Search

Emotion detection in online social networks (OSNs) can benefit kinds of applications, such as personalized advertisement services, recommendation systems, etc. Conventionally, emotion analysis mainly focuses on the sentence level polarity prediction or single emotion label classification, however, ignoring the fact that emotions might coexist from users’ perspective. To this end, in this work, we address the multiple emotions detection in OSNs from user-level view, and formulate this problem as a multilabel learning problem. First, we discover emotion labels correlations , social correlations , and temporal correlations from an annotated Twitter data set. Second, based on the above observations, we adopt a factor graph-based emotion recognition model to incorporate emotion labels correlations , social correlations , and temporal correlations into a general framework, and detect the multiple emotions based on the multilabel learning approach. Performance evaluation demonstrates that the factor graph-based emotion detection model can outperform the existing baselines.

Developing prediction models for emerging infectious diseases from relatively small numbers of cases is a critical need for improving pandemic preparedness. Using COVID-19 as an exemplar, we propose a transfer learning methodology for developing predictive models from multi-modal electronic healthcare records by leveraging information from more prevalent diseases with shared clinical characteristics. Our novel hierarchical, multi-modal model (TransMED) integrates baseline risk factors from the natural language processing of clinical notes at admission, time-series measurements of biomarkers obtained from laboratory tests, and discrete diagnostic, procedure and drug codes. We demonstrate the alignment of TransMED's predictions with well-established clinical knowledge about COVID-19 through univariate and multivariate risk factor driven sub-cohort analysis. TransMED's superior performance over state-of-the-art methods shows that leveraging patient data across modalities and transferring prior knowledge from similar disorders is critical for accurate prediction of patient outcomes, and this approach may serve as an important tool in the early response to future pandemics.

Objective: To analyze the differential expression of related serum proteins was detected in patients with AR who were sensitized to dust mites pre- and 6-day post-RIT. Methods: Serum samples were collected from 15 AR, 15 6-day post-RIT and 10 healthy control patients. This analysis was performed using iTRAQ technique and was coupled with 2-dimensional gel electrophoresis and mass spectrometry. To determine the intimacy of the screened related differential proteins, a hierarchical analysis was performed with Cluster 3.0 software, and the results were visualized using Java TreeView software. Last, an ELISA was employed to validate the selected CRP, CTHRC1 and WDR89 protein expression data. Results: Compared with the healthy control group, the most upregulated protein in the AR group and the most downregulated proteins in the 6-day post-RIT group were CTHRC1 and WDR89. Compared with the healthy control group, the most downregulated protein in the AR group and the most upregulated protein of the 6-day post-RIT group was CRP. The serum protein levels of CRP, CTHRC1 and WDR89 proteins were determined by ELISA. It was found that the differential expression of CTHRC1 and WDR89 in the AR and 6-day post-RIT groups were statistically significant (P 0.05) was found in the differential expression of serum CRP in the AR and 6-day post-RIT groups. Conclusions: We found that CTHRC1 and WDR89 might be associated with the pathogenesis of AR, and they may play a regulatory role in the RIT.

Spectral beam combining is an effective method to increase the total output power of a laser system while maintaining a high beam quality. The diffraction grating with high diffraction efficiency and high laser damage threshold is the key component in a spectral-beam-combining laser system. To meet the above requirements, we design a grating on top of a high-reflectivity HfO2/SiO2 thin-film stack. The diffraction efficiency of the grating depends mainly on the duty cycle and depth of the grating grooves. The grating was fabricated by using optical interference lithography and reactive ion beam etching. To achieve high efficiency, we controlled the duty cycle by applying an end-point detection technique during development and we controlled the groove depth by adjusting the ion-beam etching time length. In a 50 mm × 50 mm fabricated grating area, the measured diffraction efficiency in TE polarization was 95.8 ± 1.1 % at the center wavelength of 1030 nm. In the wavelength range of 1025 nm to 1035 nm, the measured average diffraction efficiency in TE polarization was 97.7% at center of the grating sample. We describe the details of our work, including design parameters, fabrication processes, and measured diffraction efficiencies.

Superparamagnetic tunnel junctions (STJs) are nanostructures with very low turnover barriers. The barrier height of an STJ is generally equal to the heat energy at room temperature; thus, it can oscillate automatically without external driving. Previous studies have shown that the randomness of an STJ can be driven by a subthreshold voltage. This synchronization can be adjusted using electrical noise, which is often considered as zero-field Gaussian white noise. However, the actual circuit and environment are inevitably associated with colored noise, which has not been considered previously. In this work, numerical simulations were performed to study the phase-locking characteristics of a single STJ with the aid of several typical types of colored noise. The results show that the phase-locked behavior of an STJ can be effectively enhanced by colored noise whose power spectral density per unit of bandwidth is proportional to its frequency. Meanwhile, colored noise whose power spectral density per unit of bandwidth and frequency are inversely proportional can suppress the synchronization of STJs by suppressing the increase in junction frequency.

The realization of ultra-low energy consumption spin wave device applications requires a material that can efficiently propagate the spin waves. Spinel-type ferrite is a promising insulator from the perspective of application in magnonic devices owing to its good crystal compatibility. In this study, thin films of spinel γ-Fe2−xAlxO3 (0 ≤ x ≤0.5) are fabricated on single-crystal MgO substrates through pulsed laser deposition. High crystalline quality and room temperature ferrimagnetism are observed in all the films. Spin wave propagation is observed in γ-Fe2−xAlxO3 films, and it is found that the spin-wave transmission properties significantly correlate with the crystallinity and magnetization of the films. The γ-Fe1.85Al0.15O3 thin film represents an excellent compromise between a sufficiently high spin-wave amplitude and long propagation distance. These results indicate that the γ-Fe2−xAlxO3 thin film is a highly promising candidate for application in spin-wave-based computing devices in the future.

