Search

This work mainly concentrated on the structural and optical properties of TiO2-doped transparent glass-ceramics embedding ZnO quantum dots (QDs). The results show that the concentration of non-bridging oxygen and related defects in the glasses increases and thermal parameters decreases by the introduction of TiO2. The surface defect concentration of ZnO QDs is greatly improved and the size of ZnO QDs is reduced by heat treatment below the Tg and the introduction of TiO2. The green emission of glasses is mainly derived from oxygen vacancy defects and the yellow emission of glass-ceramics primarily originated from oxygen interstitial defects which show a tendency of enhancing first and then reducing with the increase of TiO2 content. Among them, the optimal content for the strongest emission intensity of TiO2-doped glasses and glass-ceramics was determined to be 1.0 mol%, and the relative exceeds 10 times and 5.4 times that of TiO2-free glasses and glass-ceramics, respectively. The average size of ZnO QDs (∼2.1 nm) remarkably close to its exciton Bohr radius is successfully prepared by the optimal heat treatment condition (630 °C/8 h, below the Tg). The results show that these glass-ceramics embedding ZnO QDs are good yellow emission materials and have application prospects in WLEDs.

ZrO2-doped transparent silicate glasses and glass-ceramics embedding ZnO nano-crystallines were synthesized. The effects of ZrO2 addition on the structure and optical properties of glass-ceramics were studied by DSC, XRD, SEM, IR, and PL spectra. The DSC results show that the crystallization ability of glass enhances with the addition of ZrO2. All the glass-ceramics after being heat-treated at 730 °C for 8 min can maintain a high transmittance in the visible range (>80%). With the increase of ZrO2 content, the grain size of ZnO nano-crystallines shows a tendency of decline, while the bandgap and the defect-related emission (580 nm) intensity attributed to ZnO is enhanced. When the ZrO2-doping content is 2.5 mol%, the yellow emission (580 nm) intensity of ZnO crystals reaches the highest under excitation at 345 nm, which is 21 folds higher than that of ZrO2-free glass-ceramics. These results imply that the glass-ceramics embedding ZnO nano-crystallines doped with ZrO2 would be a prospective candidate for yellow-light emitters.

Detection of voice changes in Parkinson’s Disease (PD) patients would make it possible for early detection and intervention before the onset of disabling physical symptoms. This study explores static and dynamic speech features relating to PD detection. A comparative analysis of the articulation transition characteristics shows that the number of articulation transitions and the trend of the fundamental frequency curve are significantly different between HC speakers and PD patients. Motivated by this observation, we propose to apply Bidirectional long-short term memory (LSTM) model to capture time-series dynamic features of a speech signal for detecting PD. The dynamic speech features are measured based on computing the energy content in the transition from unvoiced to voiced segments (onset), and in the transition from voiced to unvoiced segments (offset). Under the two evaluation methods of 10-fold cross validation (CV) and splitting the dataset without samples overlap of one individual, the experimental results show that the proposed method remarkably improves the accuracy of PD detection over traditional machine learning models using static features.

In recent years, microtransformation has gradually become a renewal method adapted for the development of Chinese cities and has been promoted by many city managers. This research regards ...

This paper studies a novel recurrent neural network (RNN) with hyperbolic secant (sech) in the gate for a specific medical application task of Parkinson’s disease (PD) detection. In detail, it focuses on the fact that patients with PD have motor speech disorders, by converting the voice data into black-and-white images of a recurrence plot (RP) at specific time intervals and constructing the detection model that combines RNN and convolutional neural network (CNN), the study evaluates the performance of the RNN with sech gate compared with long short-term memory (LSTM) and gated recurrent unit (GRU) with conventional gates. As a result, the proposed model obtained similar results to LSTM and GRU (an average accuracy of about 70%) with less hyperparameters, resulting in faster learning. In addition, in the framework of the RNN with sech in gate, the accuracy obtained by using tanh as the output activation function is higher than using the relu function. The proposed method will see more improvement by increasing the data in the future. More analysis on the input sound type, the RP image size, and the deep learning structures will be included in our future work for further improving the performance of PD detection from voice.

In patients with stroke, damage to the central nervous system (CNS) can affect the postural stability and increase the risk of falling. Therefore, accurately assessing the balance is important to understand the type, extent, and causes of balance deficit, and to identify individualized interventions. Clinical assessment methods for balance function can be broadly divided into observation, scale assessment, and balance instrument testing. Here, a clinical protocol is presented for static and dynamic balance assessment in stroke patients, which includes three semiquantitative balance function scale assessments (i.e., Berg Balance Scale, Timed Up and Go Test, and Fugl-Meyer Assessment) and three quantitative instrumental balance evaluation (i.e., Stability Assessment Module, Proprioceptive Assessment Module, and Limit of Stability Module). It is recommended that clinicians consider the use of both classic clinical balance scales and instrumental balance measurements when assessing stroke patients to improve the accuracy of assessments, leading to a better individualized treatment plan.

