Learning Patch-Channel Correspondence for Interpretable Face Forgery Detection.

Beyond high accuracy, good interpretability is very critical to deploy a face forgery detection model for visual content analysis. In this paper, we propose learning patch-channel correspondence to facilitate interpretable face forgery detection. Patch-channel correspondence aims to transform the latent features of a facial image into multi-channel interpretable features where each channel mainly encoders a corresponding facial patch. Towards this end, our approach embeds a feature reorganization layer into a deep neural network and simultaneously optimizes classification task and correspondence task via alternate optimization. The correspondence task accepts multiple zero-padding facial patch images and represents them into channel-aware interpretable representations. The task is solved by step-wisely learning channel-wise decorrelation and patch-channel alignment. Channel-wise decorrelation decouples latent features for class-specific discriminative channels to reduce feature complexity and channel correlation, while patch-channel alignment then models the pairwise correspondence between feature channels and facial patches. In this way, the learned model can automatically discover corresponding salient features associated to potential forgery regions during inference, providing discriminative localization of visualized evidences for face forgery detection while maintaining high detection accuracy. Extensive experiments on popular benchmarks clearly demonstrate the effectiveness of the proposed approach in interpreting face forgery detection without sacrificing accuracy. The source code is available at https://github.com/Jae35/IFFD.

Xuesaitong injection treating acute myocardial infarction: A systematic review and meta-analysis.

BACKGROUND: Although the incidence of acute myocardial infarction (AMI) is decreasing, the mortality in AMI patients remains substantial. Traditional Chinese medicine has shown its role in the prevention and management of AMI. The purpose of this study is to evaluate the clinical efficacy of Xuesaitong injection (XST) for the treatment of AMI by a meta-analysis. METHODS: A literature search was performed in 5 medical databases up to June 1, 2020. Randomized controlled trials involving XST combined with conventional treatment versus conventional treatment were included. A meta-analysis of clinical efficacy, left ventricular function and other objective parameters was performed to evaluate the effects of XST on AMI. RESULTS: Five randomized controlled trials involving 539 participants were eventually included. Meta-analysis showed that the combination of XST and conventional treatment could achieve significantly better effect on improving clinical efficacy (risk ratio: 1.09 [1.01, 1.17]; P = .04), left ventricular ejection fraction (mean difference [MD]: 3.18 [1.69, 4.67]; P < .0001), hypersensitive C-reactive protein (MD: -2.58 [-5.04, -0.12]; P = .04), interleukin 6 (MD: -26.00 [-38.85, -13.16]; P < .0001), cardiac troponin T (MD: -15.85 [-18.09, -13.61]; P < .00001) and creatine kinase myocardial isoenzyme (MD: -73.06 [-79.74, -66.37]; P < .00001). CONCLUSION: XST combined with conventional treatment can achieve better efficacy on clinical performance and some of the AMI related parameters. However the interpretation of the results should be cautious, due to the relatively low quality of included trials. More rigorously designed, large-scaled, randomized controlled trials are warranted to support its clinical use in the future.

Traditional Chinese medicine for acute coronary syndrome: A meta-analysis of clinical manifestations and objective indicators.

BACKGROUND: Modern clinical trials and experimental researches of traditional Chinese medicine (TCM) have been conducted for decades and provided support for the prevention and treatment of acute coronary syndrome (ACS). However the level of evidence and the proper application of TCM were still barely satisfactory. METHODS: In this study, we divided ACS into 3 different stages, including unstable angina, acute myocardial infarction, and post myocardial infarction. Then we systematically reviewed and meta-analyzed the existing randomized controlled trials on both clinical manifestations and objective indicators, in these 3 aspects. RESULTS: The results indicate that TCM can both improve the clinical manifestations and ameliorate the objective parameters in different courses of ACS, including C-reactive protein in unstable angina, left ventricular ejection fraction in acute myocardial infarction and post myocardial infarction. And the incidence of short-term cardiovascular events are lower in TCM intervention group. Some of the improvements lead to potential long-term benefits. CONCLUSION: TCM treatment is beneficial to different courses of ACS. To acquire more solid and comprehensive evidence of TCM in treating ACS, more rigorously designed randomized controlled trials with longer follow-up duration are warranted.

