Human Activity Recognition via Score Level Fusion of Wi-Fi CSI Signals.

Wi-Fi signals are ubiquitous and provide a convenient, covert, and non-invasive means of recognizing human activity, which is particularly useful for healthcare monitoring. In this study, we investigate a score-level fusion structure for human activity recognition using the Wi-Fi channel state information (CSI) signals. The raw CSI signals undergo an important preprocessing stage before being classified using conventional classifiers at the first level. The output scores of two conventional classifiers are then fused via an analytic network that does not require iterative search for learning. Our experimental results show that the fusion provides good generalization and a shorter learning processing time compared with state-of-the-art networks.

Wide field-of-hearing metalens for aberration-free sound capture.

Metalenses are instruments that manipulate waves and have exhibited remarkable capabilities to date. However, an important hurdle arises due to the severe hampering of the angular response originating from coma and field curvature aberrations, which result in a loss of focusing ability. Herein, we provide a blueprint by introducing the notion of a wide field-of-hearing (FOH) metalens, designed particularly for capturing and focusing sound with decreased aberrations. Employing an aberration-free planar-thin metalens that leverages perfect acoustic symmetry conversion, we experimentally realize a robust wide FOH capability of approximately 140∘ in angular range. Moreover, our metalens features a relatively short focal length, enabling compact implementation by reducing the aperture-to-hearing plane distance. This is beneficial for space-efficient source-tracking sound sensing. Our strategy can be used across various platforms, potentially including energy harvesting, monitoring, imaging, and communication in auditory, ultrasonic, and submerged environments.

An improved analytic model for designing the polymer-composite stepped-plate transducer using the modified Mindlin plate theory.

A stepped-plate transducer (SPT) uses an extensive radiating plate to produce highly-directional ultrasound beams. In this paper, we present an improved analytical model for designing the polymer-composite stepped-plate transducer (PCSPT). The polymer-composite features the lightweight and flexible properties, and there can be little change in the resonant frequency and mode shape when the steps are attached. With the outstanding merit, it is feasible to construct SPTs with polymer-composite steps without taking the steps into consideration. The modified Mindlin plate theory (MMPT) is applied to improve the accuracy in the equivalent circuit model (ECM) that is used to predict the high-frequency vibratory responses. Our analytical model can be used to design well-tuned SPTs to achieve the desired dynamic responses such as resonant frequencies, mode shape and bandwidth for various high-power ultrasonic applications. We use several numerical design examples to illustrate that the design of the transducer can be accomplished without analyzing the sophisticated stepped-plate's behavior. We also perform a series of experiments to verify that the PCSPT is capable of functioning as a high-power ultrasonic transducer.

Influence of divalent metal ions on E2-induced ER pathway in goldfish (Carassius auratus) hepatocytes.

Metal ions existing in the environment could influence the estrogen pathway in aquatic animal, but the detailed mechanism is still delusive. We here showed that in male Carassius auratus hepatocytes, copper (Cu) or cadmium (Cd), did not directly induce vitellogenin (VTG) expression. Interestingly, co-exposure with Cd²âº (or Cu²âº) and 17-ß-estradiol (E2) greatly increased the VTG level, comparing with single treatment of E2. Meanwhile, Cd²âº or Cu²âº (but not E2) triggers HSP70 expression. But, mixture of Cd²âº or Cu²âº with E2 did not obviously raise HSP70 level. E2 also had no obvious effect on reactive oxygen species. Co-treatment of Cd²âº and E2 showed no obvious increase compared to single treatment with Cd²âº. We further assume that Cd²âº-involved oxidative stress generates misfolded proteins, resulting in the competition of HSP70 proteins from a heterocomplex (with estrogen receptor). Thus, dissociation of the heterocomplex actives the receptor-ligand binding activity and promotes the E2-induced VTG expression.

Estrogenic response in male bullfrog (Rana catesbeiana) hepatocytes after single or combined exposure to cadmium (Cd) and 17beta-estradiol (E2).