In this study, the transparent glass-ceramics containing spinel ZnGa2O4 nanocrystals were obtained by controlling crystallization. XRD, SEM, and TEM results show that after heat treatment at 750–950 °C, ZnGa2O4 nanocrystals with spinel structure can be precipitated in the glass matrix. As the heat treatment temperature increases, the nanocrystal size of ZnGa2O4 in the glass matrix shows an increasing trend, which results in a slight decrease in the transmittance of the glass-ceramics. However, it still maintains high optical transparency in the visible-light range. Under the excitation of 281 and 355 nm ultraviolet light, the transparent glass-ceramics embedding ZnGa2O4 nanocrystals show wide emission bands. Moreover, with the increase of heat-treatment temperatures, the CIE chromaticity coordinates of the glass-ceramics gradually move from the green-light region towards the blue-light region, which indicates that the transparent ZnGa2O4 glass-ceramics have potential applications in blue-green laser and color display.

Recent multi-person pose estimation networks rely on sequential downsampling and upsampling procedures to capture multi-scale features and stacking basic modules to reassess local and global contexts. However, the network parameters become huge and difficult to be trained under limited computational resource. Motived by this observation, we design a lite version of Hourglass module that uses hybrid convolution blocks to reduce the number of parameters while maintaining performance. The hybrid convolution block builds multi-context paths with dilated convolutions with different rates which not only reduces the number of parameters but also enlarges the receptive field. Moreover, due to the limitation of heatmap representation, the networks need extra and non-differentiable post-processing to convert heatmaps to keypoint coordinates. Therefore, we propose a simple and efficient operation based on integral loss to fill this gap specifically for bottom-up pose estimation methods. We demonstrate that the proposed approach achieves better performance than the baseline methods on the challenge benchmark MSCOCO dataset for multi-person pose estimation.

Bisphenol A (BPA) is an endocrine-disrupting chemical. Studies have shown that the exposure to BPA is associated with attention-deficit/hyperactivity disorder (ADHD) during adolescent development. However the direct clinical evidence is limited. To investigate the possible association between environmental BPA exposure and the altered behavior of children, a case-control study was conducted with children aged 6-12 years in Guangzhou, China. Two hundred fifteen children diagnosed with ADHD and 253 healthy children from Guangzhou were recruited as the case and control groups, respectively. Urinary BPA and 8-hydroxy-2'-deoxyguanosine (8-OHdG, a biomarker of oxidative DNA damage) concentrations were determined by high-performance liquid chromatography/tandem spectrometry. The results showed that concentrations of urinary BPA for the case group were significantly higher than those for the control group (3.44 vs 1.70 μg/L; 4.63 vs 1.71 μg/g Crt. p < .001). A stepwise increase in the odds ratios for ADHD was observed with the increasing quartiles of children's urinary BPA (first quartile: reference category; second quartile adjusted OR: 1.79, 95% CI: 0.95-3.37; third quartile adjusted OR: 7.44, 95% CI: 3.91-14.1; fourth quartile adjusted OR: 9.41, 95% CI: 4.91-18.1). When the BPA levels were stratified by gender, the odds of ADHD among boys and girls increased significantly with urinary BPA concentrations (adjusted OR: 4.58, 95% CI: 2.84-7.37; adjusted OR: 2.83, 95% CI: 1.17-6.84). Urinary 8-OHdG concentrations in the ADHD children were significantly higher than those in the control group. Furthermore, the linear regression analysis results indicated that a significant relationship existed between BPA exposure and 8-OHdG levels (R = 0.257, p < .001). Our findings provide direct evidence that childhood BPA exposure may be related to ADHD and 8-OHdG concentrations for children. Moreover, BPA exposure could increase the higher occurrence of ADHD for boy than for girls.

This research uses a very-large amount of heart rate data (approximately 10 millions entries) collected on one million persons to investigate heart rate trends in daily life. Trends were examined regarding both human characteristics (gender, age, BMI), and environmental properties (hour of the day, day of the week, season). Concerning the relation between heart rate and gender, women’s heart rate trend was significantly lower compared to men’s. As for the relation between heart rate and age, on one hand, women’s heart rate tends to become lower with aging. On the other hand, men’s heart rate tends to keep increasing until the 50’s. Also, men’s heart rate is lower than women’s in the 10’s and 20’s, the trend becomes opposite from the 30’s. In the relation between heart rate and BMI (Body Mass Index), heart rate is significantly higher for users with higher BMI. Regarding diurnal fluctuation of heart rate, it has been confirmed to take its lowest between 3 A.M. and 6 A.M., then increasing regularly until the peak between noon and 3 P.M., after which it decreases during the evening through the night. As for in-week fluctuation, the trend of higher heart rate during the week and lower heart rate during week-ends and holidays has been comfirmed. Specialy, heart rate remarkably increases from Sunday to Monday. Finally, concerning seasonal fluctuation trend, heart rate is decreasing from summer to autumn, and is at his highest level in winter. The results of this research tend to correspond with previous studies that have been carried out on smaller samples and fewer subjects, demonstrating the credibilitay and validity of the data measured by our system. Moreover, on a public health point of view, we can say that it is important to take into account various human properties and spatio-temporal characteristics when having a comparative analysis of people’s health distribution. We think the knowledge obtained in this research can be applied practically to the whole society as an indication for health care and various diseases prevention.

