Search

The intelligent data-driven screening of pathological voice signals is a non-invasive and real-time tool for computer-aided diagnosis that has attracted increasing attention from researchers and clinicians. In this paper, the authors propose multi-domain features and the hierarchical extreme learning machine (H-ELM) for the automatic identification of voice disorders. A sufficient number of sensitive features are first extracted from the original voice signal through multi-domain feature extraction (i.e., features of the time domain and the sample entropy based on ensemble empirical mode decomposition and gammatone frequency cepstral coefficients). To eliminate redundancy in high-dimensional features, neighborhood component analysis is then applied to filter out sensitive features from the high-dimensional feature vectors to improve the efficiency of network training and reduce overfitting. The sensitive features thus obtained are then used to train the H-ELM for pathological voice classification. The results of the experiments showed that the sensitivity, specificity, F1 score, and accuracy of the H-ELM were 99.37%, 98.61%, 99.37%, and 98.99%, respectively. Therefore, the proposed method is feasible for the initial classification of pathological voice signals.

Background and Purpose: Whether imaging parameters would independently predict stroke recurrence in low-risk minor ischemic stroke (MIS) or transient ischemic attack (TIA) according to traditional score system (such as ABCD 2 score, which was termed on the basis of the initials of the five factors: age, blood pressure, clinical features, duration, diabetes) remains unclear. We sought to evaluate the association between imaging parameters and 1-year stroke recurrence in patients with TIA or MIS in different risk stratum stratified by ABCD 2 score. Methods: We included patients with TIA and MIS (National Institutes of Health Stroke Scale score ≤3) with complete baseline vessel and brain imaging data from the Third China National Stroke Registry III. Patients were categorized into different risk groups based on ABCD 2 score (low risk, 0–3; moderate risk, 4–5; and high risk, 6–7). The primary outcome was stroke recurrence within 1 year. Multivariable Cox proportional-hazards regression models were used to assess whether imaging parameters (large artery stenosis, infarction number) were independently associated with stroke recurrence. Results: Of the 7140 patients included, 584 patients experienced stroke recurrence within 1 year. According to the ABCD 2 score, large artery stenosis was associated with higher stroke recurrence in both low-risk (adjusted hazard ratio, 1.746 [95% CI, 1.200–2.540]) and moderate-risk group (adjusted hazard ratio, 1.326 [95% CI, 1.042–1.687]) but not in the high-risk group ( P >0.05). Patients with multiple acute infarctions or single acute infarction had a higher risk of recurrent stroke than those with no infarction in both low- and moderate-risk groups, but not in the high-risk group. Conclusions: Large artery stenosis and infarction number were independent predictors of 1-year stroke recurrence in low-moderate risk but not in high-risk patients with TIA or MIS stratified by ABCD 2 score. This finding emphasizes the importance of early brain and vascular imaging evaluation for risk stratification in patients with TIA or MIS.

Hydraulic systems faults have the characteristics of being highly concealed and unclear. Due to the characteristics of the complex vibration transmission mechanism and strong nonlinear time-varying signals in hydraulic systems, it is extremely difficult to achieve fault diagnosis for hydraulic systems. Different components of the system can fail individually or simultaneously. Signal processing faces the problem of coupling between multi-component faults, which makes it more difficult to realise multi-component fault diagnosis. On the one hand, existing techniques rely on hand-designed features and only use a traditional single shallow machine model as the base classifier, and these do not have the ability to self-learn meaningful features. On the other hand, the diagnostic performance of a single base classifier sometimes does not meet engineering requirements. To handle the above problems, a bagging strategy based heterogeneous ensemble deep neural networks (DNNs) approach is proposed for the multiple components fault diagnosis of hydraulic systems. First, Pearson correlation coefficient and neighbourhood component analysis are developed for data channel selection and feature dimensionality reduction. Second, two distinct DNNs are constructed as base learners: a stacked sparse autoencoder and a deep hierarchical extreme-learning machine. Finally, a bagging strategy is adopted to integrate different DNNs to obtain robust diagnostic results. The results from this experiment demonstrate that the proposed method can precisely diagnose hydraulic system faults compared with comparative methods.

An original vibrational energy harvesting device with regard to the vibrational performance of far-end cables was designed under the condition that no power can be supplied to the far end of freight cableways in transmission and distribution projects. The device was designed to have a Y-shaped radial gear-rack structure. The random vibrations of the cable are converted into three one-way rotary motions of a Y-shaped device terminal using a Y-shaped rack unit and one-way gear-rack units. The device then harvests the electric energy through a super-capacitor and supplies power to equipment in areas out of the reach of the power supply along the cableway. An analysis and dynamic simulation of the Y-shaped rack unit and one-way gear-rack units demonstrate that the Y-shaped layout of the device adapts well to the random vibration of the cable. The device was proven to have successfully harvested the vibrational energy of the cable in several directions on its radial plane, showing good prospects in supplying power to far-end wireless surveillance equipment for freight cableways in remote mountainous areas.

