Semi-Supervised Deep Learning for Multi-Tissue Segmentation from Multi-Contrast MRI

Journal of Signal Processing Systems - Segmentation of thigh tissues (muscle, fat, inter-muscular adipose tissue (IMAT), bone, and bone marrow) from magnetic resonance imaging (MRI) scans is useful...     

考虑电动汽车时空负荷分布特性的主动配电网动态重构

实现电动汽车(Electric Vehicle, EV)规模化发展并与电网双赢的关键问题之一是如何提高EV充电负荷的预测准确性,并保证含大规模EV充电负荷的配电网运行的安全性和经济性。考虑EV时空负荷分布特性,建立了主动配电网动态重构与有功、无功联合优化数学模型,并给出了其求解方法。首先,根据出行链技术和马尔可夫决策理论,考虑天、人、路对EV的影响因素,构建了EV单位能耗模型和充电负荷的时空分布预测模型。其次,提出考虑储能系统、有载分接开关、投切电容器组、静止无功补偿装置和动态重构多种主动管理措施,计及经济、技术指标和各设备、系统运行约束,建立了含EV的主动配电网动态重构与有功-无功联合优化数学模型。然后,为了提高所构建模型的求解效率,通过二阶锥松弛和变量乘积线性化方法将非凸等式约束和非线性不等式约束线性化后,将原始的混合整数非线性规划问题转化为易求解计算的混合整数二阶锥问题。最后,基于修改的IEEE33节点系统进行仿真实验和对比分析,结果验证了所提方法的有效性和优越性。     

基于深度学习特征融合和联合约束的单通道语音分离方法

为了提高单通道语音分离性能,该文提出基于深度学习特征融合和联合约束的单通道语音分离方法。传统基于深度学习的分离算法的损失函数只考虑了预测值和真实值的误差,这使得分离后的语音与纯净语音之间误差较大。该文提出一种新的联合约束损失函数,该损失函数不仅约束了理想比值掩蔽的预测值和真实值的误差,还惩罚了相应幅度谱的误差。另外,为了充分利用多种特征的互补性,提出一种含特征融合层的卷积神经网络(CNN)结构。利用该CNN提取多通道输入特征的深度特征,并在融合层中将深度特征与声学特征融合用来训练分离模型。由于融合构成的特征含有丰富的语音信息,具有强的语音信号表征能力,使得分离模型预测的掩蔽更加准确。实验结果表明,从信号失真比(SDR) 、主观语音质量评估( PESQ)和短时客观可懂度(STOI)3个方面评价,相比其他优秀的基于深度学习的语音分离方法,该方法能够更有效地分离目标语音。     

基于Ad hoc网络的联合有效用户识别与信道安全编译码方法

Ad hoc网络是一种无中心的自组织网络,由于用户能量受限当信噪比较低时有效用户识别的可靠性和信道安全性下降.该文针对这个问题提出一种联合有效用户识别与信道安全编译码方法.将发送用户的正交地址码与接收用户的伪随机地址码异或产生基于发送用户与接收用户的有效用户识别码.为提高信道安全性以信道安全码作为密钥加密有效用户识别码得到正交随机安全序列.为实现扩频并提高传输效率将发送数据以6 bit信息作为一个符号进行分组,将每个符号与一个正交随机安全序列对应.接收用户采用基于子空间的方法处理接收信号,通过建立判决模型识别有效用户.实验结果表明该方法与已有方法相比,当有效用户识别的漏警概率为10–3时信噪比增益改善1.6 dB.     

长寿命路面疲劳寿命预测的频域分析方法

笔者分别从沥青路面和水泥混凝土路面两个方面分析了路面疲劳寿命预测的相关方法。重点对疲劳寿命预测的频域分析方法进行了较为详细的分析,结合车辆–道路耦合振动系统模型,对路面不平整引起的车辆与道路之间的随机振动进行了分析,指出车辆–道路耦合振动的随机功率谱密度是进行路面疲劳寿命预测的一种手段和方法,并提出了相应的预测步骤,旨在为长寿命路面的设计和性能评价提供参考。     

Poly(N‐phenylglycine)‐Based Nanoparticles as Highly Effective and Targeted Near‐Infrared Photothermal Therapy/Photodynamic Therapeutic Agents for Malignant Melanoma

