An Interpretable CNN for the Segmentation of the Left Ventricle in Cardiac MRI by Real-Time Visualization

The interpretability of deep learning models has emerged as a compelling area in artificial intelligence research.The safety criteria for medical imaging are highly stringent,and models are required for an explanation.However,existing convolutional neural network solutions for left ventricular segmentation are viewed in terms of inputs and outputs.Thus,the interpretability of CNNs has come into the spotlight.Since medical imaging data are limited,many methods to fine-tune medical imaging models that are popular in transfer models have been built using massive public Image Net datasets by the transfer learning method.Unfortunately,this generates many unreliable parameters and makes it difficult to generate plausible explanations from these models.In this study,we trained from scratch rather than relying on transfer learning,creating a novel interpretable approach for autonomously segmenting the left ventricle with a cardiac MRI.Our enhanced GPU training system implemented interpretable global average pooling for graphics using deep learning.The deep learning tasks were simplified.Simplification included data management,neural network architecture,and training.Our system monitored and analyzed the gradient changes of different layers with dynamic visualizations in real-time and selected the optimal deployment model.Our results demonstrated that the proposed method was feasible and efficient:the Dice coefficient reached 94.48%,and the accuracy reached 99.7%.It was found that no current transfer learning models could perform comparably to the ImageNet transfer learning architectures.This model is lightweight and more convenient to deploy on mobile devices than transfer learning models.

A collaborative approach to integrated energy systems that consider direct trading of multiple energy derivatives

The cooperative model of a multi-subject Regional Integrated Energy System(RIES)is no longer limited to the trading of traditional energy,but the trading of new energy derivatives such as Green Certificates(GC),Service Power(SP),and CO_(2) will be more involved in the energy allocation of the cooperative model.This study was conducted for the multi-entity RIES cooperative model considering the trading of electronics,GC,SP,and CO_(2).First,a cooperative framework including wind-photovoltaic generation system(WG),combined heat and power system(CHP),and power-carbon-hydrogen load(PCH)is proposed,and the mechanism of energy derivatives trading is also analyzed.Then,the sub-models of each agent in the cooperative model are established separately so that WG has the capability of GC generation,CHP has the capability of GC and CO_(2) absorption,and PCH can realize the effective utilization of CO_(2).Then,the WG–CHP–PCH cooperative model is established and equated into two sub-problems of cooperative benefit maximization and transaction payment negotiation,which are solved in a distributed manner by the alternating directed multiplier method(ADMM).Finally,the effectiveness of the proposed cooperative model and distributed solution is verified by simulation.The simulation results show that the WG–CHP–PCH cooperative model can substantially improve the operational efficiency of each agent and realize the efficient redistribution of energy and its derivatives.In addition,the dynamic parameter adjustment algorithm(DP)is further applied in the solving process to improve its convergence speed.By updating the step size during each iteration,the computational cost,the number of iterations,and the apparent oscillations are reduced,and the convergence performance of the algorithm is improved.

Material properties and machining performance of hybrid Ti_2AlN bulk material for micro electrical discharge machining

Mn+1AXn(MAX) phases are a family of nanolaminated compounds that possess unique combination of typical ceramic properties and typical metallic properties.As a member of MAX phase,Ti2 AlN bulk materials are attractive for some high-temperature applications.The synthesis,characteristics and machining performance of hybrid Ti2 AlN bulk materials were focused on in this work.The bulk samples mainly consisting of Ti2 AlN MAX phase with density close to theoretic one were synthesized by a spark plasma sintering method.Scanning electron microscopy results indicate homogenous distribution of Ti2 AlN grains in the samples.Micro-hardness values are almost constant under different loads (6-6.5 GPa).A machining test was carried out to compare the effect of material properties on micro-electrical discharge machining (micro-EDM) performance for Ti2 AlN bulk samples and Ti6242 alloy.The machining performance of the Ti2 AlN sample is better than that of the Ti6242 alloy.The inherent mechanism was discussed by considering their electrical and thermal conductivity.

Analytical model of IoT CoAP traffic

The Constrained Application Protocol (CoAP) is a mechanism that enables the efficient transmission of information in the context of Wireless Sensor Networks (WSNs) characteristic of the Internet of Things (IoT). CoAP is specifically intended to work well with low complexity sensors that have lim让ed access to computational resources. And because it lacks the overhead of other more sophisticated transport protocols,让 is suitable for high packet loss wireless networks. Relying on the User Datagram Protocol (UDP), CoAP introduces, among other things, confirmable and non-confirmable message types, depending on the nature of the source data. This paper presents a novel analytical model that can relate the application to network packet loss and be used to estimate the effect of lossy wireless channels in CoAP scenarios. Moreover, the model is ultimately validated by an experimental framework in which theoretical and experimental results are compared.

Progressive Collapse of RC Frames Due to Heavy Impact Loads of Tsunami

Progressive collapse is a relatively rare event, as it requires both an abnormal loading to initiate the local damage and a structure that lacks adequate continuity, ductility and redundancy to resist the spread of damage. However, significant casualties can result when collapse occurs. Heavy impact loads due to tsunami against building can be one of the scenarios of progressive collapse during tsunami disaster. Since progressive collapse includes material and geometry nonlinearity during collapse propagation, in the present research capability of 2 models for the material nonlinearity in simulating actual behavior of structures during collapse is compared with recent experimental results of a Reinforced Concrete (RC) frame. The results demonstrate that a material nonlinearity model, that is based on the idealized component load-deformation behavior, is not a proper representation for the real behavior of structures during progressive collapse and is so conservative.