Effective Antitumor of Orally Intestinal Targeting Penetrating Peptide-Loaded Tyroserleutide/PLGA Nanoparticles in Hepatocellular Carcinoma.

PURPOSE: Hepatocellular carcinoma (HCC) is a common malignant tumor that seriously threatens human life and health. Currently, the majority of antitumor drugs are administered in an injectable manner, which can cause pain and side effects to patients. Objective of this study is to establish an effective oral drug delivery system for anti hepatoma drugs. METHODS: In this study, intestinal targeting cell penetrating peptide (R6LRVG) was obtained by binding cell penetrating peptide (R6) with the polypeptide of LRVG (targeting intestinal epithelial cells). Next, R6LRVG-modified tyroserleutide-poly(lactic-co-glycolic acid) (PLGA) nanoparticles (YSL-PLGA/R6LRVG NPs) were prepared. After that, the nanoparticles were characterized and their stability was evaluated. The cellular uptake, in vitro bioactivity and in vivo antitumor activity of the nanoparticles were investigated. In addition, the mechanism, including the endocytic pathway and respiratory rate detection of mitochondria, was further investigated. RESULTS: YSL-PLGA/R6LRVG NPs were successfully prepared. Characterization revealed YSL-PLGA/R6LRVG NPs to be globular particles with smooth surfaces and an average diameter of 222.6 nm. The entrapment efficiency and drug loading of tyroserleutide were 70.27% and 19.69%, respectively. Furthermore, the YSL-PLGA/R6LRVG NPs group exhibited the largest amount of YSL uptake. We also found that cell uptake of YSL-PLGA/R6LRVG NPs could be related to the endocytosis pathways mediated by reticulin and caveolae/lipid rafts. Additionally, the YSL-PLGA/R6LRVG NPs could interfere with mitochondrial function. In vivo experiments revealed that orally administered YSL-PLGA/R6LRVG NPs exerted excellent anticancer effects in tumor-bearing mice. Hematoxylin-eosin staining did not show any histological changes in the major organs. CONCLUSION: To summarize, YSL-PLGA/R6LRVG NPs could be a useful oral delivery system of YSL and may provide a new platform for the oral delivery of anticancer drugs.

kataegis: an R package for identification and visualization of the genomic localized hypermutation regions using high-throughput sequencing.

BACKGROUND: Genomic localized hypermutation regions were found in cancers, which were reported to be related to the prognosis of cancers. This genomic localized hypermutation is quite different from the usual somatic mutations in the frequency of occurrence and genomic density. It is like a mutations "violent storm", which is just what the Greek word "kataegis" means. RESULTS: There are needs for a light-weighted and simple-to-use toolkit to identify and visualize the localized hypermutation regions in genome. Thus we developed the R package "kataegis" to meet these needs. The package used only three steps to identify the genomic hypermutation regions, i.e., i) read in the variation files in standard formats; ii) calculate the inter-mutational distances; iii) identify the hypermutation regions with appropriate parameters, and finally one step to visualize the nucleotide contents and spectra of both the foci and flanking regions, and the genomic landscape of these regions. CONCLUSIONS: The kataegis package is available on Bionconductor/Github ( https://github.com/flosalbizziae/kataegis ), which provides a light-weighted and simple-to-use toolkit for quickly identifying and visualizing the genomic hypermuation regions.

Effect and mechanism of actionin vitroof cyclodextrin derivative nanoparticles loaded with tyroserleutide on hepatoma.

