Recently, existing FCNs‐based methods have shown their advantages in processing object boundaries. However, these methods still suffer from false object interference, which appears in saliency ...predictions. To solve this problem, an edge‐interior feature fusion (EIFF) framework is proposed, which consists of an internal‐boundary decoupled generation structure with receptive field enlargement and attention mechanism enhancement, and a salient feature refinement module. Specifically, the framework first learns edge features and interior features through an internal‐boundary decoupling generation network, which is supervised by labels obtained by decoupling ground‐truth through an image erosion algorithm. Then, feature refinement module (FRM) is designed to purify the coarse prediction by focusing on the ambiguous regions through a mining strategy to generate the final saliency map. To compensate for shortcomings of the BCE and IU loss, we also introduce a weighted loss to guide our model to focus more on the error‐prone parts. Experimental results on five benchmark datasets demonstrate that the proposed method performs favorably against 19 state‐of‐the‐art approaches under four standard metrics.
Introduction
Ixekizumab, a monoclonal antibody against interleukin-17A, demonstrated effectiveness in the treatment of psoriasis in a Chinese real-world study that was consistent with previous ...randomized controlled trials. Here, we report further analyses from this study to explore the effectiveness of ixekizumab for treating patients with psoriasis and the involvement of special body areas (scalp, nail, joint, palmoplantar, or genital areas).
Methods
A multicenter, prospective, observational, single-arm, post-marketing surveillance study was conducted in patients aged ≥ 18 years with moderate-to-severe plaque psoriasis and prescribed with ixekizumab in 26 Chinese hospitals. Psoriasis Area and Severity Index (PASI) and Dermatology Life Quality Index (DLQI) scores were compared between patients with versus without psoriasis in special body areas in the overall study population and across subgroups by body area.
Results
In total, 612 patients were included. At baseline, most patients (93.6%) had psoriasis involvement in at least one special body area. Overall, patients with psoriasis in special body areas reported a worse quality of life (QoL) than those without. Patients with versus without psoriasis in special body areas achieved a comparable mean reduction from baseline in PASI score (10.9 vs. 9.2 at week 2, and 16.9 vs. 14.7 at week 12, respectively) and DLQI score (6.0 vs. 4.4 at week 2, and 9.9 vs. 7.5 at week 12, respectively); a similar proportion of patients also achieved PASI 50 at week 2, and PASI 75 and PASI 90 at week 12, and a DLQI (0/1) at weeks 2 and 12. Several significantly different results were reported between subgroups, the majority of which favored patients with special body area involvement.
Conclusion
Most patients had psoriasis involvement in a special body area which was associated with worse QoL. Ixekizumab is similarly effective in reducing disease severity and improving QoL in patients with plaque psoriasis across different special body areas.
Our previous genome-wide linkage analysis identified a susceptibility locus for generalized vitiligo on 22q12. To search for susceptibility genes within the locus, we investigated a biological ...candidate gene, X-box binding protein 1(XBP1). First, we sequenced all the exons, exon-intron boundaries as well as some 5' and 3' flanking sequences of XBP1 in 319 cases and 294 controls of Chinese Hans. Of the 8 common variants identified, the significant association was observed at rs2269577 (p_(trend) = 0.007, OR = 1.36, 95% CI = 1.09-1.71), a putative regulatory polymorphism within the promoter region of XBP1. We then sequenced the variant in an additional 365 cases and 404 controls and found supporting evidence for the association (p_(trend) = 0.008, OR = 1.31, 95% CI = 1.07-1.59). To further validate the association, we genotyped the variant in another independent sample of 1,402 cases and 1,288 controls, including 94 parent-child trios, and confirmed the association by both case-control analysis (p_(trend) = 0.003, OR = 1.18, 95% CI = 1.06-1.32) and the family-based transmission disequilibrium test (TDT, p = 0.005, OR = 1.93, 95% CI = 1.21-3.07). The analysis of the combined 2,086 cases and 1,986 controls provided highly significant evidence for the association (p_(trend) = 2.94x10(-6), OR = 1.23, 95% CI = 1.13-1.35). Furthermore, we also found suggestive epistatic effect between rs2269577 and HLA-DRB1*07 allele on the development of vitiligo (p = 0.033). Our subsequent functional study showed that the risk-associated C allele of rs2269577 had a stronger promoter activity than the non-risk G allele, and there was an elevated expression of XBP1 in the lesional skins of patients carrying the risk-associated C allele. Therefore, our study has demonstrated that the transcriptional modulation of XBP1 expression by a germ-line regulatory polymorphism has an impact on the development of vitiligo.
Video prediction is a meaningful task for it has a wide range of application scenarios. And it is also a challenging task since it needs to learn the internal representation of a given video for both ...appearance and motion dynamics. The existing methods regard this problem as a spatiotemporal sequence forecasting problem and try to resolve it in a one-shot fashion, which causes the prediction result being blurry or inaccurate. So, a more intuitional thought is to split this problem into two parts: model the dynamic pattern of the given video and learn the appearance representation of given video frames. In this paper, we develop a novel network structure named MixPred based on this idea to address this issue. We divide the prediction problem into two parts as mentioned above and build two subnets to solve these two parts separately. Instead of fusing the results of subnets at the final layer, we put forward a parallel interaction style through the whole process to merge the dynamic information and content information in a more natural way. Besides, we propose three different connection methods for exploring the most effective connection structure. We trained the model on UCF-101 and KITTI, and testing our model on UCF-101, KITTI, and Caltech. The results demonstrate that our method achieves state-of-the-art both quantitatively and qualitatively.
Video saliency prediction has received increasing attention in the field of computer vision research. How to model the spatio-temporal information in video frames is a key issue for saliency ...prediction. Most of the existing methods either show a limited temporal memory ability or fail in jointly processing spatial and temporal information. Besides, these methods take each pixel of video frames as input indiscriminately, which may result in incorrect results because of redundant pixels. To solve the above problems, we propose a novel Explicit Cyclic Attention-based Network (ECANet), which is a two-stream encoder-decoder model. Specifically, an explicit cyclic attention mechanism is proposed for temporal modeling and pixel emphasizing, in which an attentional stream is built, whose input is cropped patches based on the model’s previous prediction. For spatio-temporal information aggregation, 3D convolutional neural network is adopted as the backbone of our encoders and decoder. Stacks of video frames and attentional patches from past steps build up the input of our model. Extensive experimental results over three benchmark datasets show that our model achieves competitive performance against state-of-the-art methods.
In the transportation sector, new energy vehicles (NEVs) are critical to reduce CO2 emissions in the context of carbon neutralization. The study of dynamic evaluation and regional difference analysis ...is helpful to the NEV industry development in policy design and industrial planning. In this study, based on the provincial data in China from 2016 to 2020, the grey target model and Dagum Gini coefficient method are employed for the dynamic evaluation and regional differences of the NEV industry development. The results were as follows: (1) The overall and provincial level of the NEV industry development showed an increasing pattern. The bull’s eye degrees of Guangdong, which had the best development, were 0.4884, 0.5361, 0.6067, 0.6787, and 0.7047 during the study period. (2) The regional differences in the NEV industry development were significant. The east region had the best development, followed by the middle, the west, and the northeast regions. The intra-regional differences were expanding with different annual growth rates. (3) The inter-regional differences between the east and the other three regions were the largest. The regional differences in the NEV development are mainly derived from inter-regional dereference. (4) The D1, D2, and D3 dimensions all contributed significantly to provinces with higher levels of development, while the D4 dimension contributed significantly to provinces with lower levels of development. Based on these results, different provinces should take differentiated development strategies and enhancement paths to promote their NEV industry development.