number of genera per patient of >157, indicating high diversity. Acinetobacter, Prevotella, Neisseria, Rothia, Lactobacillus, Leptotrichia, Streptococcus, Veillonella, and Actinomyces were the most commonly identified genera, and the average total sequencing number per sputum sample was >10000 18S ITS sequences. The fungal population was typically dominated by Candia, Phialosimplex, Aspergillus, Penicillium, Cladosporium and Eutypella. Our findings highlight that COPD patients have personalized structures and varieties in sputum microbial community during hospitalization periods. [ABSTRACT FROM AUTHOR]

The adverse impact of antibiotics on the gut microbiota has attracted extensive interest, particularly due to the development of microbiome research techniques in recent years. However, a direct comparison of the dynamic effects of various types of antibiotics using the same animal model has not been available. In the present study, we selected six antibiotics from four categories with the broadest clinical usage, namely, β-lactams (Ceftriaxone Sodium, Cefoperazone/Sulbactam and meropenem), quinolones (ofloxacin), glycopeptides (vancomycin), and macrolides (azithromycin), to treat BALB/c mice. Stool samples were collected during and after the administration of antibiotics, and microbial diversity was analyzed through Illumina sequencing and bioinformatics analyses using QIIME. Both α and β diversity analyses showed that ceftriaxone sodium, cefoperazone/sulbactam, meropenem and vancomycin changed the gut microbiota dramatically by the second day of antibiotic administration whereas the influence of ofloxacin was trivial. Azithromycin clearly changed the gut microbiota but much less than vancomycin and the β-lactams. In general, the community changes induced by the three β-lactam antibiotics showed consistency in inhibiting Papillibacter, Prevotella and Alistipes while inducing massive growth of Clostridium. The low diversity and high Clostridium level might be an important cause of Clostridium difficile infection after usage of β-lactams. Vancomycin was unique in that it inhibited Firmicutes, mainly the genus Clostridium. On the other hand, it induced the growth of Escherichia and effect lasted for months afterward. Azithromycin and meropenem induced the growth of Enterococcus. These findings will be useful for understanding the potential adverse effects of antibiotics on the gut microbiome and ensuring their better usage. [ABSTRACT FROM AUTHOR]

Notch signaling pathway defines an evolutionarily conserved mechanism in cell-fate determination in a broad spectrum of developmental processes through local cell interactions. mind bomb (mib) encodes an E3 ubiquitin ligase that is involved in Notch activation through Delta ubiquitylation and internalization. To further dissect the function of Mib, two yeast two-hybrid screens for zebrafish Mib/Mib2-binding proteins with different strategies have been performed. 81 putative interesting proteins were discovered and classified into six groups: ubiquitin proteasome pathway, cytoskeleton, trafficking, replication/transcription/translation factors, cell signaling and others. Confirmed by coimmunoprecipitation (Co-IP), Mib interacted with four tested proteins: ubiquitin specific protease 1 (Usp1), ubiquitin specific protease 9 (Usp9), tumor-necrosis-factor-receptor-associated factor (TRAF)-binding domain (Trabid)/zinc finger, RAN-binding domain containing 1 (Zranb1) and hypoxia-inducible factor 1, alpha subunit inhibitor (Hif1an)/factor inhibiting HIF 1 (Fih-1). Usp1, Usp9, Trabid and Fih-1 also bound to zebrafish Mib2, a Mib homolog with similar structural domains and functions. Both Mib and Mib2 can ubiquitylate Trabid and Fih-1, indicating a potential regulating role of Mib and Mib2 on Trabid and Fih-1 and, furthermore, the possible involvement of Notch signaling in hypoxia-regulated differentiation, tumorigenesis and NF-κB pathway. Finally, functions of confirmed Mib/Mib2-interacting proteins are collated, summarized and hypothesized, which depicts a regulating network beyond Notch signaling. [ABSTRACT FROM AUTHOR]

Background: There are four cell lineages derived from intestinal stem cells that are located at the crypt and villus in the mammalian intestine the non-secretory absorptive enterocytes, and the secretory cells, which include mucous-secreting goblet cells, regulatory peptide-secreting enteroendocrine cells and antimicrobial peptide-secreting Paneth cells. Although fibroblast growth factor (Fgf) signaling is important for cell proliferation and differentiation in various tissues, its role in intestinal differentiation is less well understood. Methodology/Principal Findings: We used a loss of function approach to investigate the importance of Fgf signaling in intestinal cell differentiation in zebrafish; abnormal differentiation of goblet cells was observed when Fgf signaling was inhibited using SU5402 or in the Tg(hsp70ldnfgfr1-EGFP) transgenic line. We identified Fgfr2c as an important receptor for cell differentiation. The number of goblet cells and enteroendocrine cells was reduced in fgfr2c morphants. In addition to secretory cells, enterocyte differentiation was also disrupted in fgfr2c morphants. Furthermore, proliferating cells were increased in the morphants. Interestingly, the loss of fgfr2c expression repressed secretory cell differentiation and increased cell proliferation in the mibta52b mutant that had defective Notch signaling. Conclusions/Significance: In conclusion, we found that Fgfr2c signaling derived from mesenchymal cells is important for regulating the differentiation of zebrafish intestine epithelial cells by promoting cell cycle exit. The results of Fgfr2c knockdown in mibta52b mutants indicated that Fgfr2c signaling is required for intestinal cell differentiation. These findings provide new evidences that Fgf signaling is required for the differentiation of intestinal cells in the zebrafish developing gut. [ABSTRACT FROM AUTHOR]

Radiotherapy is the primary treatment modality used for patients with head-and-neck cancers, but inevitably causes microorganism-related oral complications. This study aims to explore the dynamic core microbiome of oral microbiota in supragingival plaque during the course of head-and-neck radiotherapy. Eight subjects aged 26 to 70 were recruited. Dental plaque samples were collected (over seven sampling time points for each patient) before and during radiotherapy. The V1–V3 hypervariable regions of bacterial 16S rRNA genes were amplified, and the high-throughput pyrosequencing was performed. A total of 140 genera belonging to 13 phyla were found. Four phyla (Actinobacteria, Bacteroidetes, Firmicutes, and Proteobacteria) and 11 genera (Streptococcus, Actinomyces, Veillonella, Capnocytophaga, Derxia, Neisseria, Rothia, Prevotella, Granulicatella, Luteococcus, and Gemella) were found in all subjects, supporting the concept of a core microbiome. Temporal variation of these major cores in relative abundance were observed, as well as a negative correlation between the number of OTUs and radiation dose. Moreover, an optimized conceptual framework was proposed for defining a dynamic core microbiome in extreme conditions such as radiotherapy. This study presents a theoretical foundation for exploring a core microbiome of communities from time series data, and may help predict community responses to perturbation as caused by exposure to ionizing radiation. [ABSTRACT FROM AUTHOR]