OPOM: Customized Invisible Cloak Towards Face Privacy Protection.

While convenient in daily life, face recognition technologies also raise privacy concerns for regular users on the social media since they could be used to analyze face images and videos, efficiently and surreptitiously without any security restrictions. In this paper, we investigate the face privacy protection from a technology standpoint based on a new type of customized cloak, which can be applied to all the images of a regular user, to prevent malicious face recognition systems from uncovering their identity. Specifically, we propose a new method, named one person one mask (OPOM), to generate person-specific (class-wise) universal masks by optimizing each training sample in the direction away from the feature subspace of the source identity. To make full use of the limited training images, we investigate several modeling methods, including affine hulls, class centers and convex hulls, to obtain a better description of the feature subspace of source identities. The effectiveness of the proposed method is evaluated on both common and celebrity datasets against black-box face recognition models with different loss functions and network architectures. In addition, we discuss the advantages and potential problems of the proposed method. In particular, we conduct an application study on the privacy protection of a video dataset, Sherlock, to demonstrate the potential practical usage of the proposed method.

A novel small molecule glycolysis inhibitor WZ35 exerts anti-cancer effect via metabolic reprogramming.

BACKGROUND: Liver cancer is the fifth leading cause of cancer death worldwide, but early diagnosis and treatment of liver cancer remains a clinical challenge. How to screen and diagnose liver cancer early and prolong the survival rate is still the focus of researchers. METHODS: Cell experiments were used to detect the effect of WZ35 on the colony formation ability and proliferation activity of hepatoma cells, nude mouse experiment to observe the in vivo anticancer activity and toxic side effects of WZ35; metabolomics analysis, glucose metabolism experiment and Seahorse analysis of liver cancer cells treated with WZ35; cell experiments combined with bioinformatics analysis to explore the mechanism of WZ35-mediated metabolic reprogramming to exert anticancer activity; tissue microarray and case analysis to evaluate the clinical significance of biomarkers for early diagnosis, treatment and prognosis evaluation of liver cancer. RESULTS: WZ35 inhibited the proliferation activity of various cell lines of liver cancer, and showed good therapeutic effect in nude mice model of hepatocellular carcinoma without obvious toxic and side effects; WZ35 inhibited the absorption of glucose in hepatoma cells, and the drug effect glycolysis, phosphorylation and purine metabolism are relatively seriously damaged; WZ35 mainly inhibits YAP from entering the nucleus as a transcription factor activator by activating oxidative stress in liver cancer cells, reducing the transcription of GLUT1, and finally reducing its GLUT1. Tissue microarray and case analysis showed that GLUT1 and YAP were highly expressed and correlated in liver cancer patients, and were associated with poor patient prognosis. The GLUT1-YAP risk model had a high score in predicting prognosis. CONCLUSION: The study confirms that WZ35 is a small molecule glycolysis inhibitor, and through its properties, it mediates metabolic reprogramming dominated by impaired glycolysis, oxidative phosphorylation and purine metabolism to inhibit the proliferation activity of liver cancer cells. Our findings present novel insights into the pathology of liver cancer and potential targets for new therapeutic strategies. GLUT1-YAP has important reference significance for predicting the stages of disease progression in liver cancer patients and have the potential to serve as novel biomarkers for the diagnosis and treatment of liver cancer.

1α, 25-Dihydroxyvitamin D3 protects gastric mucosa epithelial cells against Helicobacter pylori-infected apoptosis through a vitamin D receptor-dependent c-Raf/MEK/ERK pathway.

CONTEXT: Due to the resistance of Helicobacter pylori to antibiotics, it is difficult to eradicate this pathogenic bacterium from the host. The role of 1α, 25-dihydroxyvitamin D3 (1,25-D3) in H. pylori-infected gastric mucosa epithelial cells remains unknown. OBJECTIVE: This study investigates the protective property of 1,25-D3 against H. pylori-infected apoptosis in gastric mucosa epithelial cells and its potential molecular mechanisms. MATERIALS AND METHODS: GES-1 cells were infected with H. pylori SS1 strain (MOI: 100) and treated with 1,25-D3 at 100, 200, and 300 nM for 24 h. Mice were orally gavaged with 108 CFUs of H. pylori and 25 µg/kg 1,25-D3 every other day for 1 month. CCK-8, LDH assay, TUNEL assay and western blot were used to determine the effect of 1,25-D3 on H. pylori-induced apoptosis. RESULTS: H. pylori infection decreased cell viability to 59.2%, while 100-300 nM 1,25-D3 increased cell viability to 62.2%, 78.4% and 87.1%, respectively. Compared with positive control (4.53-fold), 1,25-D3 reduced caspase-3 activity to 4.49-, 2.88- and 1.49-fold, reduced caspase-6 activity to 2.36-, 1.88- and 1.50-fold, reduced caspase-9 activity to 4.55-, 2.91- and 2.01-fold. 1,25-D3 alters Bcl-2 family, caspase protein expression and c-Raf/MEK/ERK phosphorylation levels in vivo and in vitro. Suppression of 1,25-D3 in apoptosis was reliant on binding to vitamin D receptor. The pharmacological inhibition of c-Raf/MEK/ERK phosphorylation blocked the anti-apoptotic effect of 1,25-D3. DISCUSSION AND CONCLUSION: 1,25-D3 protected gastric mucosa epithelial cells against H. pylori-infected apoptosis through a VDR-dependent c-Raf/MEK/ERK pathway.

