Search

With the rapid growth of the world's population and urbanization, people are increasingly seeking higher-quality medical services to improve their lives. The classification method based on deep convolutional neural networks (CNNs) is widely used in smart healthcare systems along with advancements in communication and hardware technology. Unfortunately, for conventional deep CNN algorithms, most of the regions do not participate in the convolution operation, resulting in the loss of feature information and the correlation of information between the features. To address this issue, this paper proposes a new strategy of aggregation decentralized down-sampling to prevent the loss of feature information. The regions that are not involved in the convolution operation are re-convoluted and stacked onto depth information in the forward propagation layer and the short-circuit layer, ensuring gradual convergence of the feature map and avoiding the loss of feature information. The accuracy of the proposed residual network (ResNet) system for classification tasks showed an average improvement of 2.57% compared with the conventional ResNet strategies. [ABSTRACT FROM AUTHOR]

The Cl-initiated oxidation of methacrolein (MACR, C4H6O) under NOx-free conditions has been investigated in a fast flow tube by using a home-made vacuum ultraviolet (VUV) photoionization mass spectrometer complemented by high-level theoretical calculations. The key species such as intermediates and radicals together with products involved in the oxidation are observed online and confirmed in photoionization mass spectra. The reaction potential energy surfaces of the transient C4H5O and C4H6OCl radicals, formed from the hydrogen-abstraction reaction and the addition reaction of MACR with Cl atoms, with oxygen have been theoretically calculated to illuminate the formation of the peroxy radicals of C4H5OO2 and C4H6OClO2. The photoionization processes of these peroxy radicals, whose cations are not stable, and their individual self-reactions as well as bimolecular reactions with HO2 radical are studied and discussed. In addition, kinetic experiments are also performed to get the time evolution of specific products and compared with theoretical models, providing a detailed insight into the reaction mechanism of the Cl-initiated oxidation of MACR. [ABSTRACT FROM AUTHOR]

The Cl-initiated oxidation of methacrolein (MACR, C 4 H 6 O) under NO x -free conditions has been investigated in a fast flow tube by using a home-made vacuum ultraviolet (VUV) photoionization mass spectrometer complemented by high-level theoretical calculations. The key species such as intermediates and radicals together with products involved in the oxidation are observed online and confirmed in photoionization mass spectra. The reaction potential energy surfaces of the transient C 4 H 5 O and C 4 H 6 OCl radicals, formed from the hydrogen-abstraction reaction and the addition reaction of MACR with Cl atoms, with oxygen have been theoretically calculated to illuminate the formation of the peroxy radicals of C 4 H 5 OO 2 and C 4 H 6 OClO 2 . The photoionization processes of these peroxy radicals, whose cations are not stable, and their individual self-reactions as well as bimolecular reactions with HO 2 radical are studied and discussed. In addition, kinetic experiments are also performed to get the time evolution of specific products and compared with theoretical models, providing a detailed insight into the reaction mechanism of the Cl-initiated oxidation of MACR.