The CC chemokine receptor 3 (CCR3) expressed by eosinophils, mast cells and Th2 cells is closely related to allergic diseases. The objective of this study was to explore whether silencing of CCR3 with short hairpin RNAs (shRNAs) delivered by a lentiviral vector could impact the function of mast cells in a murine model of allergic rhinitis (AR) in vivo. The murine model of allergic rhinitis (AR) inducing by ovalbumin (OVA) was constructed, and the BALB/c mice were divided into normal control group, AR group, controlshRNA treated group and lentiviral CCR3-shRNA treated group. The recombinant lentivirus vectors which express a short hairpin RNA (shRNA) targeting the CCR3 were dropped into the nasal cavity of OVA-sensitized mice before the challenges. Real-time fluorescence quantitative PCR and western blotting were performed to observe inhibitory effect of CCR3 gene. Nasal symptoms of mice and OVA-specific IgE in each group were assessed. Concentrations of histamine, tryptase and Prostaglandin D2 (PGD2) in bone marrow, peripheral blood and nasal mucosa were analyzed. Furthermore, histological analysis and electron microscopy analysis were applied to detect the histology changes of nasal mucosa and the infiltration of mast cells in nasal mucosa. The results showed that administration of CCR3shRNA could effectively inhibit the expression of the CCR3 gene in bone marrow, peripheral blood and nasal mucosa, which reduced the nasal symptoms, the level of OVA-specific IgE, the inflammatory cells and mast cells infiltration into nasal cavity, and relieved the histopathological changes of nasal mucosa. In addition, intervention of CCR3shRNA could reduce the levels of the histamine, tryptase and PGD2 in bone marrow, peripheral blood and nasal mucosa. These results suggest that inhibition of CCR3 gene expression by shRNAs lentiviral vectors can effectively attenuate migration, infiltration and degranulation of mast cells in local tissues and alleviate the inflammation of allergic rhinitis mice.

PURPOSE: To compare machine learning methods for classifying mass lesions on mammography images that use predefined image features computed over lesion segmentations to those that leverage segmentation-free representation learning on a standard, public evaluation dataset. METHODS: We apply several classification algorithms to the public Curated Breast Imaging Subset of the Digital Database for Screening Mammography (CBIS-DDSM), in which each image contains a mass lesion. Segmentation-free representation learning techniques for classifying lesions as benign or malignant include both a Bag-of-Visual-Words (BoVW) method and a Convolutional Neural Network (CNN). We compare classification performance of these techniques to that obtained using two different segmentation-dependent approaches from the literature that rely on specific combinations of end classifiers (e.g. linear discriminant analysis, neural networks) and predefined features computed over the lesion segmentation (e.g. spiculation measure, morphological characteristics, intensity metrics). RESULTS: We report area under the receiver operating characteristic curve (A(Z)) values for malignancy classification on CBIS-DDSM for each technique. We find average A(Z) values of 0.73 for a segmentation-free BoVW method, 0.86 for a segmentation-free CNN method, 0.75 for a segmentation-dependent linear discriminant analysis of Rubber-Band Straightening Transform features, and 0.58 for a hybrid rule-based neural network classification using a small number of hand-designed features. CONCLUSIONS: We find that malignancy classification performance on the CBIS-DDSM dataset using segmentation-free BoVW features is comparable to that of the best segmentation-dependent methods we study, but also observe that a common segmentation-free CNN model substantially and significantly outperforms each of these (p

Although the past decade has seen important advances in prosthetic technologies, grasping household objects with an artificial hand still requires significant skill and effort for an amputee to regulating hand behaviour. A solution to this problem is to automate the process by using vision sensors that determine the object's orientation and optimal grasp procedure. In this paper, we use a neuromorphic dynamic vision sensor (DVS) to assist amputees with object grasping. Event-driven sensors such as the DVS have gained popularity in recent years as an alternative to conventional frame-based sensors due to their low-power consumption and low-latency. Here, we use event data from a DVS to find a grasp-appropriate orientation for the object and subsequently its optimal grasp type. Our estimation technique exploits general assumptions such as object symmetry and grasps preference to be along the smallest major dimension of an object. The grasp type is determined through a combination of multiple convolutional neural network (CNN) classifiers. We evaluated our grasp estimation methodology on a set of 20 household objects. The results of this study show that 96.25% of the estimated orientations were within ±10° of the actual orientations. In addition, our grasp detection method yielded a 99.47% accuracy on unseen object classes.

In this demonstration we show improved grasping capabilities of household objects by an upper limb prosthesis using the unique features of a dynamic vision sensor (DVS). With the low-power consumption and non-redundant data output of a DVS, our system is able to perform real-time object classification and optimal grasp orientation. This enables an amputee to simply approach an object to be grasped by the prosthetic, while our novel algorithm assists with task completion. Our approach is motivated with the goal to reduce cognitive load required by amputees to control prosthetic hands.

Neuromorphic image sensors produce activity-driven spiking output at every pixel. These low-power consuming imagers which encode visual change information in the form of spikes help reduce computational overhead and realize complex real-time systems; object recognition and pose-estimation to name a few. However, there exists a lack of algorithms in event-based vision aimed towards capturing invariance to transformations. In this work, we propose a methodology for recognizing objects invariant to their pose with the Dynamic Vision Sensor (DVS). A novel slow-ELM architecture is proposed which combines the effectiveness of Extreme Learning Machines and Slow Feature Analysis. The system, tested on an Intel Core i5-4590 CPU, can perform 10,000 classifications per second and achieves 1% classification error for 8 objects with views accumulated over 90 degrees of 2D pose.