Deep spatial-spectral prior with an adaptive dual attention network for single-pixel hyperspectral reconstruction.

Recently, single-pixel imaging has shown great promise in developing cost-effective imaging systems, where coding and reconstruction are the keys to success. However, it also brings challenges in capturing hyperspectral information accurately and instantly. Many works have attempted to improve reconstruction performance in single-pixel hyperspectral imaging by applying various hand-crafted priors, leading to sub-optimal solutions. In this paper, we present the deep spatial-spectral prior with adaptive dual attention network for single-pixel hyperspectral reconstruction. Specifically, the spindle structure of the parameter sharing method is developed to integrate information across spatial and spectral dimensions of HSI, which can synergistically and efficiently extract global and local prior information of hyperspectral images from both shallow and deep layers. Particularly, a sequential adaptive dual attention block (SADAB), i.e., spatial attention and spectral attention, are devised to adaptively rescale informative features of spatial locations and spectral channels simultaneously, which can effectively boost the reconstruction accuracy. Experiment results on public HSI datasets demonstrate that the proposed method significantly outperforms the state-of-the-art algorithm in terms of reconstruction accuracy and speed.

The efficient circulating immunoscore predicts prognosis of patients with advanced gastrointestinal cancer.

BACKGROUND: Immunoscore from tumor tissues was initially established to evaluate the prognosis of solid tumor patients. However, the feasibility of circulating immune score (cIS) for the prognosis of advanced gastrointestinal cancers (AGC) has not been reported. MATERIAL AND METHODS: Peripheral venous blood was collected from 64 untreated AGC patients. We utilized flow cytometry to determine several immune cell subpopulations, including CD8+ and CD4+ T cells, NK cells, and CD4 + CD25 + CD127low Tregs. The circulating immune score 1 (cIS1) was assessed according to the proportions of CD4+, CD8+ T cells, and NK cell, whereas circulating immune score 2 (cIS2) was derived from the proportions of CD4+, CD8+ T cell, and CD4 + CD25 + CD127low Tregs. The prognostic role of cIS for progression-free survival (PFS) and overall survival (OS) was analyzed using Kaplan-Meier curves and Cox multivariate models. Receiver operating characteristic (ROC) curves were depicted to compare the prognostic values of cIS1 and cIS2. RESULTS: AGC patients with high cIS1(≥ 2) and cIS2(≥ 2) had significantly longer PFS (cIS1: median PFS, 11 vs. 6.7 months, P = 0.001; cIS2: 12 vs. 5.8 months, P < 0.0001) and OS (cIS1: median OS, 12 vs. 7.9 months, P = 0.0004; cIS2: 12.8 vs. 7.4 months, P < 0.0001) than those with low cIS1 and low cIS2. The areas under ROC curves (AUROCs) of cIS1 and cIS2 for OS were 0.526 (95% confidence interval; 95% CI 0.326-0.726) and 0.603 (95% CI 0.427-0.779, P = 0.332), whereas AUROC of cIS2 for PFS was larger than that of cIS1 0.735 (95% CI 0.609-0.837) vs 0.625 (95% CI 0.495-0.743) (P = 0.04)). CONCLUSION: The cIS can be applied to predict the prognosis of untreated AGC patients. Compared with cIS1, cIS2 displayed superior prognostic value for PFS prediction.

Structural Similarity Loss for Learning to Fuse Multi-Focus Images.

Convolutional neural networks have recently been used for multi-focus image fusion. However, some existing methods have resorted to adding Gaussian blur to focused images, to simulate defocus, thereby generating data (with ground-truth) for supervised learning. Moreover, they classify pixels as 'focused' or 'defocused', and use the classified results to construct the fusion weight maps. This then necessitates a series of post-processing steps. In this paper, we present an end-to-end learning approach for directly predicting the fully focused output image from multi-focus input image pairs. The suggested approach uses a CNN architecture trained to perform fusion, without the need for ground truth fused images. The CNN exploits the image structural similarity (SSIM) to calculate the loss, a metric that is widely accepted for fused image quality evaluation. What is more, we also use the standard deviation of a local window of the image to automatically estimate the importance of the source images in the final fused image when designing the loss function. Our network can accept images of variable sizes and hence, we are able to utilize real benchmark datasets, instead of simulated ones, to train our network. The model is a feed-forward, fully convolutional neural network that can process images of variable sizes during test time. Extensive evaluation on benchmark datasets show that our method outperforms, or is comparable with, existing state-of-the-art techniques on both objective and subjective benchmarks.

