Depressive symptoms are a risk factor for mortality in elderly people, particularly in physically healthy individuals.

Electromagnetic source imaging (ESI) is a highly ill-posed inverse problem. To find a unique solution, traditional ESI methods impose a variety of priors which may not reflect the actual source properties. Such limitations of traditional ESI methods hinder their further applications. Inspired by deep learning approaches, a novel data-synthesized spatio-temporal denoising autoencoder method (DST-DAE) method was proposed to solve the ESI inverse problem. Unlike the traditional methods, we utilize a neural network to directly seek generalized mapping from the measured E/MEG signals to the cortical sources. A novel data synthesis strategy is employed by introducing the prior information of sources to the generated large-scale samples using the forward model of ESI. All the generated data are used to drive the neural network to automatically learn inverse mapping. To achieve better estimation performance, a denoising autoencoder (DAE) architecture with spatio-temporal feature extraction blocks is designed. Compared with the traditional methods, we show (1) that the novel deep learning approach provides an effective and easy-to-apply way to solve the ESI problem, that (2) compared to traditional methods, DST-DAE with the data synthesis strategy can better consider the characteristics of real sources than the mathematical formulation of prior assumptions, and that (3) the specifically designed architecture of DAE can not only provide a better estimation of source signals but also be robust to noise pollution. Extensive numerical experiments show that the proposed method is superior to the traditional knowledge-driven ESI methods.

Amorphous carbon (a-C) and carbon nitride (a-CNx) thin films prepared by sputter deposition on thin Corning glass substrates were studied using isothermal thermogravimetry and the differential method. Two typical reaction kinetics, the sigmoid and deceleratory reactions, were observed, which correspond to the films sputtered at higher (0.9–3 mTorr) and lower vacuums (7–16 mTorr), respectively. The structure bonding ratios and activation energies for thermal decompositions, which increased with the kinetic energy of the carbon species during deposition, were used to interpret the observations of reactions. The free bonded and weakly bonded compositions, the amounts of which increased with sputtering pressure, could be active nucleus sites in early reactions determining the reaction kinetics. The films with higher structure bonding ratios showed an induction period for generation of infrequently formed nuclei and had typically sigmoid-type reaction kinetics, while those with large amounts of free and weakly bonded compositions produced large amounts of quickly formed nuclei in early reactions and showed deceleratory reaction kinetics.

In this paper, numerical simulation of a new type enhanced heat transfer burner was carried out by using the CFD commercial software FLUENT. Standard κ-ε turbulent model, P-1 radiation model and PDF diffusion combustion model were used to predict the influence of the combustor’s structure change on its performance in the combustion process. The results showed that: The added necking down at the outlet of the combustion chamber can significantly enhanced the jet action of the flue gas, the high speed flow flue gas formed a forced convection cyclical field, convective heat transfer rate was increased greatly and the temperature distribution in the furnace became more uniform, which guaranteed an excellent heating effect. New type of staged air distribution can promote the mixing of the fuel and air. Further more, it can improve the flame length to prevent the local overheating phenomenon during the combustion process. On the basis of the same total sectional areas, the added number of the jet orifice can also promote the mixing of the fuel and air to enhance the thermal intensity and thermal efficiency of the furnace.

No abstract