s-TBN: A New Neural Decoding Model to Identify Stimulus Categories From Brain Activity Patterns.

Neural decoding is still a challenging and a hot topic in neurocomputing science. Recently, many studies have shown that brain network patterns containing rich spatiotemporal structural information represent the brain's activation information under external stimuli. In the traditional method, brain network features are directly obtained using the standard machine learning method and provide to a classifier, subsequently decoding external stimuli. However, this method cannot effectively extract the multidimensional structural information hidden in the brain network. Furthermore, studies on tensors have show that the tensor decomposition model can fully mine unique spatiotemporal structural characteristics of a spatiotemporal structure in data with a multidimensional structure. This research proposed a stimulus-constrained Tensor Brain Network (s-TBN) model that involves the tensor decomposition and stimulus category-constraint information. The model was verified on real neuroimaging data obtained via magnetoencephalograph and functional mangetic resonance imaging). Experimental results show that the s-TBN model achieve accuracy matrices of greater than 11.06% and 18.46% on the accuracy matrix compared with other methods on two modal datasets. These results prove the superiority of extracting discriminative characteristics using the STN model, especially for decoding object stimuli with semantic information.

The Structure Evolution of Hydroxyl-Reacted LiBC to Improve the Capacity for Li-Ion Batteries.

As a graphite-like material, the LiBC can deliver a high capacity up to 500 mA h g-1 in Li-ion batteries, which is dependent on the carbon precursor, the high-temperature treatment, and the lithium insufficiency. However, the underlying mechanism is still not clear for the electrochemical reactions of LiBC. In this work, the pristine LiBC was reacted with aqueous solutions of different alkalinity, which was delithiated chemically and retained the layered structure. According to the XPS and NMR results, the B-B bond might be produced through the aqueous reaction or the initial charge process, which can be oxidized (charged) and reduced (discharged) in the electrochemical measurements. In the Li-ion battery, the reversible capacity of LiBC increases evidently with the alkalinity of aqueous solution and significantly rises to a similar value of ca. 285 mA h g-1 under 200 cycles. Therefore, the specific capacity of LiBC should be contributed by the active sites of B-B bonds, which can be significantly increased through the reaction with the hydroxyl ions, and this strategy might be adopted to activate more graphite-like materials.

Controlled Synthesis of Single-Crystalline Ni-Rich Cathodes for High-Performance Lithium-Ion Batteries.

Single-crystalline LiNi0.8Co0.1Mn0.1O2 (NCM811) has been considered as one of the most promising cathode materials. It addresses the pulverization issue present in its polycrystalline counterpart by eliminating intergranular cracks. However, synthesis of high-performance single-crystalline NCM is still a challenge owing to the lower structure stability of NCM811 at high calcination temperatures (≥900 °C), which is often required to grow single crystals. Herein, we report a synthesis process for microsized single-crystalline NCM811 particles with exposed (010) facets on their lateral sides [named as SC-NCM(010)], which includes the preparation of a well-dispersed microblock-like Ni0.8Co0.1Mn0.1(OH)2 precursor through coprecipitation assisted with addition of PVP and Na2SiO3 and subsequent lithiation of the precursor at 800 °C. The SC-NCM(010) cathode exhibits an excellent capacity retention rate (91.6% after 200 cycles at 1 C, 4.3 V) and a high rate capability (152.2 mAh/g at 20 C, 4.4 V), much superior to those of polycrystalline NCM811 cathodes. However, despite the removal of interparticle boundaries, when cycled between 2.8 and 4.5 V, the SC-NCM(010) cathode still suffers from structural changes including lattice gliding and intragranular cracking. These structural changes are correlated with the interior structural inhomogeneity, which is evidenced by the coexistence of H2 and H3 phases in the fully deintercalated state.

Relieving the Reaction Heterogeneity at the Subparticle Scale in Ni-Rich Cathode Materials with Boosted Cyclability.

