A veracity dissemination consistency-based few-shot fake news detection framework by synergizing adversarial and contrastive self-supervised learning.

With the rapid growth of social media, fake news (rumors) are rampant online, seriously endangering the health of mainstream social consciousness. Fake news detection (FEND), as a machine learning solution for automatically identifying fake news on Internet, is increasingly gaining the attentions of academic community and researchers. Recently, the mainstream FEND approaches relying on deep learning primarily involves fully supervised fine-tuning paradigms based on pre-trained language models (PLMs), relying on large annotated datasets. In many real scenarios, obtaining high-quality annotated corpora are time-consuming, expertise-required, labor-intensive, and expensive, which presents challenges in obtaining a competitive automatic rumor detection system. Therefore, developing and enhancing FEND towards data-scarce scenarios is becoming increasingly essential. In this work, inspired by the superiority of semi-/self- supervised learning, we propose a novel few-shot rumor detection framework based on semi-supervised adversarial learning and self-supervised contrastive learning, named Detection Yet See Few (DetectYSF). DetectYSF synergizes contrastive self-supervised learning and adversarial semi-supervised learning to achieve accurate and efficient FEND capabilities with limited supervised data. DetectYSF uses Transformer-based PLMs (e.g., BERT, RoBERTa) as its backbone and employs a Masked LM-based pseudo prompt learning paradigm for model tuning (prompt-tuning). Specifically, during DetectYSF training, the enhancement measures for DetectYSF are as follows: (1) We design a simple but efficient self-supervised contrastive learning strategy to optimize sentence-level semantic embedding representations obtained from PLMs; (2) We construct a Generation Adversarial Network (GAN), utilizing random noises and negative fake news samples as inputs, and employing Multi-Layer Perceptrons (MLPs) and an extra independent PLM encoder to generate abundant adversarial embeddings. Then, incorporated with the adversarial embeddings, we utilize semi-supervised adversarial learning to further optimize the output embeddings of DetectYSF during its prompt-tuning procedure. From the news veracity dissemination perspective, we found that the authenticity of the news shared by these collectives tends to remain consistent, either mostly genuine or predominantly fake, a theory we refer to as "news veracity dissemination consistency". By employing an adjacent sub-graph feature aggregation algorithm, we infuse the authenticity characteristics from neighboring news nodes of the constructed veracity dissemination network during DetectYSF inference. It integrates the external supervisory signals from "news veracity dissemination consistency" to further refine the news authenticity detection results of PLM prompt-tuning, thereby enhancing the accuracy of fake news detection. Furthermore, extensive baseline comparisons and ablated experiments on three widely-used benchmarks demonstrate the effectiveness and superiority of DetectYSF for few-shot fake new detection under low-resource scenarios.

An Infrared Small Target Detection Method Based on Attention Mechanism.

The human visual attention system plays an important role in infrared target recognition because it can quickly and accurately recognize infrared small targets and has good scene adaptability. This paper proposes an infrared small target detection method based on an attention mechanism, which consists of three modules: a bottom-up passive attention module, a top-down active attention module, and decision feedback equalization. In the top-down active attention module, given the Gaussian characteristics of infrared small targets, the idea of combining knowledge-experience Gaussian shape features is applied to implement feature extraction, and quaternion cosine transform is performed to achieve multi-dimensional fusion of Gaussian shape features, thereby achieving complementary fusion of multi-dimensional feature information. In the bottom-up passive attention module, considering that the difference in contrast and motion between the target and the background can attract attention easily, an optimal fast local contrast algorithm and improved circular pipeline filtering are adopted to find candidate target regions. Meanwhile, the multi-scale Laplacian of the Gaussian filter is adopted to estimate the optimal size of the infrared small target. The fast local contrast algorithm based on box filter acceleration and structure optimization is employed to extract local contrast features, and candidate target regions can be obtained by using an adaptive threshold. Besides, the mean gray, target size, Gaussian consistency, and circular region constraint are used in pipeline filtering to extract motion regions, and the false-alarm rate is reduced effectively. Finally, decision feedback equalization is adopted to obtain real targets. Experiments are conducted on some real infrared images involving complex backgrounds with sea, sky, and ground clutters, and the experimental results indicate that the proposed method can achieve better detection performance than conventional baseline methods, such as RLCM, ILCM, PQFT, MPCM, and ADMD. Also, mathematical proofs are provided to validate the proposed method.