In order to measure real life prospective memory (PM) ability in laboratory setting, previous studies have used photograph-based, video-based and virtual reality (VR)-based to simulate real life environment. In this study, we used the Functional Near-Infrared Spectroscopy (fNIRS) to record participants’ brain activities when they are performing PM tasks in immersive VR environment. 10 participants were asked to perform PM tasks in a virtual environment while wearing a VR headset and an fNIRS device. By comparing the hemodynamic changes of rostral prefrontal cortex (BA10) of the PM component and the ongoing component, the result shows that VR PM task can induce the activation of BA10. In addition, by making use of the fNIRS technique, this study provides a feasible way to measure BA10 activity in simulated real life environment using immersive VR device.

This paper presents a control approach for bounding gait of quadruped robots by applying the concept of Virtual Constraints (VCs). A VC is a relative motion relation between two related joints imposed to the robots in terms of a specified gait, which can drive the robot to run with desired gait. To determine VCs for highly dynamic bounding gait, the limit cycle motions of the passive dynamic model of bounding gait are analyzed. The leg length and hip/shoulder angle trajectories corresponding to the limit cycles are parameterized by leg angles using 4 th-order polynomials. In order to track the calculated periodic motions, the polynomials are imposed on the robot as virtual motion constraints by a high-level state machine controller. A bounding speed feedback strategy is introduced to stabilize the robot running speed and enhance the stability. The control approach was applied to a newly designed lightweight bioinspired quadruped robot, AgiDog. The experimental results demonstrate that the robot can bound at a frequency up to 5 Hz and bound at a maximum speed of 1.2 m·s−1 in sagittal plane with a Froude number approximating to 1.

A drop-on-demand electromagnetic printing process has been developed to deposit metallic droplets. An external electromagnetic field and an internal pulsed current pass through the metallic liquid are introduced to make the liquid driven by the resulting electromagnetic force. Experimental results show that the printing frequency equals to the power frequency, and can be larger than 150 Hz when the pulsed width is 5 ms. The printed mercury droplets are uniform in diameter with a mean variation of 2.27%. Smaller nozzle, shorter pulse width and lower magnetic field intensity favor tiny droplets. Besides, printing of molten solder at a temperature of 290 °C is achieved. Our results show promise for electromagnetic printing as a potential tool in drop-on-demand printing fabrication of high-melting-point metals.

This study used a small wearable heart rate sensor to monitor the daily autonomic function of 600 subjects from across all age groups over a prolonged period of time. The results indicated that the LF/HF ratio (Heart Rate Variability, LF: frequencies between 0.04 Hz - 0.15 Hz, HF: frequencies between 0.15 Hz - 0.4 Hz) an indicator of balance in the autonomic nervous system, tended to peak for subjects in their 40’s and decline thereafter. This conceivably may be partially due to the causes for concern and stress changing and/or declining for the group aged 50-plus. A decline in diurnal variation of autonomic nervous activity was also exhibited in subjects aged 50 and up, showing a tendency for decline in the function of rising sympathetic nerve activity particularly in the morning. It is conceivable that this stems from a decline in the responsiveness of the autonomic nervous system. Subjects in the 50-plus group furthermore exhibited a tendency for declining variation in autonomic nervous activity between sleeping and waking hours. This phenomenon was consistent with the tendency for there to be a rise in wake after sleep onset coupled with a decline in slow-wave sleep in middle- to old-age.

This study describes the development of a cervical traction therapy simulation model that evaluates two types of the traction positions, namely the sitting position and the inclined position. An anatomically correct human skeleton model and two mechanical traction device models were constructed in simulations using a physics engine. The anterior and posterior intervertebral separations were measured at both positions with a series of traction forces (60N to 200N) and traction angles (10° to 40° ). The result suggested that the sitting position caused the subject to lean forward and as a result led to excessive anterior compression when traction angle is over 20 degrees. The inclined position creates greater intervertebral separations on both the anterior and posterior sides than the sitting position. This suggests that the inclined position may be more effective in increasing intervertebral separation than the sitting position.

Increasing attention is being paid to the research of heart-rate-based training to improve the safety and efficiency of exercises. The present study develops a walking and running exercise training system, which is able to provide a more human friendly way to adjust exercise intensity by measuring user's heart rate in real-time. Specifically, the proposed training system makes it possible for the user to control the speed of the treadmill based on the results of heart rate analysis. Two types of exercise intensity adjustment approaches are tested by four health human subjects. From the frequency analysis of heart rates, it was found that, for the same type of intensity adjustment, the subjects may feel different stress with respect to whether he/she has been used to do exercise or not.