In this study, a cyclodextrin derivative (R6RGD-CMßCD) nanoparticle with tumor targeting and cell penetration ability was successfully synthesized and loaded with tyroserleutide (YSL) to obtain YSL-loaded nanoparticles (YSL/R6RGD-CMßCD NPs). The characterization of these NPs revealed a smooth surfaces and an average diameter of approximately 170 nm. YSL/R6RGD-CMßCD NPs increased the NP uptake in Caco-2 cells. As regard the mechanism of action, the cell uptake was related to endocytosis mediated by reticulin and megacytosis. In addition, YSL/R6RGD-CMßCD NPs induced significantly higher cytotoxicity on tumor cells and better tumor targeting compared with the effect of CMßCD NPs. Most importantly, the good anti-cancer effect of YSL/R6RGD-CMßCD NPs might be due to the interference with the function of mitochondria. On the other hand, YSL/R6RGD-CMßCD NPs were not toxic for normal cells. Taken together, our results indicated that R6RGD-CMßCD could be considered as a nanopharmaceutical material with good tumor targeting abilities, and their combination with YSL could represent an effective anti-cancer system.

Design and Investigation of Penetrating Mechanism of Octaarginine-Modified Alginate Nanoparticles for Improving Intestinal Insulin Delivery.

The aim of the study is to design octaarginine (R8)-modified insulin-alginate nanoparticles (INS-SA/R8 NPs) as the oral insulin delivery system, and further investigate its penetrating mechanism. The characterization results indicated that the surface of INS-SA/R8 NPs was smooth and the average diameter was about 300 nm. INS-SA/R8 NPs exhibited a stronger stability in the simulated gastrointestinal fluids and had a better controlled release than unmodified alginate nanoparticles (INS-SA NPs). Moreover, INS-SA/R8 NPs group had the strongest insulin transport capacity and the largest amount of insulin uptake in all experimental groups. Most importantly, the improvement of insulin intestinal uptake was further confirmed in rat intestine in vivo, and its penetrating mechanism might be involved in the production of endogenous nitric oxide (NO) signal molecule. In addition, in vivo hypoglycemic studies showed that orally administrated INS-SA/R8 NPs produced a better hypoglycemic effect as compared with INS-SA NPs in diabetic rats. Meanwhile, from the cytotoxicity analysis, INS-SA/R8 NPs were safe for oral administration. Taken together, INS-SA/R8 NPs was a good oral insulin delivery system, which might also be suitable for other protein drugs.

Effects of lifestyle interventions on rural patients with type 2 diabetes mellitus.

BACKGROUND: The prevalence of type 2 diabetes mellitus (T2DM) is rising rapidly in rural areas, and lifestyle interventions can effectively reduce the blood glucose levels of patients with T2DM. However, current dietary and exercise guidelines are still at experimental stages and are difficult for subjects to understand and implement. The Human Metabolism Analyzer provides real life interventions for the prevention and treatment of T2DM, and our pilot research has demonstrated its effectiveness and good compliance. AIM: To investigate the effect of and compliance with lifestyle interventions in rural patients with T2DM. METHODS: A total of ten rural villages were randomly selected in Chaoshui Township, Penglai City, Shandong Province, China, to conduct health screening among residents aged 50 years or older. Each rural village represented a group, and 12 patients with T2DM were randomly selected from each group (total: 120) to participate in this study and receive real life lifestyle interventions and medication guidance. Lifestyle interventions included changing the meal order (A), postprandial activities (B), resistance exercise (C), and reverse abdominal breathing (D). Diabetes education was conducted at least once a month with a weekly phone follow-up to monitor exercise and diet. Waist circumference, blood pressure, body mass index (BMI), motor function, body composition, fasting blood glucose, and glycated hemoglobin (HbA1c) were analyzed before and 3 mo after the intervention. Moreover, patient compliance and adjustments of hypoglycemic drugs were evaluated. RESULTS: A total of 109 subjects completed the study. The compliance rates for lifestyle interventions A, B, C, and D were 57.79%, 60.55%, 64.22%, and 75.23%, respectively. Among the subjects who received hypoglycemic drugs, the dose was reduced 2 to 3 times based on blood glucose in 54 (67.50%) subjects and was tapered and discontinued in 5 (6.25%) subjects within 3 mo, with no significant fluctuations in blood glucose after dose reduction and withdrawal. After lifestyle interventions, waist circumference, BMI, fasting blood glucose, and HbA1c significantly decreased (P < 0.001); motor function and body composition also significantly improved (P < 0.001). CONCLUSION: For patients with T2DM, compliance to real-life lifestyle interventions is good, and the interventions significantly improve metabolic indicators such as waist circumference, BMI, blood pressure, HbA1c, body composition, and motor function. Some patients are able to taper or discontinue hypoglycemic drugs.

