High doses of tumor necrosis factor-α (TNF-α) suppress osteogenic differentiation of human dental pulp stem cells (hDPSCs). In the present study, we aimed to explore the role and potential regulatory ...mechanism of microRNA-138 (miR-138) in the osteogenic differentiation of hDPSCs after treatment with a high dose of TNF-α. The hDPSCs were cultured in osteogenic medium with or without 50 ng/ml TNF-α. The miR-138 levels were upregulated during osteogenic differentiation of the hDPSCs following TNF-α treatment. The miR-138 overexpression accelerated but miR-138 knockdown alleviated the TNF-α-induced suppression of the alkaline phosphatase activity, calcium deposition, and protein abundance of dentin sialophosphoprotein, dentin matrix protein 1, bone sialoprotein, and osteopontin during osteogenic differentiation induction of hDPSCs. Additionally, miR-138 overexpression accelerated but miR-138 knockdown alleviated the suppression of the focal adhesion kinase- (FAK-) extracellular signal-regulated kinase 1/2 (ERK1/2) signaling pathway during osteogenic differentiation induction of hDPSCs under TNF-α treatment. In conclusion, miR-138 accelerates TNF-α-induced suppression of osteogenic differentiation of hDPSCs. Inactivation of the FAK-ERK1/2 signaling pathway may be one of the mechanisms underlying the effect of miR-138. Inhibition of miR-138 expression may be a strategy to weaken the inhibitory effect of high-dose TNF-α on the osteogenic differentiation of hDPSCs.
Flexible pressure sensors with excellent performance have broad application potential in wearable devices, motion monitoring, and human–computer interaction. In this paper, a flexible pressure sensor ...with a porous structure is proposed by coating molybdenum disulfide (MoS2) and hydroxyethyl cellulose (HEC) on a polyurethane (PU) sponge skeleton. The obtained sensor has excellent sensitivity (0.746 kPa–1), a wide detection range (250 kPa), fast response (120 ms), and outstanding repeatability over 2000 cycles. It is proven that the sensor can realize human motion detection and distinguish the touch of varying strength. In addition, a pressure sensing array was fabricated to reflect the pressure distribution and recognize the writing of Arabic numerals. Finally, the sensor performs speech detection through throat muscle movements, and high-accuracy (97.14%) speech recognition for seven words was achieved by a machine learning algorithm based on the support vector machine (SVM). This work provides an opportunity to fabricate simple flexible pressure sensors with potential applications in next-generation electronic skin, health detection, and intelligent robotics.
Visual question answering (VQA) has become a hot study topic with challenging motivation of correctly answering the videos or images questions in recent years. However, the existing VQA model mostly ...aimed at answering questions about images and performed poorly in the video question answering (VideoQA) domain. VideoQA needs to simultaneously consider the correlations between video frames and the dynamic information of multiple objects in video. Therefore, we propose a novel Cascade Transformers with Dynamic Attention for Video Question Answering (CTDA-QA), which aims to simultaneously solve the above considerations. Specifically, the proposed CTDA-QA model utilizes multiple transformers structure to encode videos for reasoning complex spatial and temporal information, which is different from the previous recurrent neural network methods. Besides, in order to effectively capture the dynamic information from various scenarios in videos, a flexible attention module has been proposed to explore the essential relations between objects in a dynamic timeline. Finally, to avoid spurious answers and fully explore the cross-modal relationships, a mixed-supervised learning strategy is designed for optimizing the reasoning tasks. The experiments on several benchmark video question–answer datasets clearly verify the performance and effectiveness of CTDA-QA, which contains the results in contrast to the state-of-the-art methods. Besides, the provided ablation study and visualization results further reveal the potential of CTDA-QA.
•This paper proposes a novel framework for video question answering.•Learn about objects in terms of their spatiality, temporality and interrelationships.•The dynamic attention focuses on visual features related to text in videos.
Multi-view clustering, which aims at dividing data with similar structures into their respective groups, is a popular research subject in computer vision and machine learning. In recent years, ...Non-negative matrix factorization (NMF) has received constant concern in multi-view clustering due to its ability to deal with high-dimensional data. However, most existing NMF methods may fail to integrate valuable information from multi-view data adequately, and the local geometry structure in data is also not fully considered. Thus, it’s still a crucial but challenging problem, which effectively extracts multi-view information while maintaining the low-dimensional geometry structure. In this paper, we propose an innovative multi-view clustering method, referred to as re-weighted multi-view clustering via triplex regularized non-negative matrix factorization (SMCTN), which is a unified framework and provides the following contributions: 1) pairwise regularization can extract complementary information between views and is suitable for both homogeneous and heterogeneous perspectives; 2) consensus regularization can process the consistent information between views; 3) graph regularization can preserve the geometric structure of data. Specifically, SMCTN applies a re-weighted strategy to assign suitable weights for multiple views according to their contributions. Besides, an effective iterative updating algorithm is developed to solve the non-convex optimization problem in SMCTN. Extensive experimental results on textual and image datasets indicate that the superior performance of the proposed method.
Multi-view clustering is to divide data into distinct clusters according to their different features. Tensor-based multi-view clustering can capture higher order connections between various views for ...better clustering results. However, they have limitations: (1) the higher-order local geometric structure in non-linear subspaces is not considered; (2) significant differences in singular values are not reflected. To resolve the above issues, we introduce a novel method called Enhanced Tensor Multi-view Clustering via Dual Constraints(ETMC-DC). ETMC-DC utilizes Hyper-Laplacian regularization to maintain higher-order local geometric structure in the raw space. The Schatten-p norm is used to the tensor stacked by the obtained affinity matrix to process unequal singular values, and larger singular values carry more structural information, and vice versa. Moreover, the complexity of the model is reduced by the rotation of the tensors constructed from Markov transition probabilities. Finally, an iterative update technique is used for optimizing the presented ETMC-DC. We have conducted extensive experiments on real-world datasets in various forms to demonstrate that ETMC-DC can perform exceptionally well in comparison to other multi-view clustering approaches.