A computer simulation model was developed to study the behavior of the cervical spine during cervical traction therapy in inclined and sitting traction positions. The model improved upon an old model with additional components to represent the behavior of the intervertebral discs and the posterior ligaments. The simulation result of the new model was compared against the cervical traction data from a radiographic experiment in both positions. The simulation results of the old model and new model were compared to illustrate the improvement. Using the new model, we compared the timing response of cervical traction in the inclined and sitting positions.

[Purpose] The purpose of this study was to clarify whether walking speed affects acceleration variability of the head, lumbar, and lower extremity by simultaneously evaluating of acceleration. [Subjects and Methods] Twenty young individuals recruited from among the staff at Kurashiki Heisei Hospital participated in this study. Eight accelerometers were used to measure the head, lumbar and lower extremity accelerations. The participants were instructed to walk at five walking speeds prescribed by a metronome. Acceleration variability was assessed by a cross-correlation analysis normalized using z-transform in order to evaluate stride-to-stride variability. [Results] Vertical acceleration variability was the smallest in all body parts, and walking speed effect had laterality. Antero-posterior acceleration variability was significantly associated with walking speed at sites other than the head. Medio-lateral acceleration variability of the bilateral hip alone was smaller than the antero-posterior variability. [Conclusion] The findings of this study suggest that the effect of walking speed changes on the stride-to-stride acceleration variability was individual for each body parts, and differs among directions.

This research uses a large amount of autonomic nervous system data (approximately 100,000 entries) to investigate the relationship between human autonomic nervous activity and behaviors, daily and regional changes. Data were measured via a heart rate variability analysis system that utilizes the camera of smartphones. This system was developed by the authors during previous research. The relations between autonomic nervous system and behaviors, total power and sympathetic nervous activity were found to rise after waking, while during leisure time, the total power rises and sympathetic nervous activity is inhibited. Concerning the relationship between autonomic nervous system and day of the week, it was found that total power decreases from the middle through the latter part of the week (namely, Wednesday, Thursday, and Friday), while it rises on Saturday, while the sympathetic nervous activity is suppressed on Saturday. Regarding the relationship between autonomic nervous system and region, it was found that total power is lower in the Kanto region of Japan than in others. This study also shows statistical proof (using a large amount of measurement data) to ideas held by the public for years. Thus, the data can be considered meaningful to the society, and the authors hope that it helps to improve work-life balance.

We developed a smart-phone based system to measure the activities of autonomic nervous system during everyday life. Using commonly marketed smart phones, by touching your fingertips on the phone’s camera over a short time of about 30 seconds, it will detect changes in the brightness of the blood flow and in turn analyze your heart rate variability. By using this system, about 100,000 cases were measured and from this large amount of data regarding heart rate variability, we evaluated the autonomic nervous function in their daily life. As a result, for the correlation between autonomic nervous system and age, we found that as the increase of age, the total power becomes decreased and the sympathetic nervous system tends to increase between thirties and fifties. For the correla-tion between autonomic nervous system and BMI (Body Mass Index), it is found that in general, the higher the BMI, the lower the total power and the stronger the sympathetic nervous system. In other words, people who are fat are lower about the total power and stronger about the sympathetic nervous system. In addition, for the correlation between autonomic nervous system and one day life, it is that found that total power and sympathetic function tend to increase, while as evening ap-proaches, sympathetic function tends to become suppressed.

[Purpose] This study examined age-related differences in muscle control for support and propulsion during walking in both males and females in order to develop optimal exercise regimens for muscle control. [Subjects and Methods] Twenty elderly people and 20 young people participated in this study. Coordinates of anatomical landmarks and ground reaction force during walking were obtained using a 3D motion analysis system and force plates. Muscle forces during walking were estimated using OpenSim. Muscle modules were obtained by using non-negative matrix factorization analysis. A two-way analysis of covariance was performed to examine the difference between the elderly and the young in muscle weightings using walking speed as a covariate. The similarities in activation timing profiles between the elderly and the young were analyzed by cross-correlation analysis in males and females. [Results] In the elderly, there was a change in the coordination of muscles around the ankle, and muscles of the lower extremity exhibited co-contraction in late stance. Timing and shape of these modules were similar between elderly and young people. [Conclusion] Our results suggested that age-related alteration of muscle control was associated with support and propulsion during walking.

This research uses our previously-developed smartphone camera-based heart rate change analysis system to survey the correlation between weather patterns and the autonomic nervous activity across a big data set of approximately 200,000 entries. The results showed a trend in which a significant decrease was seen in sympathetic nervous activity in both males and females—the higher the temperature. In addition, a significant increase was seen in the sympathetic nervous system in both males and females—the higher the atmospheric pressure. Lastly, a significant decrease was seen in the sympathetic nervous system in both males and females—the more precipitation there was. These results accord with prior research and with human biological phenomena, and we were able to use a data set of approximately 200,000 entries to statistically demonstrate our hypothesis. We believe this represents a valuable set of reference data for use in the health care.