Self-Supervised Learning to Detect Key Frames in Videos.

Detecting key frames in videos is a common problem in many applications such as video classification, action recognition and video summarization. These tasks can be performed more efficiently using only a handful of key frames rather than the full video. Existing key frame detection approaches are mostly designed for supervised learning and require manual labelling of key frames in a large corpus of training data to train the models. Labelling requires human annotators from different backgrounds to annotate key frames in videos which is not only expensive and time consuming but also prone to subjective errors and inconsistencies between the labelers. To overcome these problems, we propose an automatic self-supervised method for detecting key frames in a video. Our method comprises a two-stream ConvNet and a novel automatic annotation architecture able to reliably annotate key frames in a video for self-supervised learning of the ConvNet. The proposed ConvNet learns deep appearance and motion features to detect frames that are unique. The trained network is then able to detect key frames in test videos. Extensive experiments on UCF101 human action and video summarization VSUMM datasets demonstrates the effectiveness of our proposed method.

Single Infrared Image-Based Stripe Nonuniformity Correction via a Two-Stage Filtering Method.

The presence of stripe nonuniformity severely degrades the image quality and affects the performance in many infrared (IR) sensing applications. Prior works correct the nonuniformity by using similar spatial representations, which inevitably damage some detailed structures of the image. In this paper, we instead take advantage of spectral prior of stripe noise to solve its correction problem in single IR image. We first analyse the significant spectral difference between stripes and image structures and utilize this knowledge to characterize stripe nonuniformity. Then a two-stage filtering strategy is adopted combining spectral and spatial filtering. The proposed method enables stripe nonuniformity to be eliminated from coarse to fine, thus preserving image details well. Extensive experiments on simulated images and raw IR images demonstrate that the proposed method achieves superior correction performance over the recent state-of-the-art methods.

Multi-focus image fusion based on dictionary learning with rolling guidance filter.

We present a new multi-focus image fusion method based on dictionary learning with a rolling guidance filter to fusion of multi-focus images with registration and mis-registration. First, we learn a dictionary via several classical multi-focus images blurred by a rolling guidance filter. Subsequently, we present a new model for focus regions identification via applying the learned dictionary to input images to obtain the corresponding focus feature maps. Then, we determine the initial decision map via comparing the difference of the focus feature maps. The latter is to optimize the initial decision map and perform it on input images to obtain fused images. Experimental results demonstrate that the suggested algorithm is competitive with the current state of the art and superior to some representative methods when input images are well registered and mis-registered.

Multi-focus image fusion using a guided-filter-based difference image.

The aim of multi-focus image fusion technology is to integrate different partially focused images into one all-focused image. To realize this goal, a new multi-focus image fusion method based on a guided filter is proposed and an efficient salient feature extraction method is presented in this paper. Furthermore, feature extraction is primarily the main objective of the present work. Based on salient feature extraction, the guided filter is first used to acquire the smoothing image containing the most sharpness regions. To obtain the initial fusion map, we compose a mixed focus measure by combining the variance of image intensities and the energy of the image gradient together. Then, the initial fusion map is further processed by a morphological filter to obtain a good reprocessed fusion map. Lastly, the final fusion map is determined via the reprocessed fusion map and is optimized by a guided filter. Experimental results demonstrate that the proposed method does markedly improve the fusion performance compared to previous fusion methods and can be competitive with or even outperform state-of-the-art fusion methods in terms of both subjective visual effects and objective quality metrics.

Infrared and visible image fusion with spectral graph wavelet transform.

Infrared and visible image fusion technique is a popular topic in image analysis because it can integrate complementary information and obtain reliable and accurate description of scenes. Multiscale transform theory as a signal representation method is widely used in image fusion. In this paper, a novel infrared and visible image fusion method is proposed based on spectral graph wavelet transform (SGWT) and bilateral filter. The main novelty of this study is that SGWT is used for image fusion. On the one hand, source images are decomposed by SGWT in its transform domain. The proposed approach not only effectively preserves the details of different source images, but also excellently represents the irregular areas of the source images. On the other hand, a novel weighted average method based on bilateral filter is proposed to fuse low- and high-frequency subbands by taking advantage of spatial consistency of natural images. Experimental results demonstrate that the proposed method outperforms seven recently proposed image fusion methods in terms of both visual effect and objective evaluation metrics.

