RESUMO
Pigeon Newcastle disease (ND) is a serious infectious illness caused by the pigeon Newcastle disease virus (NDV) or Paramyxovirus type 1 (PPMV-1). Genotype VI NDV is a primary factor in ND among Columbiformes (such as pigeons and doves). In a recent study, eight pigeon NDV strains were discovered in various provinces in China. These viruses exhibited mesogenic characteristics based on their MDT and ICPI values. The complete genome sequences of these eight strains showed a 90.40% to 99.19% identity match with reference strains of genotype VI, and a 77.86% to 80.45% identity match with the genotype II vaccine strain. Additionally, analysis of the F gene sequence revealed that these NDV strains were closely associated with sub-genotypes VI.2.2.2, VI.2.1.1.2.1, and VI.2.1.1.2.2. The amino acid sequence at the cleavage site of the F protein indicated virulent characteristics, with the sequences 112KRQKRF117 and 112RRQKRF117 observed. Pigeons infected with these sub-genotype strains had a low survival rate of only 20% to 30%, along with lesions in multiple tissues, highlighting the strong spread and high pathogenicity of these pigeon NDV strains. Molecular epidemiology data from the GenBank database revealed that sub-genotype VI.2.1.1.2.2 strains have been prevalent since 2011. In summary, the findings demonstrate that the prevalence of genotype VI NDV is due to strains from diverse sub-genotypes, with the sub-genotype VI.2.1.1.2.2 strain emerging as the current epidemic strain, highlighting the significance of monitoring pigeon NDV in China.
RESUMO
Recent efforts on learning-based image denoising approaches use unrolled architectures with a fixed number of repeatedly stacked blocks. However, due to difficulties in training networks corresponding to deeper layers, simply stacking blocks may cause performance degradation, and the number of unrolled blocks needs to be manually tuned to find an appropriate value. To circumvent these problems, this paper describes an alternative approach with implicit models. To our best knowledge, our approach is the first attempt to model iterative image denoising through an implicit scheme. The model employs implicit differentiation to calculate gradients in the backward pass, thus avoiding the training difficulties of explicit models and elaborate selection of the iteration number. Our model is parameter-efficient and has only one implicit layer, which is a fixed-point equation that casts the desired noise feature as its solution. By simulating infinite iterations of the model, the final denoising result is given by the equilibrium that is achieved through accelerated black-box solvers. The implicit layer not only captures the non-local self-similarity prior for image denoising, but also facilitates training stability and thereby boosts the denoising performance. Extensive experiments show that our model leads to better performances than state-of-the-art explicit denoisers with enhanced qualitative and quantitative results.