Hydrogen station siting plays an important role in hydrogen-energy infrastructure construction, and it's different from gas station siting. A gas station has a unitary way of fuel transport and a unitary fuel supplier, hence no consideration given to factors like fuel supplier and way of fuel transport at the time of siting it. However, hydrogen for a hydrogen fueling station can be supplied jointly from a couple of different sources nearby. Since there is a diversity of hydrogen price and productivity between different sources, hydrogen fueling station siting also entails consideration of the effect of the proportions of hydrogen supplied by the sources on hydrogen's life cycle cost. With the purpose of minimizing hydrogen's life cycle cost, this paper creates a mathematical model for station siting, largely for the case that each station can get hydrogen supply from combined multiple sources, and considers the effect of geographical information factors on station siting. The effect of geographical information factors on such siting is described herein in two cases to avoid selecting a must-not-build location and rebuilding into a gasoline-hydrogen fueling station at an existing gas station location. The latter can reduce station construction and operating costs. By creating a particle swarm optimization (PSO) example for station siting with Shanghai-Nanjing Expressway and constructing a position particle swarm in the form of 5D vector in order to optimize 5 station locations at the same time as well a weight particle swarm in the form of 2D matrix in order to optimize the multi-source hydrogen supply programs, the paper works out optimal station construction locations on condition of multi-source hydrogen supply, multi-source hydrogen supply programs, ways of storage and transport and corresponding hydrogen's optimal life cycle cost.

Collection, transfer and storage of solid waste cannot do without hoisting work. Hoisting operators are usually exposed to bad working circumstances. Studying the precise on-ground locating of solid waste carriers and auto offset correcting gripper design with no need of precise locating are the key to accomplishing unattended operation. This paper proposes a combination of slidable flat washer and four-jaw auto-centering chuck to meet the clamping requirement and provide millimeter-grade carrier locating precision; realization of auto offset correction by use of the clamping jaws of a quasi-planetary gear structure is a guarantee of hoisting work on the carrier in the condition of gripper swinging. The feasibility of the fruit of the paper has been verified by hoisting work in the disposal of nuclear waste and hazardous waste.

Composite pressure vessel has been applied by NASA for the first time. Safe low-cost efficient composite hydrogen-storing device is a key technique affecting the generalization and application of hydrogen energy. By combining the shear theory and based on the highly heave-stable shear behavior shell structure lightweight design, this paper describes researches on the lightweight design of pressure vessel lining and, on the basis of completing the numerical analysis of hydrogen storage tank, extracts its stress isogram, achieving the conformity or tangency of fiber weaving type with stress load isoline and accomplishing lightweight design of the tank and reduced production cost. The method herein can be further generalized to the storage of hydrogen energy and pipe component design.

In the teaching design of electromechanics integration for mechanism, three designs (machinery, motor and control) work in parallel. Before the mechanical team completes the design, the motor and control team needs to obtain information about the machinery in advance. The integrated teaching of the three elements of machinery, motor and control of the planar linkage mechanism is discussed in this article, creating an enhanced environment for electromechanics integration in teaching mechanism combining the virtual simulation and physical production, integrating the five aspects of design, simulation, control, manufacturing and testing, therefore establishing an innovation model of mechanical principles of research and exploratory type teaching, problem-oriented learning-type teaching.

The relationship between hyperuricemia (HUA) and prognosis of acute ischemic stroke (AIS) is unclear. This prospective cohort study aims to evaluate potential value of HUA as a prognostic factor for AIS independent of diabetic status.A total of 1041 consecutive patients aged from 25 to 96 with AIS were included. 340 (32.7%) had diabetes and 246 (23.6%) had HUA. Diabetic patients were stratified by gender or age. Multivariate logistic regression was used to analyse the association between HUA and prognosis of AIS.HUA independently predicted poor discharge outcome of AIS in diabetic patients [OR (95% CI): 2.061 (1.042-4.077), p 0.05]. Among diabetics, HUA selectively predicted a poor functional outcome of AIS at discharge in patients aged ≤75 years [OR (95% CI): 2.381 (1.115-5.085), p 0.05]. Furthermore, in patients aged ≤75 years, HUA independently predicted poor discharge outcome of AIS in male diabetics [OR (95% CI): 2.684 (1.001-7.200), p 0.05]. No association between HUA and prognosis of AIS was observed at 3-month, 6-month and 12-month follow-up, either in diabetics or nondiabetics.HUA independently predicted poor in-hospital outcome of AIS in diabetic patients, especially in patients aged ≤ 75 years.

