Designing ergonomic interventions for EMS workers - part II: lateral transfers

The objective of the current work was to test ergonomic interventions aimed at reducing the low back musculoskeletal loads experienced by firefighters/paramedics (FFPs) providing emergency medical services (EMS) when performing lateral transfers between a bed and a stretcher or between a stretcher and a hospital gurney. The interventions, developed using focus groups, were a bridgeboard to reduce the frictional force resisting the lateral sliding of the patient, the use of rods along each side of the patient to facilitate the grasping and handling of the bedsheet on which the patient is typically transferred, and a single rod that, when rolled in the bedsheet, resulted in the task being changed from a lifting task to a pulling task. Eleven two-person teams laterally transferred a 75 kg dummy with each intervention between a bed and simulated stretcher. Two roles were defined. For the two-sided transfers, the FFP roles were termed "stretcher-side" and "bed-side." Surface electromyographic (EMG) data were collected from 8 trunk muscles from each participant along with spine kinematic data. Additionally, kinetic data were obtained for the FFP in the stretcher-side role. Trunk flexion moments and Erector Spinae activity were reduced for the FFP in the stretcher-side role when using the bridgeboard and the single rod both individually and in combination. The single rod reduced the Erector Spinae activity in the FFP who typically would have been on the bed. For FFPs in both roles the single rod increased Latissimus Dorsi activation relative to the standard bedsheet transfer condition, although, this effect was moderated when the single rod was used in combination with the bridgeboard. Ratings of perceived exertion also supported the use of the single rod relative to the corresponding control condition.     

Designing ergonomic interventions for EMS workers, Part I: transporting patients down the stairs

The objective of the current work was to test ergonomic interventions aimed at reducing the magnitude of trunk muscle exertions in firefighters/paramedics (FFPs) providing emergency medical services (EMS) when transporting patients down the stairs. The interventions, developed using focus groups, were a footstrap to prevent the patient from sliding down on the backboard, a change in the handle configuration on the stairchair, and 2 devices, the "backboard wheeler" and a tank tread-like device (descent control system, DCS) for a stretcher, that change the backboard and stretcher carrying tasks into rolling and sliding tasks. Eleven two-person teams transported a 75 kg dummy with each intervention and its corresponding control condition down a flight of steps. Surface electromyographic (EMG) data were collected from 8 trunk muscles from each participant. Results showed that the backboard footstrap reduced the erector spinae (ERS) activity for the FFP in the "leader" role by 15 percent, on average. The change in handle configuration on the stairchair had no effect on the variables measured. The backboard wheeler reduced the ERS activity bilaterally in the FFP in the leader role and unilaterally for the FFP in the "follower" role, by 28 and 24 percent, respectively. The DCS reduced the 90th percentile ERS activity for both FFPs from 26 to 16 percent MVC, but increased the latissimus dorsi activity in the follower from 11 to 15 percent MVC. The DCS was the only intervention tested that resulted in a reduced rating of perceived exertion relative to the corresponding control condition. In summary, the hypotheses that the proposed interventions could reduce trunk muscle loading were supported for 3 of the 4 transport interventions tested.     

基于回跳层数的SAT求解器学习子句删除策略

目前学习子句删除策略广泛采用的是基于LBD的评估方式,LBD评估方式在每次执行删除时都会删除前一半LBD值大的学习子句,这种方式对LBD值大的学习子句的删除过于激进。针对此问题,提出了一种利用冲突回跳层数（back-jump levels）的评估方式来保留LBD值较大的相关学习子句。以CDCL（conflict driven clause learning）完备算法为框架,在子句删除环节形成了BJL删除算法。通过测试2017年SAT国际竞赛例,对新改进的版本与原版求解器进行了对比实验。实验表明,所提策略可显著提高求解器的求解性能和求解效率。     

基于高精度室内位置感知的大数据研究与应用

随着室内定位技术的发展,室内位置数据和用户消费行为数据的大量产生为室内位置大数据(LBD)研究和应用提供了可能。基于高精度室内位置感知,突破了室内定位位置数据不准确的瓶颈。通过对室内位置数据聚类、降维等预处理,建立挖掘模型分析并提取了室内商圈区域的聚散和流动等特性,进一步通过特征关联预测用户的消费等行为,提出了室内位置大数据协同挖掘的方法和架构。在某机场商圈、西单某商场亿级用户位置数据集上进行了有效性实验和应用,通过实测数据对比验证了基于此架构室内定位数据的精准性和挖掘方法的可行性。     

LBD29基因在拟南芥侧根发生过程中的作用

以拟南芥LBD29基因功能缺失突变体lbd29为材料,采用显微及亚显微技术,研究了拟南芥LBD29基因对于中柱鞘细胞分化和侧根发生的影响。结果显示：（1）与野生型相比,lbd29侧根发生晚,侧根数量少;（2）拟南芥根成熟区中柱鞘细胞是由两类分化程度不同的细胞组成,侧根原基形成前,位于木质部极中柱鞘的侧根原基建成细胞（founder cells）首先经历了脱分化,进而分裂形成侧根原基;（3）突变体中柱鞘细胞生长素响应不敏感,具有一定程度的脱分化障碍。以上结果初步表明,LBD29基因可能通过生长素响应通路调控中柱鞘founder cells的脱分化及分裂,从而影响侧根发生。     

基于频次的SAT问题学习子句混合评估算法

为了有效管理学习子句,避免学习子句规模呈几何级增长,减少冗余学习子句对系统内存占用,从而提高布尔可满足性问题SAT求解器的求解效率,需要对学习子句进行评估,然后删减学习子句。传统的评估方式是基于学习子句的长度,保留较短的子句。当前主流的做法一个是变量衰减和VSIDS的子句评估方式,另外一个是基于文字块距离LBD的评估方式,也有将二者结合使用作为子句评估的依据。通过对学习子句参与冲突分析次数与问题求解的关系进行分析,将学习子句使用频率与LBD评估算法混合使用,既反映了学习子句在冲突分析中的作用,也充分利用了文字与决策层之间的信息。以Syrup求解器（GLUCOSE 4.1并行版本）为基准,在评估算法与并行子句共享策略方面做改进测试,通过实验对比发现,混合评估算法比LBD评估算法有优势,求解问题个数明显增多。