The exponentially increasing size of biomedical literature and the limited ability of manual curators to discover protein&ndash, protein interactions (PPIs) in text has led to delays in keeping PPI databases updated with the current findings. The state-of-the-art text mining methods for PPI extraction are primarily based on deep learning (DL) models, and the performance of a DL-based method is mainly affected by the architecture of DL models and the feature embedding methods. In this study, we compared different architectures of DL models, including convolutional neural networks (CNN), long short-term memory (LSTM), and hybrid models, and proposed a hybrid architecture of a bidirectional LSTM+CNN model for PPI extraction. Pretrained word embedding and shortest dependency path (SDP) embedding are fed into a two-embedding channel model, such that the model is able to model long-distance contextual information and can capture the local features and structure information effectively. The experimental results showed that the proposed model is superior to the non-hybrid DL models, and the hybrid CNN+Bidirectional LSTM model works well for PPI extraction. The visualization and comparison of the hidden features learned by different DL models further confirmed the effectiveness of the proposed model.

In this paper, we propose a cluster-wise feature aggregation network that exploits multi-level contextual association for multi-person pose estimation. The recent popular approach for pose estimation is extracting the local maximum response from each detection heatmap that trained for a specific keypoint type. To exploit more contextual information, our network simultaneously learns complementary semantic information to encourage the detected keypoints subject to a certain contextual constraint. Specifically, our network uses dense and sparse branches to generate paired multi-peak detection heatmaps for clusters of keypoints. To enhance the feature passing through the network, we aggregate information from different branches. The in-branch aggregation enriches the detection features in each branch by absorbing the holistic human region attention. The cross-branch aggregation further strengthens the detection features by fusing global and local context information between dense and sparse branches. We demonstrate competitive performance of our network on the benchmark dataset for multi-person pose estimation.

Familial hypercholesterolemia (FH) is mostly caused by low-density lipoprotein receptor (LDLR) mutations and results in an increased risk of early-onset cardiovascular disease due to marked elevation of LDL cholesterol (LDL-C) in blood. Statins are the first line of lipid-lowering drugs for treating FH and other types of hypercholesterolemia, but new approaches are emerging, in particular PCSK9 antibodies, which are now being tested in clinical trials. To explore novel therapeutic approaches for FH, either new drugs or new formulations, we need appropriate in vivo models. However, differences in the lipid metabolic profiles compared to humans are a key problem of the available animal models of FH. To address this issue, we have generated a human liver chimeric mouse model using FH induced pluripotent stem cell (iPSC)-derived hepatocytes (iHeps). We used Ldlr(-/-)/Rag2(-/-)/Il2rg(-/-) (LRG) mice to avoid immune rejection of transplanted human cells and to assess the effect of LDLR-deficient iHeps in an LDLR null background. Transplanted FH iHeps could repopulate 5-10% of the LRG mouse liver based on human albumin staining. Moreover, the engrafted iHeps responded to lipid-lowering drugs and recapitulated clinical observations of increased efficacy of PCSK9 antibodies compared to statins. Our human liver chimeric model could thus be useful for preclinical testing of new therapies to FH. Using the same protocol, similar human liver chimeric mice for other FH genetic variants, or mutations corresponding to other inherited liver diseases, may also be generated.

Solder jetting is a growing technology for microelectronic packaging in diverse range of applications. In this paper, an electromagnetic jetting device is developed to direct deposit solder droplets. Experimental results show that the electromagnetic jetting is feasible to deposit solder, and uniform solder balls can be obtained. The process of jetting and deposition of solder droplets is analyzed, and the influence of the nozzle diameter on the droplet diameter is obtained. Our results show promise for electromagnetic jetting as a potential tool in fabrication of metal droplets.

Featured Application: The proposed Uncertainty Flow framework may benefit the facial analysis with its promised elevation in discriminability in multi-label affective classification tasks. Moreover, this framework also allows the efficient model training and between tasks knowledge transfer. The applications that rely heavily on continuous prediction on emotional valance, e.g., to monitor prisoners’ emotional stability in jail, can be directly benefited from our framework. Abstract: To lower the single-label dependency on affective facial analysis, it urges the fruition of multi-label affective learning. The impediment to practical implementation of existing multi-label algorithms pertains to scarcity of scalable multi-label training datasets. To resolve this, an inductive transfer learning based framework, i.e.,Uncertainty Flow, is put forward in this research to allow knowledge transfer from a single labelled emotion recognition task to a multi-label affective recognition task. I.e., the model uncertainty—which can be quantified in Uncertainty Flow—is distilled from a single-label learning task. The distilled model uncertainty ensures the later efficient zero-shot multi-label affective learning. On the theoretical perspective, within our proposed Uncertainty Flow framework, the feasibility of applying weakly informative priors, e.g., uniform and Cauchy prior, is fully explored in this research. More importantly, based on the derived weight uncertainty, three sets of prediction related uncertainty indexes, i.e., soft-max uncertainty, pure uncertainty and uncertainty plus are proposed to produce reliable and accurate multi-label predictions. Validated on our manual annotated evaluation dataset, i.e., the multi-label annotated FER2013, our proposed Uncertainty Flow in multi-label facial expression analysis exhibited superiority to conventional multi-label learning algorithms and multi-label compatible neural networks. The success of our proposed Uncertainty Flow provides a glimpse of future in continuous, uncertain, and multi-label affective computing.