We report a case of a 105-year-old woman who was treated with Alteplase for acute ischaemic stroke (AIS). On arrival, this centenarian woman had sudden-onset aphasia and right-sided hemiparesis, her baseline National Institutes of Health Stroke Scale (NIHSS) score was 21. An acute noncontrast brain CT scan showed only mild white matter disease. After assessing the patient's risk-benefit ratio, the patient was administered intravenous Alteplase (0.9 mg/kg) 2.5 hours after symptom onset. Her right hemiparesis and aphasia eased within 24 hours after Alteplase treatment. At discharge, her NIHSS score had decreased to 1, leaving only mild right facial palsy unresolved. Few clinical data are available on centenarian patients with AIS who undergo intravenous thrombolysis. There are no data on what is the best method to use to manage centenarians. Our successful thrombolysis case indicates that biologically young centenarians may profit from thrombolysis, despite the very high age.

OBJECTIVE To determine the sensitivity and specificity of the ucN13-P15 (CV2-Fz) and IcN13-P15 (CV7-Fz) interpeak latencies when used to predict acute posterior circulation ischemia. METHODS A total of 426 consecutive patients who were hospitalized within 3 days of the onset of stroke were prospectively enrolled in the study. Of these patients, 110 had infarct lesions in the posterior circulation territory. The ucN13-P15 and IcN13-P15 interpeak latencies on admission were assessed, and the chronic outcome was assessed at 3 months. RESULTS The ucN13-P15 and IcN13-P15 interpeak latencies (both P < 0.001) were prolonged in patients with posterior circulation ischemia than in those with anterior circulation ischemia. The optimal cutoff score of the baseline ucN13-P15 IPL (≥ 3.35 milliseconds) had a sensitivity of 76.8% and a specificity of 79.9%. The IcN13-P15 IPL (≥ 4.225 milliseconds) had a sensitivity of 56.1% and a specificity of 93.2%. CONCLUSIONS The ucN13-P15 interpeak latency must be appraised in light of its diagnostic value in acute posterior circulation ischemia. However, the ucN13-P15 interpeak latency was not independently predictive of a favorable outcome.

Parameterization of discrete surface is a fundamental and widely-used operation in graphics, required, for instance, for texture mapping or remeshing. As 3D data becomes more and more detailed, there is an increased need for fast and robust techniques to automatically compute least-distorted parameterizations of large meshes. In this paper, for a surface patch on a smooth, two-dimensional surface in R3 low-distortion Parameterizations are achieved by an optimal parameterization algorithm. Our proposed algorithm is described by initial parameterization, domain discretization and numerical optimization. The parameterization can be optimized with respect to the varying importance of conformality, length preservation and area preservation. The initial parameterization is provided for the subsequent optimization of the parameter domain. The domain discretization is introduced and a constrained Newton method is used to minimize a corresponding discrete energy. The numerical optimization method is provided for the optimal deformation to generate the final, optimal parameterization. Experiment results prove that our proposed algorithm is effective and robust. (6 pages)

The existence of artifacts in polygonal meshes makes it difficult for model operations, especially when comes to model rebuilding, rapid prototyping or other downstream applications. Model repair is the task of removing artifacts from a geometric model to produce an output model that is suitable for further processing by downstream applications that have certain quality requirements on their input. Existing model repair algorithms are capable of solving single artifact in a model. For different types of artifacts in a model, they may not result in satisfactory results. This paper proposes a surface oriented algorithm for model repair. This algorithm includes five contents: consistent normal orientation, hole filling on surface, manifolds mapping on surface, gap bridging between surface patches, mesh topology simplification. The proposed algorithm works well for different types of artifacts and can better preserve the detailed features of the original mesh. (5 pages)

The optimization of mechanisms is generally done in the field of mechanism itself. In this paper, a multidisciplinary design optimization (MDO) concept is applied to synthesize optimum mechanism. Two disciplines of the mechanism and the control are involved in the design optimization of mechanism. The sensitivity analysis technique of constraints is firstly proposed to coordinate complex coupled relation between disciplines. A MDO algorithm, based on genetic algorithm (GA), is proposed to get the optimum mechanism. A six-bar mechanism design problem is given to verify the coordination method and the MDO algorithm. In this study, the objective function of the mechanism is to minimize the difference of the coupler curve and a desired path. When an electric motor is connected to drive the mechanism, the angular speed of the input crank isn't constant. So, the objective function of the control is to minimize the crank speed fluctuation by using PID feedback control. The optimized six-bar mechanism not only satisfies mechanism and control constraints but also has synthesized optimum function value.

Nowadays, the grinding disc design of big-screen glass screen follows the grinding disc design concept and equation of small-screen glass screen, so the entire glass screen surface was unable to be grinded uniformly and smoothly. Though the drive of the lapping mechanism is constant, it is difficult to describe its motion and set up a mathematical model for the constant dynamic contact between the grinding disc and the glass screen. The grinding disc movement locus and parameter of any time step can be obtained through kinematics simulation. Combing the removal rate, mapping the parameter of operation condition for discrete plane motion into a 3D space, substitute several quasi-static models for the lapping process and solve the grinding disc surface design by scattering the continuous lapping process. A simulation model is established though the understanding of the actual lapping process. After the simulation, revise the numerical simulation model by comparison between the simulation result and the actual result, and then simulate again, thus the simulation for the glass screen lapping process is realized.