Dermoscopic image segmentation based on Pyramid Residual Attention Module.

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.

Gated Skip-Connection Network with Adaptive Upsampling for Retinal Vessel Segmentation.

Segmentation of retinal vessels is a critical step for the diagnosis of some fundus diseases. Methods: To further enhance the performance of vessel segmentation, we propose a method based on a gated skip-connection network with adaptive upsampling (GSAU-Net). In GSAU-Net, a novel skip-connection with gating is first utilized in the extension path, which facilitates the flow of information from the encoder to the decoder. Specifically, we used the gated skip-connection between the encoder and decoder to gate the lower-level information from the encoder. In the decoding phase, we used an adaptive upsampling to replace the bilinear interpolation, which recovers feature maps from the decoder to obtain the pixelwise prediction. Finally, we validated our method on the DRIVE, CHASE, and STARE datasets. Results: The experimental results showed that our proposed method outperformed some existing methods, such as DeepVessel, AG-Net, and IterNet, in terms of accuracy, F-measure, and AUCROC. The proposed method achieved a vessel segmentation F-measure of 83.13%, 81.40%, and 84.84% on the DRIVE, CHASE, and STARE datasets, respectively.

MFI-Net: A multi-resolution fusion input network for retinal vessel segmentation.

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.

Genetic characterization of novel fowl aviadenovirus 4 isolates from outbreaks of hepatitis-hydropericardium syndrome in broiler chickens in China.

Since May 2015, severe outbreaks of hepatitis-hydropericardium syndrome (HHS) associated with infections of fowl aviadenovirus (FAdV) have emerged in broiler chickens in several Chinese provinces. To identify the genotype and gain a better understanding of the genetic properties of the FAdV strains responsible for the recent HHS outbreaks in China, the complete genome sequences of five isolates from outbreaks of HHS in broiler chickens in five provinces were determined. The results demonstrated that a novel fowl aviadenovirus 4 (FAdV-4) genotype was epidemic in China. To investigate the molecular characteristics of these Chinese FAdV-4 isolates, their genome contents were compared with those of reported pathogenic and non-pathogenic FAdV-4 strains. The comparative analysis revealed that the novel Chinese FAdV-4 isolates contain various genomic deletions and multiple distinct amino-acid mutations in their major structural genes. Two additional putative genetic virulence markers in the fiber 2 gene were identified. These findings confirmed some of the genetic differences between the pathogenic and non-pathogenic FAdV-4 isolates. The data presented in this report will enhance the current understanding of the molecular epidemiology and genetic diversity of FAdV-4 isolates in China and will provide additional insight into the critical factors that determine the pathogenicity of FAdV-4 strains. Finally, the emergence of this novel and highly pathogenic FAdV-4 genotype emphasizes that preventive measures against FAdV-4 infections on poultry farms should be implemented in China.

[Isolation, identification and full-length genome sequence analysis of encephalomyocarditis virus from local aardvarks].

Encephalomyocarditis virus (EMCV) is a natural epidemic zoonotic pathogen. However, no reports have been published regarding the isolation, identification and full-length genome of EMCV from a local aardvark population. In present study, an EMCV isolate HNXX13 was isolated from aardvarks named Huainan-pig in Henan Province. The systematic identification, full-length genome sequencing and molecular characteristic analysis of the isolate HNXX13 were conducted. The result showed that the isolate was spherical with a diameter of 24-30 nm, neither heat- nor acid-resistant, sensitive to trypsin, insensitive to chloroform, not protected by bivalent cationic, and the specific fluorescence was observed in the cytoplasm of BHK-21 cells infected with the isolate by using indirect fluorescence assay. The full-length genome of EMCV HNXX13 generated a 7 725bp sequence (GenBank: F771002), with 81.0%-99.9% nucleotide identity to reference strains from different animals, and 99.5% with a Chinese reference strain isolated earlier from a commercial pig herd. The phylogenetic tree based on the full-length genome and ORF sequences identified that all EMCV strains were divided into three groups G1, G2 and G3, and strain HNXX13 belonging to the G1 group with other Chinese reference strains. The result also identified that this EMCV infection could cause severe clinical signs in a local aardvark population, and enriches the molecular epidemiological data of EMCV in China. Regional differences exist in EMCV genome and transmission is limited within a certain area. However, the cross-infection and transmission of EMCV between aardvark and mice appears most likely. Mutations have occurred in some amino acids of EMCV strain HNXX13 during the transmission in local aardvark herd and these mutations might make the virus easier to infect the aardvark.

[Investigation of etiology of massive infection with porcine pseudorabies virus in Henan and neighboring Provinces].