Firefly algorithm is widely used in the tackling of optimization problems. This paper proposed a global path planning algorithm based on the modified firefly algorithm (PPMFA) in order to find an optimal path under multiple objective functions. Owning to the low convergence speed and inaccurate local search ability of the standard firefly algorithm (SFA), the Gaussian random walk is proposed to replace the fixed step size of the SFA so as to improve the random search ability. Incorporating a double check method during the iteration process, the success rate of the movement of each firefly is increased. In order to measure the distance between any two fireflies (which represent two paths), the conception and calculation method of Path Center (PC) is proposed. Simulation results show that compared with the particle swarm optimization (PSO) and SFA, the proposed algorithm outperforms both of the algorithms in terms of convergence speed and accuracy.

Vibro-acoustic modulation technique is a nonlinear ultrasonic method in nondestructive testing. This technique detects the defects by monitoring the modulation components generated by the interaction between the vibration and the ultrasound wave due to the nonlinear material behaviour caused by the damage. In this work, a swept frequency signal was used as high frequency excitation, then the Hilbert transform based amplitude and phase demodulation and synchronous demodulation (SD) were used to extract the modulation information from the received signal, the results were graphed in the time-frequency domain after the short time Fourier transform. The demodulation results were quite different from each other. The reason for the difference was investigated by analysing the demodulation process of the two methods. According to the analysis and the subsequent verification test, it was indicated that the SD method was more proper for the test and a new index called MISD was defined to evaluate the structure qua...

[Purpose] This study examined the relationships between joint moment and the control of the vertical ground reaction force during walking in the elderly and young male and female individuals. [Subjects and Methods] Forty elderly people, 65 years old or older (20 males and 20 females), and 40 young people, 20 to 29 years old (20 males and 20 females), participated in this study. Joint moment and vertical ground reaction force during walking were obtained using a 3D motion analysis system and force plates. Stepwise linear regression analysis determined the joint moments that predict the amplitude of the vertical ground reaction force. [Results] Knee extension moment was related to the vertical ground reaction force in the young males and females. On the other hand, in the elderly females, hip, ankle, and knee joint moments were related to the first peak and second peak forces, and the minimum value of vertical ground reaction force, respectively. [Conclusion] Our results suggest that the young males and females make use of the knee joint moment to control of the vertical ground reaction force. There were differences between the elderly and the young females with regard to the joints used for the control of the vertical ground reaction force.

Sit to stand (STS) transfer is probably the most common task during the daily life. We have developed an assistive robot which can provide physical support to aid patients and the elderly with the physical impairment during both the walk and the STS transfer. In this paper, we will focus on the STS transfer function which is realized by autonomous control and posture control that means the robot tracks human intention and motion during the STS transfer. The assistive robot consists of almost the mobile platform mechanism, STS support mechanism and the interaction module. Several robot-assisted STS transfer experiments were carried out. By analyzing the signals of the sensors, the experimental results show that the STS transfer function of the assistive robot are determined by the linear actuator and the control algorithm for the STS transfer. The acceptable smoothness and compliance are achieved. Finally, future research work to improve the performances is addressed.

In order to defend adversarial attacks in computer vision models, the conventional approach arises on actively incorporate such samples into the training datasets. Nonetheless, the manual production of adversarial samples is painful and labor intensive. Here we propose a novel generative model: Convolutional Autoencoder Model to add unsupervised adversarial training, i.e., the production of adversarial images from the encoded feature representation, on conventional supervised convolutional neural network training. To accomplish such objective, we first provide a novel statistical understanding of convolutional neural network to translate convolution and pooling computations equivalently as a hierarchy of encoders, and sampling tricks, respectively. Then, we derive our proposed Convolutional Autoencoder Model with the ‘adversarial decoders’ to automate the generation of adversarial samples. We validated our proposed Convolutional Autoencoder Model on MNIST dataset, and achieved the clear-cut performance improvement over the normal Convolutional Neural Network.