Display omitted
The common goal of the studies is to map any emotional states encoded from electroencephalogram (EEG) into 2-dimensional arousal-valance scores. It is still challenging due to each emotion having its ...specific spatial structure and dynamic dependence over the distinct time segments among EEG signals. This paper aims to model human dynamic emotional behavior by considering the location connectivity and context dependency of brain electrodes. Thus, we designed a hybrid EEG modeling method that mainly adopts the attention mechanism, combining a multi-domain spatial transformer (MST) module and a dynamic temporal transformer (DTT) module, named MSDTTs. Specifically, the MST module extracts single-domain and cross-domain features from different brain regions and fuses them into multi-domain spatial features. Meanwhile, the temporal dynamic excitation (TDE) is inserted into the multi-head convolutional transformer to form the DTT module. These two blocks work together to activate and extract the emotion-related dynamic temporal features within the DTT module. Furthermore, we place the convolutional mapping into the transformer structure to mine the static context features among the keyframes. Overall results show that high classification accuracy of 98.91%/0.14% was obtained by the <inline-formula><tex-math notation="LaTeX">\beta</tex-math></inline-formula> frequency band of the DEAP dataset, and 97.52%/0.12% and 96.70%/0.26% were obtained by the <inline-formula><tex-math notation="LaTeX">\gamma</tex-math></inline-formula> frequency band of SEED and SEED-IV datasets. Empirical experiments indicate that our proposed method can achieve remarkable results in comparison with state-of-the-art algorithms.
Clustering is a notable research topic, but it is still challenging when facing massive multi-view data from different ways or multiple feature extractors. The crucial problem is how to promote ...cooperative learning between views via subspace reconstruction while preserving the underlying geometric structure of data. Moreover, most existing methods habitually utilize K-means to achieve the final results, which is not conducive to dealing with intricate non-convex patterns in multi-perspective data. Based on the above consideration, in this paper, we present a fresh multi-view clustering approach called Adaptive Multi-view Multiple-Means Clustering via Subspace Reconstruction(AM2CSR). AM2CSR aims to simultaneously capture compatible, complementary, geometric, and discrimination information among multiple views. Subsequently, a low-rank restriction is forced on the low-dimensional representation to reduce redundancy, and K-Multiple-Means(KMM) is adopted as the clustering technique to achieve satisfying results. Additionally, an effective iteration updating method with a convergence guarantee is applied to settle the optimization matter of AM2CSR. Extensive empirical experiments on eight benchmark datasets exhibit the superiority of AM2CSR.
It is highly desirable that speech enhancement algorithms can achieve good performance while keeping low latency for many applications, such as digital hearing aids, mobile phones, acoustically ...transparent hearing devices, and public address systems. To improve the performance of traditional low-latency speech enhancement algorithms, a deep filter-bank equalizer (FBE) framework was proposed that integrated a deep learning-based subband noise reduction network with a deep learning-based shortened digital filter mapping network. In the first network, a deep learning model was trained with a controllable small frame shift to satisfy the low-latency demand, i.e., no greater than 4 ms, so as to obtain (complex) subband gains that could be regarded as an adaptive digital filter in each frame. In the second network, to reduce the latency, this adaptive digital filter was implicitly shortened by a deep learning-based framework and was then applied to noisy speech to reconstruct the enhanced speech without the overlap-add method. Experimental results on the WSJ0-SI84 corpus indicated that the proposed DeepFBE with only 4-ms latency achieved much better performance than traditional low-latency speech enhancement algorithms across several objective metrics. Listening test results further confirmed that our approach achieved higher speech quality than other methods.
It remains a tough challenge to recover the speech signals contaminated by various noises under real acoustic environments. To this end, we propose a novel system for denoising in the complicated ...applications, which is mainly comprised of two pipelines, namely a two-stage network and a post-processing module. The first pipeline is proposed to decouple the optimization problem w.r.t. magnitude and phase, i.e., only the magnitude is estimated in the first stage and both of them are further refined in the second stage. The second pipeline aims to further suppress the remaining unnatural distorted noise, which is demonstrated to sufficiently improve the subjective quality. In the ICASSP 2021 Deep Noise Suppression (DNS) Challenge, our submitted system ranked top-1 for the real-time track 1 in terms of Mean Opinion Score (MOS) with ITU-T P.808 framework.
Flexible pressure sensors have been widely concerned because of their great application potential in the fields of electronic skin, human–computer interaction, health detection, and so on. In this ...paper, a flexible pressure sensor is designed, with polydimethylsiloxane (PDMS) films with protruding structure as elastic substrate and poly(3,4-ethylenedioxythiophene)/poly(styrenesulfonate) (PEDOT:PSS)/cellulose nanocrystals (CNC) as conductive-sensitive material. The flexible pressure sensor has a wide linear detection range (0–100 kPa), outstanding sensitivity (2.32 kPa–1), and stability of more than 2000 cycles. The sensor has been proven to be able to detect a wide range of human movements (finger bending, elbow bending, etc.) and small movements (breathing, pulse, etc.). In addition, the pressure sensor array can detect the pressure distribution and judge the shape of the object. A smart wristband equipped with four flexible pressure sensors is designed. Among them, the k-nearest neighbor (KNN) algorithm is used to classify sensor data to achieve high accuracy (99.52%) recognition of seven kinds of wrist posture. This work provides a new opportunity to fabricate simple, flexible pressure sensors with potential applications in the next-generation electronic skin, health detection, and intelligent robotics.
PDF