Combined treatment with dihydrotestosterone and lipopolysaccharide modulates prostate homeostasis by upregulating TNF-α from M1 macrophages and promotes proliferation of prostate stromal cells.

Androgens and chronic inflammation, which play essential roles in the development of benign prostatic hyperplasia (BPH), are considered to be important factors in disorders of prostate homeostasis. These two factors may lead to pathological hyperplasia in the prostate transition zone of patients with BPH. However, few studies have examined the mechanism of how dihydrotestosterone (DHT) affects chronic inflammation in prostate tissue during the progression of BPH. This study examined the performance of DHT in lipopolysaccharide-treated M1 macrophages and the subsequent effects on the proliferation of prostate stromal and epithelial cells. We found that DHT increased secretion of the pro-inflammatory factor tumor necrosis factor (TNF)-α from M1 macrophages differentiated from THP-1 cells. The supernatant of M1 macrophages promoted the proliferation of WPMY-1 prostate stromal cells by upregulating B-cell lymphoma-extra large (Bcl-xL) and cellular Myc (c-Myc) levels by activating TNF-α-mediated nuclear factor-kappa B (NF-κB) and mitogen-activated protein kinase (MAPK) pathways. Moreover, this supernatant increased the expression of androgen receptor in WPMY-1 cells, which was TNF-α-independent. Additionally, TNF-α protein expression was significantly higher in patients with BPH and a large prostate volume than that in those with a small prostate volume. Further analysis showed that higher serum testosterone combined with prostate-specific androgen concentrations was related to TNF-α expression. This study suggests that DHT modulates the inflammatory environment of BPH by increasing TNF-α expression from lipopolysaccharide-treated M1 macrophages and promotes the proliferation of prostate stromal cells. Targeting TNF-α, but not DHT, may be a promising strategy for patients with BPH.

Oncogenic potential of BEST4 in colorectal cancer via activation of PI3K/Akt signaling.

BEST4 is a member of the bestrophin protein family that plays a critical role in human intestinal epithelial cells. However, its role and mechanism in colorectal cancer (CRC) remain largely elusive. Here, we investigated the role and clinical significance of BEST4 in CRC. Our results demonstrate that BEST4 expression is upregulated in clinical CRC samples and its high-level expression correlates with advanced TNM (tumor, lymph nodes, distant metastasis) stage, LNM (lymph node metastasis), and poor survival. Functional studies revealed that ectopic expression of BEST4 promoted CRC cell proliferation and metastasis, whereas the depletion of BEST4 had the opposite effect both in vitro and in vivo. Mechanistically, BEST4 binds to the p85α regulatory subunit of phosphatidylinositol-3-kinase (PI3K) and promotes p110 kinase activity; this leads to activation of Akt signaling and expression of MYC and CCND1, which are critical regulators of cell proliferation and metastasis. In clinical samples, the expression of BEST4 is positively associated with the expression of phosphorylated Akt, MYC and CCND1. Pharmacological inhibition of Akt activity markedly repressed BEST4-mediated Akt signaling and proliferation and metastasis of CRC cells. Importantly, the interaction between BEST4 and p85α was also enhanced by epidermal growth factor (EGF) in CRC cells. Therapeutically, BEST4 suppression effectively sensitized CRC cells to gefitinib treatment in vivo. Taken together, our findings indicate the oncogenic potential of BEST4 in colorectal carcinogenesis and metastasis by modulating BEST4/PI3K/Akt signaling, highlighting a potential strategy for CRC therapy.