Hand segmentation is one of the most fundamental and crucial steps for egocentric human-computer interaction. The special egocentric view brings new challenges to hand segmentation task, such as the unpredictable environmental conditions. The performance of traditional hand segmentation methods depend on abundant manually labeled training data. However, these approaches do not appropriately capture the whole properties of egocentric human-computer interaction for neglecting the user-specific context. It is only necessary to build a personalized hand model of the active user. Based on this observation, we propose an online-learning hand segmentation approach without using manually labeled data for training. Our approach consists of top-down classifications and bottom-up optimizations. More specifically, we divide the segmentation task into three parts, a frame-level hand detection which detects the presence of the interactive hand using motion saliency and initializes hand masks for online learning, a superpixel-level hand classification which coarsely segments hand regions from which stable samples are selected for next level, and a pixel-level hand classification which produces a fine-grained hand segmentation. Based on the pixel-level classification result, we update the hand appearance model and optimize the upper layer classifier and detector. This online-learning strategy makes our approach robust to varying illumination conditions and hand appearances. Experimental results demonstrate the robustness of our approach.

Taxi plays an important role in public transportation. However, in real life, it usually takes a long time for passengers to wait for an unoccupied taxi, while number of no-load taxi is very high. Optimized scheme is required to handle global taxi dispatching. The paper presents a novel method of taxi dispatching based on Artificial Fish Swarm Algorithm (AFSA). To avoid slow convergence and falling into local optimum, standard AFSA is improved using strategies of optimizing visual range, step size and swimming behavior. We compare our improved algorithm with standard AFSA and Particle Swarm Optimization through the evaluation with real data of GPS trajectory of 12,000 taxis in Beijing and make performance analysis of effects of parameter configuration to dispatching algorithm. The experimental results show good effectiveness of our method.

This study aims to verify the accuracy of a cervical traction therapy simulation model by comparing the intervertebral separations of six normal male adults when traction applied to their cervical spines. The subjects were tested on two mechanical traction devices, representing the inclined and sitting positions. Radiographic images of their cervical spines were taken before, during and after traction. The observed changes of the cervical spine were used as input to adjust parameters of the traction therapy model, which contains a cervical spine built with springs and dampers and two traction devices in inclined and sitting positions. A series of traction forces and traction angles were applied to the model to simulate the actual parameters used in the experiment and the new model was used to evaluate the two traction positions. The result suggests that inclined position not only creates greater intervertebral separations on the posterior sides, it also provides a more accurate and smoother transition when switching between different traction angles.

To improve the jetting performance of liquid metals, an electromagnetic micro-jetting (EMJ) valve that realizes drop-on-demand (DOD) jetting while not involving any valve core or moving parts was designed. The influence of the lead angle of the nozzle on the jetting of liquid metal gallium (Ga) was investigated. It was found that the Lorentz force component parallel to the nozzle that jets the electrified liquid Ga is always larger than its internal friction; thus, jet can be generated with any lead angle but with different kinetic energies. Experimental results show that the mass of the jetting liquid, the jetting distance, the initial velocity of the jet, and the resulting kinetic energy of the jet increase first and then decrease. When the lead angle is 90°, the mass of the jetting liquid and the kinetic energy are at their maximum. When the angle is 80°, the initial velocity achieves its maximum, with a calculated value of 0.042 m/s. Moreover, very close and comparatively high kinetic energies are obtained at 80° and 90°, indicating that angles in between this range can produce a preferable performance. This work provides an important theoretical basis for the design of the EMJ valve, and may promote the development and application of micro electromagnetic jetting technology.

Techniques for measuring physiological signals are gathering a lot of attention in recent years. In many cases, the techniques require to make direct contact between the physiological signal sensors with the subject's skin. Several studies were proposed to measure heart rate (HR) with standard web cameras under the condition that the subject is keeping the quiescent state. In this paper, we propose a new technique for measuring HR which allows the subject head movements during water-bed massage. For comparison purpose, HR was measured simultaneously using an electrocardiography (ECG) device during all sessions of the experiment. The result shows that our proposed technique is able to improve the accuracy of HR measurement and reduce the processing time to nearly one-third of the existing method. Our approach could be beneficial for applications that require real-time HR monitoring during subject's head is moving.

A cane-type walking-aid robot with a ball joint is developed for assisting old people's walking. Considering both the overturn of robot and the stability of human user, this paper presents a new concept of Human-Robot Coordination Stability (HRCS) which can describe the stability of the integrated human-robot system during the user operating the cane robot. HRCS is characterized by a stability region generated from the user's stability measured by posture information and the robot stability obtained by Force-Angle (FA) measure method. If the human-robot system runs out of the stability region, the fall tendency is then detected. To guarantee the safety of both the user and robot, a fall prevention control strategy is also proposed. The fall prevention control is executed to ensure the system staying in the stability region when any fall tendency is detected. Finally, we conducted experiments in real cane robot and the results illustrate the validity of the fall detection and prevention method.

