An EEG Channel Selection Framework for Driver Drowsiness Detection via Interpretability Guidance.

Drowsy driving has a crucial influence on driving safety, creating an urgent demand for driver drowsiness detection. Electroencephalogram (EEG) signal can accurately reflect the mental fatigue state and thus has been widely studied in drowsiness monitoring. However, the raw EEG data is inherently noisy and redundant, which is neglected by existing works that just use single-channel EEG data or full-head channel EEG data for model training, resulting in limited performance of driver drowsiness detection. In this paper, we are the first to propose an Interpretability-guided Channel Selection (ICS) framework for the driver drowsiness detection task. Specifically, we design a two-stage training strategy to progressively select the key contributing channels with the guidance of interpretability. We first train a teacher network in the first stage using full-head channel EEG data. Then we apply the class activation mapping (CAM) to the trained teacher model to highlight the high-contributing EEG channels and further propose a channel voting scheme to select the top N contributing EEG channels. Finally, we train a student network with the selected channels of EEG data in the second stage for driver drowsiness detection. Experiments are designed on a public dataset, and the results demonstrate that our method is highly applicable and can significantly improve the performance of cross-subject driver drowsiness detection.

Encapsulation of atomically dispersed Pt clusters in porous TiO2 for semi-hydrogenation of phenylacetylene.

The synthesis of atomically dispersed metal clusters with strong interaction with the support is attractive for the design of high-efficiency catalysts. Here, we report a multilayered catalyst (1.91%Pt@TiO2), in which atomically dispersed Pt clusters are encapsulated in the porous TiO2. As a result, 1.91%Pt@TiO2 exhibited high activity, selectivity (92.9%), and excellent stability in the semi-hydrogenation of phenylacetylene.

Neurotoxicity of perfluorooctanoic acid and post-exposure recovery due to blueberry anthocyanins in the planarians Dugesia japonica.

Perfluorooctanoic acid (PFOA) is a widely used synthetic industrial chemical which accumulates in ecosystems and organisms. Our study have investigated the neurobehavioral effects of PFOA and the alleviation effects of PFOA-induced neurotoxicity by blueberry anthocyanins (ANT) in Dugesia japonica. The planarians were exposed to PFOA and ANT for ten days. Researchs showed that exposure to PFOA affected locomotor behavior and ANT significantly alleviated the reduction in locomotion induced by PFOA. The regeneration of eyespots and auricles was suppressed by PFOA and was promoted by ANT. Following exposure to PFOA, acetylcholinesterase activity continually decreased and was unaffected in the ANT group, but was elevated after combined administration of PFOA and ANT. Oxidative DNA damage was found in planarians exposed to PFOA and was attenuated after administration of ANT by the alkaline comet assay. Concentrations of three neurotransmitters increased following exposure to PFOA and decreased after administration of ANT. Furthermore, ANT promoted and PFOA inhibited neuronal regeneration. DjotxA, DjotxB, DjFoxG, DjFoxD and Djnlg associated with neural processes were up-regulated following exposure to PFOA. Our findings indicate that PFOA is a neurotoxicant while ANT can attenuate these detrimental effects.

Distance Effect of Ni-Pt Dual Sites for Active Hydrogen Transfer in Tandem Reaction.

Unveiling the distance effect between different sites in multifunctional catalysts remains a major challenge. Herein, we investigate the distance effect by constructing a dual-site distance-controlled tandem catalyst with a five-layered TiO2/Pt/TiO2/Ni/TiO2 tubular nanostructure by template-assisted atomic layer deposition. In this catalyst, the Ni and Pt sites are separated by a porous TiO2 interlayer, and the distance between them can be precisely controlled on the subnanometer scale by altering the thickness of the interlayer, while the inner and outer porous TiO2 layers are designed for structural stability. The catalyst exhibits superior performance for the tandem hydrazine hydrate decomposition to hydrogen and subsequent nitrobenzene hydrogenation when the Ni and Pt site distance is on the subnanometer level. The performance increases with the decrease of the distance and is better than the catalyst without the TiO2 interlayer. Isotopic and kinetic experiments reveal that the distance effect controls the transfer of active hydrogen, which is the rate-determining step of the tandem reaction in a water solvent. Reduced Ti species with oxygen vacancies on the TiO2 interlayer provide the active sites for hydrogen transfer with -Ti-OH surface intermediates via the continuous chemisorption/desorption of water. A smaller distance induces the generation of more active sites for hydrogen transfer and thus higher efficiency in the synergy of Ni and Pt sites. Our work provides new insight for the distance effect of different active sites and the mechanism of intermediate transfer in tandem reactions.

The impact of chlorine disinfection on biochemical oxygen demand levels in chemically enhanced primary treatment effluent.

The response trends of biochemical oxygen demand (BOD) and organic strength after the chlorination/dechlorination process were explored through a 2-year, 5-month chemically enhanced primary treatment (CEPT) effluent onsite monitoring program and a 2-month laboratory-scale study. The monitoring results showed that better instantaneous mixing at the chlorine injection point reduced the effect of chlorination/dechlorination on the 5-day BOD levels. The laboratory study results demonstrated that chlorination did not change the particle size distribution, dissolved organic carbon, or chemical oxygen demand of the organic content of the effluent. Nevertheless, chlorination/dechlorination strongly affected the BOD measurement when nitrification was inhibited by changing bioactivity/biodegradation rates.

Quantitative analyses of ammonia-oxidizing Archaea and bacteria in the sediments of four nitrogen-rich wetlands in China.

With the rapid development of ammonia-synthesizing industry, the ammonia-nitrogen pollution in wetlands acting as the sink of point and diffuse pollution has been increased dramatically. Most of ammonia-nitrogen is oxidized at least once by ammonia-oxidizing prokaryotes to complete the nitrogen cycle. Current research findings have expanded the known ammonia-oxidizing prokaryotes from the domain Bacteria to Archaea. However, in the complex wetlands environment, it remains unclear whether ammonia oxidation is exclusively or predominantly linked to Archaea or Bacteria as implied by specific high abundance. In this research, the abundance and composition of Archaea and Bacteria in sediments of four kinds of wetlands with different nitrogen concentration were investigated by using quantitative real-time polymerase chain reaction, cloning, and sequencing approaches based on amoA genes. The results indicated that AOA distributed widely in wetland sediments, and the phylogenetic tree revealed that archaeal amoA functional gene sequences from wetlands sediments cluster as two major evolutionary branches: soil/sediment and sediment/water. The bacteria functionally dominated microbial ammonia oxidation in different wetlands sediments on the basis of molecule analysis, potential nitrification rate, and soil chemistry. Moreover, the factors influencing AOA and AOB abundances with environmental indicator were also analyzed, and the results addressed the copy numbers of archaeal and bacterial amoA functional gene having the higher correlation with pH and ammonia concentration. The pH had relatively great negative impact on the abundance of AOA and AOB, while ammonia concentration showed positive impact on AOB abundance only. These findings could be fundamental to improve understanding of the importance of AOB and AOA in nitrogen and other nutrients cycle in wetland ecosystems.