Contamination by heavy metals and sex hormones in a water environment is an important health issue. In this study, we investigated the estrogenic effects of cadmium (Cd) administration alone and in combination with 17beta-estradiol (E2) on the hepatocytes of male Bullfrog (Rana catesbeiana). Their vitellogenin (VTG) expression and reactive oxygen species (ROS) were analyzed upon exposure to Cd alone or to both Cd and E2. Our results suggest that the VTG levels induced by the co-treatment of 100 nM E2 and 100 nM CdCl(2) were significantly higher than those induced by 100 nM E2 alone (p < 0.05), and were comparable to vitellogenin induction observed with 1 µM E2. A similar result was observed by western blot analysis in the culture medium of hepatocytes. Meanwhile, Cd (but not E2) increased the ROS levels. These results suggest that Cd has a cooperative effect with E2 in the induction of VTG, thus acting as an estrogenic disruptor. Cd also causes oxidative stress that occurs with the enhanced vitellogenesis.

An Analytic Gabor Feedforward Network for Single-Sample and Pose-Invariant Face Recognition.

Gabor magnitude is known to be among the most discriminative representations for face images due to its space- frequency co-localization property. However, such property causes adverse effects even when the images are acquired under moderate head pose variations. To address this pose sensitivity issue and other moderate imaging variations, we propose an analytic Gabor feedforward network which can absorb such moderate changes. Essentially, the network works directly on the raw face images and produces directionally projected Gabor magnitude features at the hidden layer. Subsequently, several sets of magnitude features obtained from various orientations and scales are fused at the output layer for final classification decision. The network model is analytically trained using a single sample per identity. The obtained solution is globally optimal with respect to the classification total error rate. Our empirical experiments conducted on five face data sets (six subsets) from the public domain show encouraging results in terms of identification accuracy and computational efficiency.

Manifold ranking based scoring system with its application to cardiac arrest prediction: A retrospective study in emergency department patients.

BACKGROUND: The recently developed geometric distance scoring system has shown the effectiveness of scoring systems in predicting cardiac arrest within 72h and the potential to predict other clinical outcomes. However, the geometric distance scoring system predicts scores based on only local structure embedded by the data, thus leaving much room for improvement in terms of prediction accuracy. METHODS: We developed a novel scoring system for predicting cardiac arrest within 72h. The scoring system was developed based on a semi-supervised learning algorithm, manifold ranking, which explores both the local and global consistency of the data. System evaluation was conducted on emergency department patients׳ data, including both vital signs and heart rate variability (HRV) parameters. Comparison of the proposed scoring system with previous work was given in terms of sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV). RESULTS: Out of 1025 patients, 52 (5.1%) met the primary outcome. Experimental results show that the proposed scoring system was able to achieve higher area under the curve (AUC) on both the balanced dataset (0.907 vs. 0.824) and the imbalanced dataset (0.774 vs. 0.734) compared to the geometric distance scoring system. CONCLUSIONS: The proposed scoring system improved the prediction accuracy by utilizing the global consistency of the training data. We foresee the potential of extending this scoring system, as well as manifold ranking algorithm, to other medical decision making problems. Furthermore, we will investigate the parameter selection process and other techniques to improve performance on the imbalanced dataset.

The PARK2 gene is involved in the maintenance of pancreatic ß-cell functions related to insulin production and secretion.

Several association studies have implicated the PARK2 gene that encodes parkin--the key molecule orchestrating the mitochondrial quality control system--as a candidate susceptibility gene for diabetes. A total of 7551 unrelated Korean KARE cohort subjects were analyzed to investigate the association between the PARK2 single nucleotide polymorphism (SNP) and quantitative glycemic traits. Two SNPs, rs10455889 and rs9365294, were significantly associated with fasting plasma glucose level (p=∼1.2×10(-4)) and insulin secretion indices (p=∼7.4×10(-5)) in male KARE subjects. Parkin was expressed predominantly in the rat pancreatic islets. Downregulation of the Park2 gene in rat INS-1 ß-cells resulted in a significant decrease in the glucose-stimulated insulin secretion, intracellular insulin gene expression, and intracellular ATP level. The Park2-depleted ß-cells also exhibited increased mitochondrial fragmentation and ROS production and decreased mitochondrial membrane potential. Both population-based statistical evaluation and experimental evidence demonstrated a fundamental role of the PARK2 gene in the maintenance of ß-cell function.