Introduction: To date, the anti-gravity treadmill (AlterG), as a representative method of Lower body positive pressure (LBPP) treadmills, has been rarely reported for knee osteoarthritis (KOA) rehabilitation. The purpose of this case study was to setup the clinical protocol example for AlterG intervention on KOA and evaluate treatment effectiveness by 3D gait analysis combined with free EMG to explore the kinematic gait parameter changes. Patient concerns: A 65-year-old female patient (BMI = 26, mild obesity) undergoing “more than 7 years of KOA.” The activity of the right knee joint was obviously limited and she suffered from severe pain over the past month. Diagnosis: Due to the patient's symptoms and radiographic findings, she was diagnosed with acute attack of KOA. Interventions: The patient has performed clinical function evaluation and gait analysis combined at pretreatment, post-treatment, and 4 months follow-up assessment. AlterG training was performed 6 days/week for 2 weeks, with up to 30 min of training per session. The training protocol included two major parts, walking and squatting in AlterG. Outcomes: After 2 weeks of AlterG intervention, the 10-m walking test (10 MWT) and Timed-up-and-go (TUG) test improved significantly post-treatment, whereas the Visual Analog Scale (VAS) score decreased post-treatment. The Modified Barthel Index improved post-treatment and the patient restored basic community walk after treatment. The temporal parameter results showed that stride length (%height), mean velocity (%height), and cadence gradually increased before treatment, after treatment, and at 4-month follow-up. The right range of motion (ROM) of knee flexion-extension were gradually increased. Meanwhile, the synchronized EMG data showed that the RMS (root means square) values of the rectus femoris, semitendinosus, and biceps femoris at post-treatment were improved to different degrees than at pretreatment. Conclusion: We found that for this patient with KOA, AlterG relieved pain, and was also effective at improving spatio-temporal parameters, knee flexion/extension gait pattern, and corresponding muscle strength, thereby restoring certain community activities.

Oxynitride glasses with five different compositions were prepared by melting batches at 1600 °C for 2 h under N 2 atmosphere in a Si–Mo-heated resistance furnace. The amorphous nature of these samples was verified by an X-ray diffractometer with Cu/Kα radiation. UV/visible wavelength region light transmission spectra of samples showed that the transmittance of the obtained glasses (1 mm, 2 mm and 3 mm thickness) can reach up to 80% in the visible region (above 400 nm). The Vickers indentations were analyzed by the scanning electron microscopy (SEM). The ballistic resistance represented by D rel value was assessed by physical and mechanical properties including density, hardness, elastic modulus and fracture toughness. It was found that the D rel value of oxynitride glasses can reach up to 2.56 and is higher than that of sapphire, which is 2.4.

Fundamental Research Funds for the Central Universities [2010121039]; National Natural Science Foundation for Youth of China [51105320]

To make up for the insufficiency and instability of contact dispensers that are used for fluid dispensing in microelectronic packaging, a noncontact jetting dispenser driven by a piezostack actuator is introduced in this paper. After describing the structural components and operating principle of the dispensing mechanism, a fluid model is presented to discuss the dynamic properties of the fluid and analyze the key parameters of the proposed dispenser. The ANSYS simulation software is used to design the displacement amplifier, which is an important component of the dispenser. The maximum displacement output of 323 μm is obtained by optimizing. Subsequently, the dynamic behavior of the displacement amplifier is measured by an optical displacement sensor. The displacement change with the driving voltage amplitude and frequency is also investigated; the maximum displacement is 320 μm, and the error between the simulation and the measurement result is just 0.75%. In order to verify the practicality of dispenser, experiments are conducted to examine the effects of the driving voltage, backpressure, working temperature, and distance between the nozzle and the collector on the jetting performance and droplet diameter. The dispenser can dispense droplets uniformly and steadily. Its maximum jetting frequency is 65 Hz, and droplets of 1.07-mm diameter are produced by a stainless steel nozzle of 0.25-mm diameter in the experimental study, with the variation of the droplet diameter being within ±2%.

This research proposes a Asymmetry Parametric Excitation for a Biped Robot. In our research, we succeeded in developing an experimental parametric excited walking robot with counter weights and achieved the improvement of energy efficiency. However, the mechanism of this improvement has not been clarified. In this paper, we clarify the relation between the equipment of counter weights and the improvement of energy efficiency. It is found that counter weights create the asymmetry of the center of mass while the robot's posture during the double support phase is symmetric. This asymmetry contributes to the efficiency of restoring mechanical energy based on parametric excitation of the inverted pendulum. Therefore, a biped robot with counter weights can improve walking speed, step length, restored mechanical energy and walking energy efficiency.

This research studies optimal swimming movement, specifically it focuses on the optimal kick timing during dolphin style swimming. In the rule of Breast Stroke Style, it is allowed to hit one dolphin kick during underwater movement. Until December 2014, we couldn't do dolphin kick before hands start to move. Now, dolphin kick can be performed at any time by the rule amendment. In this research, we modeled this style of movement and searched for optimum timing of dolphin kick. We used computational simulation model called SWUM (SWimming hUman Model). From the simulation results, we got the conclusion that we should do dolphin kick 0.4 seconds faster than hands move. In addition, we indicate the timing affects the body posture. These results agree with the current top swimmer's timing.

The present paper investigates pinching movements using an index finger and a thumb actuated by redundant nonlinear digitorum muscles mimicking the configuration of human fingers. A dual-finger mod...