Single Failure Routing Protection Algorithm in the Hybrid SDN Network

Loop free alternate(LFA)is a routing protection scheme that is currently deployed in commercial routers.However,LFA cannot handle all single network component failure scenarios in traditional networks.As Internet service providers have begun to deploy software defined network(SDN)technology,the Internet will be in a hybrid SDN network where traditional and SDN devices coexist for a long time.Therefore,this study aims to deploy the LFA scheme in hybrid SDN network architecture to handle all possible single network component failure scenarios.First,the deployment of LFA scheme in a hybrid SDN network is described as a 0-1 integer linear programming(ILP)problem.Then,two greedy algorithms,namely,greedy algorithm for LFA based on hybrid SDN(GALFAHSDN)and improved greedy algorithm for LFA based on hybrid SDN(IGALFAHSDN),are proposed to solve the proposed problem.Finally,both algorithms are tested in the simulation environment and the real platform.Experiment results show that GALFAHSDN and IGALFAHSDN can cope with all single network component failure scenarios when only a small number of nodes are upgraded to SDN nodes.The path stretch of the two algorithms is less than 1.36.

Efficient Routing Protection Algorithm in Large-Scale Networks

With an increasing urgent demand for fast recovery routing mechanisms in large-scale networks,minimizing network disruption caused by network failure has become critical.However,a large number of relevant studies have shown that network failures occur on the Internet inevitably and frequently.The current routing protocols deployed on the Internet adopt the reconvergence mechanism to cope with network failures.During the reconvergence process,the packets may be lost because of inconsistent routing information,which reduces the network’s availability greatly and affects the Internet service provider’s(ISP’s)service quality and reputation seriously.Therefore,improving network availability has become an urgent problem.As such,the Internet Engineering Task Force suggests the use of downstream path criterion(DC)to address all single-link failure scenarios.However,existing methods for implementing DC schemes are time consuming,require a large amount of router CPU resources,and may deteriorate router capability.Thus,the computation overhead introduced by existing DC schemes is significant,especially in large-scale networks.Therefore,this study proposes an efficient intra-domain routing protection algorithm(ERPA)in large-scale networks.Theoretical analysis indicates that the time complexity of ERPA is less than that of constructing a shortest path tree.Experimental results show that ERPA can reduce the computation overhead significantly compared with the existing algorithms while offering the same network availability as DC.

Energy-Efficient Routing Algorithm Based on Multipath Routing in Large-Scale Networks

A reduction in network energy consumption and the establishment of green networks have become key scientific problems in academic and industrial research.Existing energy efficiency schemes are based on a known traffic matrix,and acquiring a real-time traffic matrix in current complex networks is difficult.Therefore,this research investigates how to reduce network energy consumption without a real-time traffic matrix.In particular,this paper proposes an intra-domain energy-efficient routing scheme based on multipath routing.It analyzes the relationship between routing availability and energy-efficient routing and integrates the two mechanisms to satisfy the requirements of availability and energy efficiency.The main research focus is as follows:(1)A link criticality model is evaluated to quantitatively measure the importance of links in a network.(2)On the basis of the link criticality model,this paper analyzes an energy-efficient routing technology based on multipath routing to achieve the goals of availability and energy efficiency simultaneously.(3)An energy-efficient routing algorithm based on multipath routing in large-scale networks is proposed.(4)The proposed method does not require a real-time traffic matrix in the network and is thus easy to apply in practice.(5)The proposed algorithm is verified in several network topologies.Experimental results show that the algorithm can not only reduce network energy consumption but can also ensure routing availability.

Novel technique of controlled laser air-force detection for rheological properties of polymers

A novel controlled laser air-force detection(CLAFD)technique was developed to detect the rheological properties of polymers with the characteristics of non-destruction and cross-contamination free.Dynamic testing and static testing were carried out in the technique.Back propagation neural network algorithm was used to establish the air-force control model.The replicability of CLAFD system was analyzed,the viscoelastic properties of polyurethane were investigated using alternating load testing.A comparative analysis of performances was made between the CLAFD and the texture analysis(TA)on the testing of creep-recovery and stress relaxation.The results demonstrated that the CLAFD system had good replicability.The lagging phase angle was between 0°-90°in the testing of alternating load.This illustrated that the CLAFD technique could be used to detect viscoelasticity.The parameters of response speed and the precision of the CLAFD entirely surpassed the TA on the creep-recovery testing.The CLAFD technique will provide a new real-time,non-destruction and cross-contamination-free detection method for material science,especially for those materials such as artificial biological tissue and function food products.

Bacterial extracellular vesicle applications in cancer immunotherapy

Cancer therapy is undergoing a paradigm shift toward immunotherapy focusing on various approaches to activate the host immune system.As research to identify appropriate immune cells and activate anti-tumor immunity continues to expand,scientists are looking at microbial sources given their inherent ability to elicit an immune response.Bacterial extracellular vesicles(BEVs)are actively studied to control systemic humoral and cellular immune responses instead of using whole microorganisms or other types of extracellular vesicles(EVs).BEVs also provide the opportunity as versatile drug delivery carriers.Unlike mammalian EVs,BEVs have already made it to the clinic with the meningococcal vaccine(Bexsero®).However,there are still many unanswered questions in the use of BEVs,especially for chronic systemically administered immunotherapies.In this review,we address the opportunities and challenges in the use of BEVs for cancer immunotherapy and provide an outlook towards development of BEV products that can ultimately translate to the clinic.