Influence of biased feedback on performance in a Vernier discrimination task.

The influence of feedback on performance is a topic of ongoing debate, with some previous studies finding it to be ineffective, while others have discovered that it can be helpful or harmful. One possible reason for these inconsistent results may be that feedback can create a conflict between a person's beliefs and the sensory information they receive. In the present study, we used a Vernier discrimination task to examine the influence of biased feedback on performance, as this type of feedback is most likely to create conflict. Biased feedback refers to feedback that does not align with the subjects' choices. The Vernier discrimination task is a type of psychophysical task that is often used to measure an individual's ability to perceive differences in the position or orientation of two visual stimuli. The task involves presenting two stimuli, one of which is slightly offset from the other, and asking the individual to determine the direction and magnitude of the offset. In Experiment 1, feedback was provided after each trial using large-offset verniers as guidance. The large-offset verniers always received correct feedback, but the small and medium-offset verniers might receive biased feedback. In Experiment 2, feedback was provided after each block of eight verniers. In Experiment 3, we removed the large offset vernier to investigate the influence of block feedback on the signal and noise. The results showed that the accuracy for the target vernier decreased due to biased feedback in both the trial feedback (Experiment 1) and the block feedback (Experiment 2). However, in Experiments 1 and 2, performance improved when feedback was absent. Moreover, if the difference between the two types of stimuli is great, the individual will engage in encoding learning rather than decision learning (Experiments 1 and 2). If the discrimination between the two types of stimuli is low, an individual's ability to discriminate noise is more vulnerable to the influence of biased feedback than the ability to discriminate the signal (Experiment 3). These results are discussed in relation to the mechanism of biased feedback, the process of encoding learning, the monitoring of internal feedback, and the generalization of false decisions.

SFace: Sigmoid-Constrained Hypersphere Loss for Robust Face Recognition.

Deep face recognition has achieved great success due to large-scale training databases and rapidly developing loss functions. The existing algorithms devote to realizing an ideal idea: minimizing the intra-class distance and maximizing the inter-class distance. However, they may neglect that there are also low quality training images which should not be optimized in this strict way. Considering the imperfection of training databases, we propose that intra-class and inter-class objectives can be optimized in a moderate way to mitigate overfitting problem, and further propose a novel loss function, named sigmoid-constrained hypersphere loss (SFace). Specifically, SFace imposes intra-class and inter-class constraints on a hypersphere manifold, which are controlled by two sigmoid gradient re-scale functions respectively. The sigmoid curves precisely re-scale the intra-class and inter-class gradients so that training samples can be optimized to some degree. Therefore, SFace can make a better balance between decreasing the intra-class distances for clean examples and preventing overfitting to the label noise, and contributes more robust deep face recognition models. Extensive experiments of models trained on CASIA-WebFace, VGGFace2, and MS-Celeb-1M databases, and evaluated on several face recognition benchmarks, such as LFW, MegaFace and IJB-C databases, have demonstrated the superiority of SFace.

Curcumin micelles suppress gastric tumor cell growth by upregulating ROS generation, disrupting redox equilibrium and affecting mitochondrial bioenergetics.

Gastric cancer is the fourth most common cancer and the second most frequent cause of cancer death worldwide. Chemotherapy is an important treatment. However, traditional chemotherapy drugs have low bioavailability and targeting ability. Therefore, we developed curcumin-encapsulated micelles for the treatment of gastric cancer and investigated their antitumor efficacy and active mechanism. Gastric cancer cells were treated with different concentrations of curcumin micelles. MTS cell proliferation assays, flow cytometry (FCM), real time cellular analysis (RTCA) and nude mice xenografts were used to evaluate the effects of curcumin micelles on gastric cancer cell growth in vitro and in vivo. Western blotting was performed to analyze the protein levels of the indicated molecules. A Seahorse bioenergetics analyzer was used to investigate alterations in oxygen consumption and the aerobic glycolysis rate. Curcumin micelles significantly inhibited proliferation and colony formation and induced apoptosis in gastric tumor cells compared to the control groups. We further investigated the mechanism of curcumin micelles on gastric tumor cells and demonstrated that curcumin micelles acted on mitochondrial proteins, causing changes in mitochondrial function and affecting mitochondrial bioenergetics. Furthermore, curcumin micelles decreased mitochondrial membrane potential, increased reactive oxygen species (ROS) generation and disrupted redox equilibrium. The nude mouse model verified that curcumin micelle treatment significantly attenuated tumor growth in vivo. Curcumin micelles suppress gastric tumor cell growth in vitro and in vivo. The mechanism may be related to increasing ROS generation, disrupting redox equilibrium and affecting mitochondrial bioenergetics.