Malignant melanoma is a highly aggressive tumor resistant to chemotherapy. Therefore, the development of new highly effective therapeutic agents for the treatment of malignant melanoma is highly desirable. In this study, a new class of polymeric photothermal agents based on poly(N‐phenylglycine) (PNPG) suitable for use in near‐infrared (NIR) phototherapy of malignant melanoma is designed and developed. PNPG is obtained via polymerization of N‐phenylglycine (NPG). Carboxylate functionality of NPG allows building multifunctional systems using covalent bonding. This approach avoids complicated issues typically associated with preparation of polymeric photothermal agents. Moreover, PNPG skeleton exhibits pH‐responsive NIR absorption and an ability to generate reactive oxygen species, which makes its derivatives attractive photothermal therapy (PTT)/photodynamic therapy (PDT) dual‐modal agents with pH‐responsive features. PNPG is modified using hyaluronic acid (HA) and polyethylene glycol diamine (PEG‐diamine) acting as the coupling agent. The resultant HA‐modified PNPG (PNPG‐PEG‐HA) shows negligible cytotoxicity and effectively targets CD44‐overexpressing cancer cells. Furthermore, the results of in vitro and in vivo experiments reveal that PNPG‐PEG‐HA selectively kills B16 cells and suppresses malignant melanoma tumor growth upon exposure to NIR light (808 nm), indicating that PNPG‐PEG‐HA can serve as a very promising nanoplatform for targeted dual‐modality PTT/PDT of melanoma.     

Isolation and characterization of seven quorum quenching bacteria for biofouling control in MBR

Recently, interspecies quorum quenching (QQ) by bacteria has been reported as a novel approach in biofouling control in membrane bioreactor (MBR) for wastewater treatment. In this study, seven QQ bacteria, named RO1 to RO7, were isolated from fouling reverse osmosis (RO) membrane and were encapsulated in the microbial filters for biofouling control in MBR. Results showed that all isolates have extracellular QQ activity, and RO1 to RO6 were characterized as heat-stable enzymatic degradation of N-acyl homoserine lactones (AHLs), while RO7 was mainly pH-dependent. Analysis of the 16S rRNA gene indicated that RO7 was a species of Pseudomonas sp., others belonged to the species of Bacillus, Stenotrophomonas maltophilia and Delftia tsuruhatensis, respectively. Then, the microbial filters encapsulated each isolate were applied in a laboratory-scaled MBR and proved the substantial inhibition effect on membrane biofouling except for RO7, indicating that QQ bacteria employing enzymes to degrade AHLs have great potential in biofouling control in MBR.     

A two-phase variable neighbourhood search algorithm for assembly line worker assignment and balancing problem type-II: an industrial case study

The assembly line worker assignment and balancing problem type-II (ALWABP-2) occurs when workers and tasks (where task times depend on workers’ skills) are to be simultaneously assigned to a fixed number of workstations with the goal of minimising the cycle time. In this study, a two-phase variable neighbourhood search (VNS) algorithm is proposed to solve the ALWABP-2 due to the NP-hard nature of this problem. In the first phase of the algorithm, a VNS approach is applied to assign tasks to workstations with the aim of minimising the cycle time while in the second phase, a variable neighbourhood descent method is applied to assign workers to workstations. The performance of the proposed algorithm is tested on well-known benchmark instances. In addition, the proposed algorithm has been used to solve a real case study from a consumer electronics company that manufactures LCD TVs. The results show that the algorithm is superior to the methods reported in the literature in terms of its higher efficiency and robustness. Furthermore, the algorithm is easy to implement and significantly improves the performance of the final assembly line for the investigated LCD TV real case study.     

Measurement Uncertainty Budget of the PMV Thermal Comfort Equation

Fanger’s predicted mean vote (PMV) equation is the result of the combined quantitative effects of the air temperature, mean radiant temperature, air velocity, humidity activity level and clothing thermal resistance. PMV is a mathematical model of thermal comfort which was developed by Fanger. The uncertainty budget of the PMV equation was developed according to GUM in this study. An example is given for the uncertainty model of PMV in the exemplification section of the study. Sensitivity coefficients were derived from the PMV equation. Uncertainty budgets can be seen in the tables. A mathematical model of the sensitivity coefficients of \(T_{\mathrm{a}}\), \(h_{\mathrm{c}}\), \(T_{\mathrm{mrt}}\), \(T_{\mathrm{cl}}\), and \(P_{\mathrm{a}}\) is given in this study. And the uncertainty budgets for \(h_{\mathrm{c}}\), \(T_{\mathrm{cl}}\), and \(P_{\mathrm{a}}\) are given in this study.     

5-5′型木质素二聚体高温热解微观化学反应机理研究

选用5-5′键型连接的木质素二聚体为研究对象,基于B3LYP/6-31g基组和密度泛函理论(DFT)对其进行了高温热解微观化学反应机理研究,设计了5-5′型木质素二聚体高温热解反应的5条初反应路径,并在分析初反应结果的基础上设计了7条次反应路径,对每条路径的反应物及其产物进行了结构优化,计算了1500K温度下每条反应路径的热力学及动力学参数。结果表明,5-5′型木质素苯环位上的甲氧基具有很强的活性,甲氧基中O-CH3键以及Caromatic-OCH3键较易断裂;通过计算得出初反应的R3、R4反应焓变较小,分别为192.7kJ/mol和270.5kJ/mol,认为该热解反应的主反应路径以R3及R4为主;该热解反应得到的最终产物有3-4,二羟基甲苯、对羟基甲苯、2-乙烯-4-甲基-3,5-二烯环己酮、甲烷、甲醇及甲醛。