Human Activity Recognition Based on Dynamic Active Learning.

Activity of daily living is an important indicator of the health status and functional capabilities of an individual. Activity recognition, which aims at understanding the behavioral patterns of people, has increasingly received attention in recent years. However, there are still a number of challenges confronting the task. First, labelling training data is expensive and time-consuming, leading to limited availability of annotations. Secondly, activities performed by individuals have considerable variability, which renders the generally used supervised learning with a fixed label set unsuitable. To address these issues, we propose a dynamic active learning-based activity recognition method in this work. Different from traditional active learning methods which select samples based on a fixed label set, the proposed method not only selects informative samples from known classes, but also dynamically identifies new activities which are not included in the predefined label set. Starting with a classifier that has access to a limited number of labelled samples, we iteratively extend the training set with informative labels by fully considering the uncertainty, diversity and representativeness of samples, based on which better-informed classifiers can be trained, further reducing the annotation cost. We evaluate the proposed method on two synthetic datasets and two existing benchmark datasets. Experimental results demonstrate that our method not only boosts the activity recognition performance with considerably reduced annotation cost, but also enables adaptive daily activity analysis allowing the presence and detection of novel activities and patterns.

Clinical spectrum and gene mutations in a Chinese cohort with anoctaminopathy.

Recessive mutations in anoctamin-5 (ANO5) are causative for limb-girdle muscular dystrophy (LGMD) 2 L and non-dysferlin Miyoshi-like distal myopathy (MMD3). ANO5 mutations are highly prevalent in European countries; however it is not common in patients of Asian origin, and there is no data regarding the Chinese population. We retrospectively reviewed the clinical manifestations and gene mutations of Chinese patients with anoctaminopathy. A total of five ANO5 mutations including four novel mutations and one reported mutation were found in four patients from three families. No hotspot mutation was found. Three patients presented with presymptomatic hyperCKemia and one patient had limb muscle weakness. Muscle imaging of lower limbs showed preferential adductor magnus and medial gastrocnemius involvement. No hotspot mutation has been identified in Chinese patients to date.

Mesial temporal extraventricular neurocytoma: a rare cause of refractory complex partial seizure.

AIM: Mesial temporal extraventricular neurocytoma (mtEVN) is a rare cause of refractory complex focal seizures. The characteristics of this clinical entity are discussed in this article. METHODS: We report two cases of mtEVN and review the related literature, with particular emphasis on radiological characteristics, clinical features, and operative techniques. RESULTS: After successful surgery, our two cases of mtEVN achieved excellent outcome. Including the cases presented here, a total of three cases of mtEVNs and 11 of neocortical temporal extraventricular neurocytoma (ntEVNs) are reported in the literature. mtEVNs are distinct from ntEVNs with regards to demographics, aetiology, radiological features, and operative techniques. CONCLUSION: mtEVNs and ntEVNs exhibit distinguishing features. Under electrocorticographic monitoring, tailored resection of the neocortical epileptogenic focus, as well as the entire tumour and mesial temporal structures, can yield excellent outcome and satisfactory seizure control.

Surgical treatment and radiotherapy of epidermoid cyst with malignant transformation in cerebellopontine angle.

We report surgical treatment and radiotherapy of an extremely rare case of malignant epidermoid cyst located in cerebellopontine angle. MRI and CT demonstrated the lesion with partial enhancement and calcification. During operation, we found the tumor attached tightly to surrounding tissue. Finally we achieved near total resection of it. Histopathology confirmed the diagnosis of epidermoid cyst with malignant transformation. With adjuvant radiotherapy, the patient underwent excellent recovery, and follow-up MRI demonstrated no obvious residue or recurrence of the tumor. Malignant epidermoid cyst can be diagnosed radiologically in combination with clinical presentation. Maximal removal plus adjuvant radiotherapy is the treatment of choice, although the general prognosis of it is poor.