Ni-rich layered LiNi0.8Co0.1Mn0.1O2 (NCM811), as a highly suitable candidate for commercialized cathode materials, inevitably suffers from reaction inhomogeneity during electrochemical processes owing to the polycrystalline aggregate particle morphology, especially at high voltages. With the cycles proceeding, intergranular microcracks induced by an anisotropic volume change emerge and accumulate, leading to contact loss of the internal grains. Subsequently, a decrease in accelerated diffusion kinetics and internal Li+ deactivation take place, which further deteriorate the reaction heterogeneity between the surface and bulk phases within polycrystalline subparticles, ultimately leading to rapid capacity failure. To deal with these issues, a microstructural tailored NCM811 with a suitable subparticle size and ordered primary grain arrangement is employed as an alternative cathode. Owing to the optimized microstructure, reaction homogeneity has been significantly promoted, which causes enhanced electrochemical properties with long-term cycling. It is revealed that the mechanically strengthened microstructure contributes to maintaining contact between the surface and bulk phases, resulting in a reversible H2-H3 phase transition and superior Li+ kinetics upon cycling. This microstructural engineering route based on the rational electrode architecture can boost reaction homogeneity and provide guidance for the design of advanced cathode materials.

A Collaboratively Polar Conductive Interface for Accelerating Polysulfide Redox Conversion.

In order to alleviate the inferior cycle stability of the sulfur cathode, a self-assembled SnO2-doped manganese silicate nanobubble (SMN) is designed as a sulfur/polysulfide host to immobilize the intermediate Li2S x, and nitrogen-doped carbon (N-C) is coated on SMN (SMN@C). The exquisite N-C conductive network not only provides sufficient free space for the volume expansion during the phase transition of solid sulfur into lithium sulfide but also reduces Rct of SMN. During cycling, the soluble polysulfide could be fastened by the silicate with an oxygen-rich functional group and heteronitrogen atoms through chemical bonding, enabling a confined shuttle effect. The synergistic effect between N-C and SMN could also effectively facilitate the interconversion between lithium polysulfides and Li2S, reducing the potential barrier and accelerating the redox kinetics. With an areal sulfur loading of 2 mg/cm2, the S-SMN@C cathodes demonstrate a high initial capacity of 1204 mA·h/g at 0.1 C, and an outstanding cycle stability with a capacity fading rate of 0.0277%, ranging from the 2nd cycle to the 1000th cycle at 2 C.

Synthesis of TiC Nanoparticles Anchored on Hollow Carbon Nanospheres for Enhanced Polysulfide Adsorption in Li-S Batteries.

A novel spatial confinement strategy based on a carbon/TiO2 /carbon sandwich structure is proposed to synthesize TiC nanoparticles anchored on hollow carbon nanospheres (TiC@C) through a carbothermal reduction reaction. During the synthesis process, two carbon layers not only serve as reductant to convert TiO2 into TiC nanoparticles, but also create a spatial confinement to suppress the aggregation of TiO2 , resulting in the formation of well-dispersed TiC nanoparticles. This unique TiC@C structure shows an outstanding long-term cycling stability at high rates owing to the strong physical and chemical adsorption of lithium polysulfides (i.e., a high capacity of 732.6 mA h g-1 at 1600 mA g-1 ) and it retains a capacity of 443.2 mA h g-1 after 1000 cycles, corresponding to a decay rate of only 0.0395 % per cycle. Therefore, this unique TiC@C composite could be considered as an important candidate for the cathode material in Li-S batteries.

Synthesis of Double-Shell SnO2@C Hollow Nanospheres as Sulfur/Sulfide Cages for Lithium-Sulfur Batteries.

Double-shell SnO2@C hollow nanospheres were synthesized by a template method, and then the sulfur was loaded to form a cathode material of S/SnO2@C composite. In Li-S batteries, it delivered a high initial specific capacity of 1473.1 mAh/g at a current density of 200 mA/g, and the capacity retention was even up to 95.7% over 100 cycles at 3200 mA/g, i.e., a capacity fade rate of only 0.043% per cycle. These good electrochemical performances should be attributed to the SnO2@C hollow nanospheres. They can enhance the electronic conductivity by the outside carbon shell, and confine the lithium polysulfides by S-Sn-O and S-C chemical bonds to suppress the shuttle effect. Besides, the hollow nanospheres can readily accommodate the sulfur/sulfides to prevent the electrical/mechanical failure of the cathode, instead of their agglomeration on the external surface of SnO2@C.