Influence of humic acid on the bioaccumulation, elimination, and toxicity of PFOS and TBBPA co-exposure in Mytilus unguiculatus Valenciennes.

Tetrabromobisphenol A (TBBPA) and Perfluorooctane sulfonate (PFOS) are emerging contaminants which coexist in marine environments, posing significant risks to ecosystems and human health. The behavior of these contaminants in the presence of dissolved organic matter (DOM), specifically the co-contamination of TBBPA and PFOS, is not well understood. The bioaccumulation, distribution, elimination, and toxic effects of TBBPA and PFOS on thick-shell mussels (Mytilus unguiculatus V.), with the absence and presence of humic acid (HA), a typical DOM, were studied. The results showed that the uptake of TBBPA decreased and the uptake of PFOS increased when exposed to 1 mg/L HA. However, at higher concentrations of HA (5 and 25 mg/L), the opposite effect was observed. Combined exposure to HA, TBBPA, and PFOS resulted in oxidative stress in the digestive gland, with the severity of stress dependent on exposure time and HA dose. Histological analysis revealed a positive correlation between HA concentration and tissue damage caused by TBBPA and PFOS. This study provides insights into the influence of HA on the bioaccumulation-elimination patterns and toxicity of TBBPA and PFOS in marine bivalves, offering valuable data for ecological and health risk assessments of combined pollutants in aquatic environments rich in DOM.

Insights into the combined toxicity and mechanisms of BDE-47 and PFOA in marine blue mussel: An integrated study at the physiochemical and molecular levels.

The coexistence of multiple emerging contaminants imposes a substantial burden on the ecophysiological functions in organisms. The combined toxicity and underlying mechanism requires in-depth understanding. Here, marine blue mussel (Mytilus galloprovincialis L.) was selected and exposed to 2,2',4,4'-tetrabromodiphenyl ether (BDE-47) and perfluorooctanoic acid (PFOA) individually and in combination at environmental related concentrations to elucidate differences in stress responses and potential toxicological mechanisms. Characterization and comparison of accumulation, biomarkers, histopathology, transcriptomics and metabolomics were performed. Co-exposure resulted in differential accumulation patterns, exacerbated histopathological alterations, and different responses in oxidative stress and biomarkers for xenobiotic transportation. Moreover, the identified differentially expressed genes (DEGs) and differential metabolites (DEMs) in mussels were found to be annotated to different metabolic pathways. Correlation analyses further indicated that DEGs and DEMs were significantly correlated with the above biomarkers. BDE-47 and PFOA altered the genes and metabolites related to amino acid metabolism, energy and purine metabolism, ABC transporters, and glutathione metabolism to varying degrees, subsequently inducing accumulation differences and combined toxicity. Furthermore, the present work highlighted the pivotal role of Nrf2-keap1 detoxification pathway in the acclimation of M. galloprovincialis to reactive oxygen species (ROS) stress induced by BDE-47 and PFOA. This study enabled more comprehensive understanding of combined toxic mechanism of multi emerging contaminants pollution.

Meta-analysis of the relationship between microRNA-499 rs3746444 polymorphism and hepatocellular carcinoma risk in Asians.

OBJECTIVE: Numerous studies have reported the role of microRNA-499 rs3746444 polymorphism with hepatocellular carcinoma. (HCC) risk in Asians, but the specific association is still controversial. MATERIALS AND METHODS: To derive a more precise estimation of the relationship, a meta-analysis was performed. An electronic search of PubMed database was conducted to select relevant studies. Odds ratios. (ORs) with 95% confidence intervals. (CIs) were estimated to assess the relationship between microRNA-499 rs3746444 polymorphism and HCC risk in Asians. RESULTS: Eight relevant studies including 2196 cases and 2701 controls were included in this meta-analysis. When all the eligible studies were pooled into this meta-analysis, no significant association of HCC risk with microRNA-499 rs3746444 polymorphism was found in Asians (TT vs. CC: OR = 1.00 95% CI = 0.54-1.83; TT vs. TC: OR = 0.89, 95% CI = 0.66-1.20; dominant model: OR = 1.13, 95% CI = 0.81-1.58; recessive model: OR = 0.99, 95% CI = 0.56-1.76). CONCLUSION: This meta-analysis suggests that the microRNA-499 rs3746444 polymorphism is not contributed to the risk of HCC in Asians. However, more studies should be performed in the future to confirm this association.