Collagen/Chitosan Complexes: Preparation, Antioxidant Activity, Tyrosinase Inhibition Activity, and Melanin Synthesis.

Bioactive collagen/chitosan complexes were prepared by an ion crosslinking method using fish skin collagen and chitosan solution as raw materials. Scanning electron microscopy observation confirmed that the collagen/chitosan complexes were of a uniform spherical shape and uniform particle size. The complexes were stable at different pH values for a certain period of time through swelling experiments. Differential scanning calorimetry (DSC) showed the collagen/ chitosan complexes were more stable than collagen. X-ray diffraction (XRD) showed that the complexes had a strong crystal structure, and Fourier transform infrared spectroscopy (FTIR) data revealed the changes in the secondary structure of the protein due to chitosan and TPP crosslinking. The content of malondialdehyde (MDA) in the complex treatment group was considerably lower, but the content of SOD was significantly higher than that of the collagen group or chitosan group. In addition, the collagen/chitosan complexes could considerably reduce melanin content, inhibit tyrosinase activity, and down-regulate tyrosinase mRNA expression. In conclusion, the collagen/chitosan complexes were potential oral protein preparation for antioxidant enhancement and inhibiting melanin synthesis.

Distilling Channels for Efficient Deep Tracking.

Deep trackers have proven success in visual tracking. Typically, these trackers employ optimally pre-trained deep networks to represent all diverse objects with multi-channel features from some fixed layers. The deep networks employed are usually trained to extract rich knowledge from massive data used in object classification and so they are capable to represent generic objects very well. However, these networks are too complex to represent a specific moving object, leading to poor generalization as well as high computational and memory costs. This paper presents a novel and general framework termed channel distillation to facilitate deep trackers. To validate the effectiveness of channel distillation, we take discriminative correlation filter (DCF) and ECO for example. We demonstrate that an integrated formulation can turn feature compression, response map generation, and model update into a unified energy minimization problem to adaptively select informative feature channels that improve the efficacy of tracking moving objects on the fly. Channel distillation can accurately extract good channels, alleviating the influence of noisy channels and generally reducing the number of channels, as well as adaptively generalizing to different channels and networks. The resulting deep tracker is accurate, fast, and has low memory requirements. Extensive experimental evaluations on popular benchmarks clearly demonstrate the effectiveness and generalizability of our framework.

Oligoarginine mediated collagen/chitosan gel composite for cutaneous wound healing.

In this study, the collagen/chitosan gel composite supplemented with a cell-penetrating peptide (CPP) (Oligoarginine, R8) was prepared. Then, the physicochemical properties of the new collagen/chitosan/CPPs gel obtained were analyzed and the related characteristics were evaluated by scanning electron microscopy (SEM), fourier transform infrared (FTIR), differential scanning calorimetry (DSC), differential thermal analyzer (DTA). Furthermore, we found that collagen/chitosan/CPPs gel composite was capable of inhibiting Staphylococcus aureus growth and had good ability to heal wounds. The mice test results showed that collagen/chitosan/CPPs gel had the highest healing rate, fastest healing speed in all the treatments. After 14 days, the group treated by collagen/chitosan/CPPs gel showed nearly complete wound surface healing rate of 98 ±â€¯4.71%. In addition, histopathological examination suggested that collagen/chitosan/CPPs could promote cutaneous wound healing through enhancing granulation tissue formation, increasing collagen deposition and promoting angiogenesis in the wound tissue. Meanwhile, no significant cytotoxicity of the gel was observed. In conclusion, the collagen/chitosan/CPPs gel composite which has antibacterial activity renders a high therapeutic efficiency to heal wounds.