Determinants of HIV-induced brain changes in three different periods of the early clinical course: A data mining analysis.

To inform an understanding of brain status in HIV infection, quantitative imaging measurements were derived at structural, microstructural and macromolecular levels in three different periods of early infection and then analyzed simultaneously at each stage using data mining. Support vector machine recursive feature elimination was then used for simultaneous analysis of subject characteristics, clinical and behavioral variables, and immunologic measures in plasma and CSF to rank features associated with the most discriminating brain alterations in each period. The results indicate alterations beginning in initial infection and in all periods studied. The severity of immunosuppression in the initial virus host interaction was the most highly ranked determinant of earliest brain alterations. These results shed light on the initial brain changes induced by a neurotropic virus and their subsequent evolution. The pattern of ongoing alterations occurring during and beyond the period in which virus is suppressed in the systemic circulation supports the brain as a viral reservoir that may preclude eradication in the host. Data mining capabilities that can address high dimensionality and simultaneous analysis of disparate information sources have considerable utility for identifying mechanisms underlying onset of neurological injury and for informing new therapeutic targets.

A review of heterogeneous data mining for brain disorder identification.

With rapid advances in neuroimaging techniques, the research on brain disorder identification has become an emerging area in the data mining community. Brain disorder data poses many unique challenges for data mining research. For example, the raw data generated by neuroimaging experiments is in tensor representations, with typical characteristics of high dimensionality, structural complexity, and nonlinear separability. Furthermore, brain connectivity networks can be constructed from the tensor data, embedding subtle interactions between brain regions. Other clinical measures are usually available reflecting the disease status from different perspectives. It is expected that integrating complementary information in the tensor data and the brain network data, and incorporating other clinical parameters will be potentially transformative for investigating disease mechanisms and for informing therapeutic interventions. Many research efforts have been devoted to this area. They have achieved great success in various applications, such as tensor-based modeling, subgraph pattern mining, and multi-view feature analysis. In this paper, we review some recent data mining methods that are used for analyzing brain disorders.

Identifying HIV-induced subgraph patterns in brain networks with side information.

Investigating brain connectivity networks for neurological disorder identification has attracted great interest in recent years, most of which focus on the graph representation alone. However, in addition to brain networks derived from the neuroimaging data, hundreds of clinical, immunologic, serologic, and cognitive measures may also be documented for each subject. These measures compose multiple side views encoding a tremendous amount of supplemental information for diagnostic purposes, yet are often ignored. In this paper, we study the problem of subgraph selection from brain networks with side information guidance and propose a novel solution to find an optimal set of subgraph patterns for graph classification by exploring a plurality of side views. We derive a feature evaluation criterion, named gSide, to estimate the usefulness of subgraph patterns based upon side views. Then we develop a branch-and-bound algorithm, called gMSV, to efficiently search for optimal subgraph patterns by integrating the subgraph mining process and the procedure of discriminative feature selection. Empirical studies on graph classification tasks for neurological disorders using brain networks demonstrate that subgraph patterns selected by the multi-side-view-guided subgraph selection approach can effectively boost graph classification performances and are relevant to disease diagnosis.

Tensor-based Multi-view Feature Selection with Applications to Brain Diseases.

In the era of big data, we can easily access information from multiple views which may be obtained from different sources or feature subsets. Generally, different views provide complementary information for learning tasks. Thus, multi-view learning can facilitate the learning process and is prevalent in a wide range of application domains. For example, in medical science, measurements from a series of medical examinations are documented for each subject, including clinical, imaging, immunologic, serologic and cognitive measures which are obtained from multiple sources. Specifically, for brain diagnosis, we can have different quantitative analysis which can be seen as different feature subsets of a subject. It is desirable to combine all these features in an effective way for disease diagnosis. However, some measurements from less relevant medical examinations can introduce irrelevant information which can even be exaggerated after view combinations. Feature selection should therefore be incorporated in the process of multi-view learning. In this paper, we explore tensor product to bring different views together in a joint space, and present a dual method of tensor-based multi-view feature selection (dual-Tmfs) based on the idea of support vector machine recursive feature elimination. Experiments conducted on datasets derived from neurological disorder demonstrate the features selected by our proposed method yield better classification performance and are relevant to disease diagnosis.