In early 2011, the serious outbreak of porcine pseudorabies virus (PRV) infection suddenly recurred in Henan and neighboring Provinces. To investigate the etiology of massive infection with PRV, 16 800 serum samples, 905 porcine epidemic diarrhea virus (PEDV) back-feeding tissues, and 56 PR gene deleted live vaccines were colleted from January 2011 to May 2013 to detect PRV field infection using a PRV gE antibody test kit. The gE and TK genes of 11 new epidemic PRV strains were sequenced by PCR, and their molecular characteristics were analyzed. Moreover, virus titer determination, protective test against PRV, and vaccine potency testing were performed. The results showed that the detection rate of PRV field infection-positive pig farms was 68.06%, and the overall positive rate of PRV field infection in serum was 38.47%; the positive rates in breeding sows, breeding boars, reserve pigs, and commercial pigs were 40.12%, 30.88%, 54.67%, and 26.52%, respectively. The new epidemic strains were in the same evolutionary branch and belonged to the virulent strain group. Compared with the classical PRV strain, the virulence of new epidemic strains changed a little. The length of gE gene was 1 787 bp, and the length of TK gene was 963 bp. The nucleotide homologies of gE and TK genes to Chinese reference strains were 98.2%-99.8% and 98.90%-99.6%, respectively, and the amino acid homologies were 97.1%-99.8% and 97.5%-99.4%, respectively. Commercial vaccine had a 100% protective effect against the new epidemic strains. The positive rate of PRV field infection was 0% in vaccine and 40.44% in back-feeding tissues. The results confirmed that PRV field infection rates were rising sharply among pigs in Henan and neighboring Provinces after 2011. The main virulence genes of new epidemic PRV strains did not change significantly over the years. PR gene deleted live vaccines had no PRV field infection and could completely resist the attack of new strains. The virus carriage of breeding boars and reserve pigs and the serious PRV field infection in PEDV back-feeding tissues were the main causative factors for massive infection with PRV and epidemic outbreak in Henan and neighboring Provinces from 2011 to 2013.

Complete genome sequence of porcine encephalomyocarditis virus from an aardvark in china.

A strain of encephalomyocarditis virus, HNXX13, was isolated from an aardvark in central China. The complete genome was sequenced and analyzed, and phylogenetic analysis suggests that HNXX13 belongs to encephalomyocarditis virus group 1.

Development and preliminary application of a quantitative PCR assay for detecting gtxA-containing Gallibacterium species in chickens.

A quantitative PCR (qPCR) assay using SYBR Green I was developed based on the published sequence of the gtxA gene from Gallibacterium anatis. This method produced reliable specificity, sensitivity, and repeatability. The detection rate of Gallibacterium in 181 clinical samples was 36.5% (66/181) by qPCR, which was superior to the detection rate of Gallibacterium-specific PCR (0/181) and an isolation and identification assay (18.2% or 33/181). No association was found between the prevalence of Gallibacterium and the age of the chickens. Gallibacterium infection was detected in one 4-day-old chicken, showing that infection can occur much earlier than the previously stated fourth week of life. Tissue sample analysis showed that Gallibacterium is mainly located in the trachea and ovaries, based on results from three groups of chicken with different health statuses. Furthermore, a titer analysis suggested that Gallibacterium loads in different organs may correlate with different clinical manifestations of disease. Thus, the qPCR assay developed in the present study is useful for identification and quantitative analysis of gtxA-containing Gallibacterium in various tissue samples from birds and for the assessment of the pathogenic mechanisms of Gallibacterium.

Prevalence and molecular characterization of Cyclospora cayetanensis, Henan, China.

To determine prevalence of Cyclospora cayetanensis infection in Henan, China, we conducted a study of 11,554 hospital patients. Prevalence was 0.70% (95% confidence interval 0.70% ± 0.15%), with all age groups infected. Most cases were found in the summer. Minor sequence polymorphisms were observed in the 18S rRNA gene of 35 isolates characterized.

Immune responses of chickens inoculated with a recombinant fowlpox vaccine coexpressing HA of H9N2 avain influenza virus and chicken IL-18.

Control of the circulation of H9N2 avian influenza virus (AIV) is a major concern for both animal and public health, and H9N2 AIV poses a major threat to the chicken industry worldwide. Here, we developed a recombinant fowlpox virus (rFPV-HA) expressing the haemagglutinin (HA) gene of the A/CH/JY/1/05 (H9N2) influenza virus and a recombinant fowlpox virus (rFPV-HA/IL18) expressing the HA gene and chicken interleukin-18 (IL-18) gene. Recombinant plasmid pSY-HA/IL18 was constructed by cloning chicken IL-18 expression cassette into recombinant plasmid pSY-HA containing the HA gene. Two rFPVs were generated by transfecting two recombinant plasmids into the chicken embryo fibroblast cells pre-infected with S-FPV-017, and assessed for their immunological efficacy on one-day-old White Leghorn specific-pathogen-free chickens challenged with the A/CH/JY/1/05 (H9N2) strain. There was a significant difference in HI antibody levels (P<0.05) elicited by either rFPV-HA or rFPV-HA/IL18. The level of splenocyte proliferation response in the rFPV-HA/IL18-vaccinated group was significantly higher (P<0.05) than that in the rFPV-HA group. After challenge with 10(6.5)ELD(50) H9N2 AIV 43days after immunization, rFPVs vaccinated groups could prevent virus shedding and replication in multiple organs in response to H9N2 AIV infection, and rFPV-HA/IL18 vaccinated group had better inhibition of viruses than rFPV-HA vaccinated group. Our results show that the protective efficacy of the rFPV-HA vaccine could be enhanced significantly by simultaneous expression of IL-18.