As the raising of traffic accidents caused by commercial vehicle drivers, more regulations have been issued for improving their safety status. Driving record instruments are required to be installed on such vehicles in China. The obtained naturalistic driving data offer insight into the causal factors of hazardous events with the requirements to identify where hazardous events happen within large volumes of data. In this study, we develop a model based on a low-definition driving record instrument and the vehicle kinematic data for postaccident analysis by multi-modal deep learning method. With a higher camera position on commercial vehicles than cars that can observe further distance, motion profiles are extracted from driving video to capture the trajectory features of front vehicles at different depths. Then random forest is used to select significant kinematic variables which can reflect the potential crash. Finally, a multi-modal deep convolutional neural network (DCNN) combined both video and kinematic data is developed to identify potential collision risk in each 12-second vehicle trip. The analysis results indicate that the proposed multi-modal deep learning model can identify hazardous events within a large volumes of data at an AUC of 0.81, which outperforms the stateof-the-art random forest model and kinematic threshold method.

Neutron Imaging System (NIS) has been used to image the burn volume and cold fuel volume of imploding fusion capsules. In this work, we present a design of neutron imaging aperture for inertial confinement fusion in Laser Fusion Research Center. Since the total neutron yield should be less than 1014, the penumbral aperture has been chosen. A geometric model has been developed to assess the performance of the neutron imaging system, including the spatial resolution, the field of view and the signal-to-noise ratio. This model reproduces the performances of neutron image systems on OMEGA. The spatial resolution of designed NIS is about 22 μm for a field of view of 250 μm. The signal-to-noise ratio can be better than 10, if the neutron yield is higher than 1013.

A high quality hot spot is crucial in the laser driven inertial confinement fusion. The hot spot self-emitted X-ray images in a high spatial resolution may be used to analyze the hot spot asymmetry and some fine structures induced by mix. The high spatial resolved X-ray imaging diagnostics can also serve in the hydrodynamic instability growth radiography and some other physical research in the inertial confinement fusion. The Kirkpatrick-Baez microscope can provide a higher resolution and throughput efficiency diagnostic. A new four-channels KB microscope was designed and built for the < 10 keV X-ray imaging. The Pt coated reflective mirror pairs were used to obtain a wide grazing angle bandwidth. The variation of the X-ray reflectivity was small in a large field of view. The microscope had a magnification of about 20. The spatial resolution in the central field of view was about 7 µm. The similarities between the different channel images were about 97%. The KB microscope is in operation in the directly or indirectly driven implosions by 10-100 kJ lasers on Shenguang laser facility in China. The time-integral hot spot asymmetry has been diagnosed, and the time-resolved imaging will be implemented in the following work. K: Plasma diagnostics -interferometry, spectroscopy and imaging; Plasma diagnosticshigh speed photography 1Corresponding author.

The role of CD8 + T cells in asthma has not been fully discussed. The mechanisms of CD4 + and CD8 + cells in severe asthma (SA) development were compared. The microarray data (GSE31773) was down loaded from the Gene Expression Omnibus (GEO) database, including 20 samples of CD4 + and CD8 + T cells, which were collected from 8 health controls (HC), 4 non severe asthma (NSA) and 8 SA patients. DEGs of CD4 + and CD8+ T cells in the HC vs. NSA and HC vs. SA groups were identified using the limma package in R. GO and pathway enrichment analysis of the common DEGs between the two groups were ana lyzed using DAVID. The interactive network of DEGs and significant modules were further explored. In CD4 + cells, there were 168 DEGs in HC vs. NSA group and 685 DEGs in HC vs. SA group, while for CD8 + T cells there were 719 DEGs in the HC vs. NSA groups and 1255 DEGs in the HC vs. SA groups. Besides, 80 common DEGs from CD4 + samples were enriched in the MAPKKK cascade and molecular metabolism, and 385 common DEGs of CD8 + T cells were significantly related with cell apoptosis and transformation. Moreover, two significant modules of DEGs in CD4 + were found to be involved with MPO and BPI. One module of CD8 + T cells containing PDHA1 and MRPL42 was identified to be related with glycolysis. In con clusion, MPO and BPI in CD4 + , and PDHA1 and MRPL42 in CD8 + T cells might be used as specific biom arkers of SA progression. Therapy targeting the functions of CD4 + and CD8 + T cells may provide a novel per spective for SA treatment.