PSTCNN: Explainable COVID-19 diagnosis using PSO-guided self-tuning CNN

Since 2019, the coronavirus disease-19 (COVID-19) has been spreading rapidly worldwide, posing an unignorable threat to the global economy and human health. It is a disease caused by severe acute respiratory syndrome coronavirus 2, a single-stranded RNA virus of the genus Betacoronavirus. This virus is highly infectious and relies on its angiotensin-converting enzyme 2-receptor to enter cells. With the increase in the number of confirmed COVID-19 diagnoses, the difficulty of diagnosis due to the lack of global healthcare resources becomes increasingly apparent. Deep learning-based computer-aided diagnosis models with high generalisability can effectively alleviate this pressure. Hyperparameter tuning is essential in training such models and significantly impacts their final performance and training speed. However, traditional hyperparameter tuning methods are usually time-consuming and unstable. To solve this issue, we introduce Particle Swarm Optimisation to build a PSO-guided Self-Tuning Convolution Neural Network (PSTCNN), allowing the model to tune hyperparameters automatically. Therefore, the proposed approach can reduce human involvement. Also, the optimisation algorithm can select the combination of hyperparameters in a targeted manner, thus stably achieving a solution closer to the global optimum. Experimentally, the PSTCNN can obtain quite excellent results, with a sensitivity of 93.65% ± 1.86%, a specificity of 94.32% ± 2.07%, a precision of 94.30% ± 2.04%, an accuracy of 93.99% ± 1.78%, an F1-score of 93.97% ± 1.78%, Matthews Correlation Coefficient of 87.99% ± 3.56%, and Fowlkes-Mallows Index of 93.97% ± 1.78%. Our experiments demonstrate that compared to traditional methods, hyperparameter tuning of the model using an optimisation algorithm is faster and more effective.     

FORAGING IN CHEMICALLY DIVERSE ENVIRONMENTS: ENERGY,PROTEIN, AND ALTERNATIVE FOODS INFLUENCE INGESTION OF PLANT SECONDARY METABOLITES BY LAMBS

Interactions among nutrients and plant secondary metabolites (PSM) may influence how herbivores mix their diets and use food resources. We determined intake of a food containing a mix of terpenoids identified in sagebrush (Artemisia tridentata) when present in isoenergetic diets of increasing concentrations of protein (6, 9, 15, or 21% CP) or in isonitrogenous diets of increasing concentrations of energy (2.17, 2.55, 3.30, or 3.53 Mcal/kg). Lambs were offered choices between those diets with or without terpenes or between diets with terpenes and alfalfa hay. Intake of the diets with terpenes was lowest with the lowest concentrations of protein (6%) and energy (2.17 Mcal/kg) in the diets, and highest with diets of 15% CP and 3.53~Mcal/kg. In contrast, when terpenes were absent from the diets, lambs consumed similar amounts of all four diets with different concentrations of protein, and more of the diets with intermediate amounts of energy. When given a choice between the diet with or without terpenes, lambs preferred the diet without terpenes. When lambs were offered choices between terpene-containing diets and alfalfa, energy and protein concentrations influenced the amount of terpenes animals ingested. Energy densities higher than alfalfa, and protein concentrations higher than 6%, increased intake of the terpene-containing diet. Thus, the nutritional environment interacted with terpenes to influence preference such that lambs offered diets of higher energy or protein concentration ate more terpenes when forced, but not when offered alternative food without terpenes. The nutrients supplied by a plant and its neighbors likely influence how much PSM an animal can ingest, which in turn may affect the dynamics of plant communities, and the distribution of herbivores in a landscape. We discuss implications of these findings for traditional views of grazing refuges and varied diets in herbivores.     

Source Allocation of Aliphatic and Polycyclic Aromatic Hydrocarbons in Particulate-Phase (PM10) in the City of Valdivia,Chile

The concentration and signature of n-alkanes (n-C₁₀ to n-C₃₃) and 18 PAHs were determined in air filters across a year period (2010) in an urban area of the city of Valdivia, Chile. Filter samples were extracted using sohxlet apparatus and analyzed by GC-MS techniques. Concentrations of total hydrocarbons ranged from 45–352 ng.m^?3 and total PAHs ranged from 2.93–78.01 ng.m^?3. Concentrations of hydrocarbons during the summer were high (288–352 ng.m^-3) and reduced when the autumn began (45–79 ng.m^?3) to then increase almost linearly to the next summer. The drop in concentration was attributed in part to the significant reduction of traffic when summer ends as tourists leave the city (about 9–15% of the total cars circulating). Results from the chemometric technique of Polytopic Vector Analysis (PVA) indicated three main sources for the alkanes: biogenic (terrestrial plants), signatures of oil combustion, and an unconfirmed source which is thought to come from non specific organic matter degradation. Total PAHs correlated well with total particulate matter with a R² = 0.94. Levels of PAHs in the atmosphere were higher during the winter (6.85–78.01 ng.m^?3) period than the rest of the year (2.93–36.30 ng.m^?3). PVA results indicate three key sources of PAHs and two of those sources derived from oil combustion and biomass burning.     

基于蓝牙技术的无线数据采集系统的设计

针对数据采集场合电缆连接复杂、易受电磁干扰等问题,设计了一种基于蓝牙模块进行无线传输的单片机数据采集系统,实现了对现场数据的采集和短距离内数据的无线传输,对于类似的数据采集系统的设计具有很好的借鉴意义.     

拥挤收费政策交通影响评价

以拥挤收费政策交通影响评价为研究对象,运用非集计出行需求预测理论,建立了拥挤收费政策交通影响评价指标体系和居民政策响应行为预测模型系统,从出行方式、时间、空间分布等多方面综合评价拥挤收费政策对交通系统产生的影响,并对佛山市南海区实施拥挤收费政策的可行性进行了实证分析。结果表明,该政策的实施有利于南海区平衡出行时间分布,降低拥挤区域道路饱和度,但在交通结构优化和出行时间调整方面效果不甚理想。本文建立的以非集计理论为基础的"交通状况分析-管理政策制定-响应行为预测-实施效果评价"分析模式对我国城市制定、评价和实施相关交通需求管理政策有一定借鉴意义。     

基于多模态时间序列建模的机器人安全监控

针对多模态时间序列建模的机器人安全监控问题,基于多模态粘性层级狄利克雷过程隐马尔可夫模型（MD-sHDPHMM）提出了一种能准确、可靠、快速、通用地实现机器人执行过程识别与异常事件监测的方法.该方法首先把机器人的执行任务分割成序列化的执行过程,然后通过对比执行过程中累积观察数据的对数似然函数值的大小实现执行过程的识别. 在此基础上,根据正常执行过程训练得到的对数似然函数值的梯度阈值,实现了机器人执行过程中的实时异常监测. 实验结果表明,该方法能有效地实现基于多模态时间序列建模的机器人安全监控.