Search

PurposeTo remedy the inherent defect in current research that focuses only on a single type of participants, this paper endeavors to look into the situation as an evolutionary game between a representative Logistics Service Integrator (LSI) and a representative Functional Logistics Service Provider (FLSP) in an environment with sudden crisis and tries to analyze how LSI supervises FLSP's operations and how FLSP responds in a recurrent pattern with different interruption probabilities.Design/methodology/approachRegarding the risks of supply chain interruption in emergencies, this paper develops a two-level model of single LSI and single FLSP, using Evolutionary Game theory to analyze their optimal decision-making, as well as their strategic behaviors on different risk levels regarding the interruption probability to achieve the optimal return with bounded rationality.FindingsThe results show that on a low-risk level, if LSI increases the degree of punishment, it will fail to enhance FLSP's operational activeness in the long term; when the risk rises to an intermediate level, a circular game occurs between LSI and FLSP; and on a high level of risk, FLSP will actively take actions, and its functional probability further impacts LSI's strategic choices. Finally, this paper analyzes the moderating impact of punishment intensity and social reputation loss on the evolutionary model in emergencies and provides relevant managerial implications.Originality/valueFirst, by taking both interruption probability and emergencies into consideration, this paper explores the interactions among the factors relevant to LSI's and FLSP's optimal decision-making. Second, this paper analyzes the optimal evolutionary game strategies of LSI and FLSP with different interruption probability and the range of their optimal strategies. Third, the findings of this paper provide valuable implications for relevant practices, such that the punishment intensity and social reputation loss determine the optimal strategies of LSI and FLSP, and thus it is an effective vehicle for LSSC system administrator to achieve the maximum efficiency of the system.

To investigate the influence of epileptogenic cortex (Rolandic areas) with executive functions in Rolandic epilepsy using structural covariance analysis of structural magnetic resonance imaging (MRI). Structural MRI data of drug-naive patients with Rolandic epilepsy (n = 70) and typically developing children as healthy controls (n = 83) were analyzed using voxel-based morphometry. Gray matter volumes in the patients were compared with those of healthy controls, and were further correlated with epilepsy duration and cognitive score of executive function, respectively. By applying Granger causal analysis to the sequenced morphometric data according to disease progression information, causal network of structural covariance was constructed to assess the causal influence of structural changes from Rolandic cortices to the regions engaging executive function in the patients. Compared with healthy controls, epilepsy patients showed increased gray matter volume in the Rolandic regions, and also the regions engaging in executive function. Covariance network analyses showed that along with disease progression, the Rolandic regions imposed positive causal influence on the regions engaging in executive function. In the patients with Rolandic epilepsy, epileptogenic regions have causal influence on the structural changes in the regions of executive function, implicating damaging effects of Rolandic epilepsy on human brain.

Although Rolandic epilepsy (RE) has been regarded as a brain developmental disorder, neuroimaging studies have not yet ascertained whether RE has brain developmental delay. This study employed deep learning–based neuroanatomic biomarker to measure the changed feature of “brain age” in RE. The study constructed a 3D-CNN brain age prediction model through 1155 cases of typically developing children’s morphometric brain MRI from open-source datasets and further applied to a local dataset of 167 RE patients and 107 typically developing children. The brain-predicted age difference was measured to quantitatively estimate brain age changes in RE and further investigated the relevancies with cognitive and clinical variables. The brain age estimation network model presented a good performance for brain age prediction in typically developing children. The children with RE showed a 0.45-year delay of brain age by contrast with typically developing children. Delayed brain age was associated with neuroanatomic changes in the Rolandic regions and also associated with cognitive dysfunction of attention. This study provided neuroimaging evidence to support the notion that RE has delayed brain development. • The children with Rolandic epilepsy showed imaging phenotypes of delayed brain development with increased GM volume and decreased WM volume in the Rolandic regions. • The children with Rolandic epilepsy had a 0.45-year delay of brain-predicted age by comparing with typically developing children, using 3D-CNN-based brain age prediction model. • The delayed brain age was associated with morphometric changes in the Rolandic regions and attentional deficit in Rolandic epilepsy.

A stacked voltage domain SRAM is proposed as an effective leakage reduction technique where bit-cell arrays are split into two voltage domains (top and bottom) connecting in series between VDD and GND to generate a subthreshold retention voltage directly from a nominal supply with no area penalty or efficiency loss compared to the conventional voltage regulator approach. The Zigzag 8T bit-cell structure is chosen with an optimal transistor sizing to balance among hold stability, leakage, and area density. SRAM peripherals remains at full supply domain resulting in super-cutoff read for improved sensing margin and word-line overdrive for better write margin. A novel array swapping mechanism with a comprehensive timing control ensure stable access to arbitrary arrays within one system clock cycle. The proposed SRAM achieves 1.03-pW/b leakage at 0.58 V in 40 nm.

PurposeThis study proposed and examined a unique but underresearched strategic response to coercive institutional pressure, covert neutralization, in which organizations neutralize coercive institutional pressure by manipulating influential factors within their discretion while conforming to the pressure. In addition, this study pointed out that the extent of covert neutralization is contingent on organizational decision-makers' willingness and ability to take countermeasures.Design/methodology/approachTo find evidence for covert neutralization and its strategic nature, a historical review with secondary data was conducted in the context of the coercive introduction of the independent director system in China. The board structure adjustments of 318 Chinese A-share listed companies from 2000 to 2004 were examined using univariateT-tests, robust ordinary least squares (OLS) regression models and ordered logit regression models to test the hypotheses.FindingsThis study found that in response to the coercive introduction of the independent director system, Chinese listed companies adjusted their board structures while conforming to the system, which neutralized the constraints on controlling shareholders' board control. Specifically, listed companies reduced the number of dependent directors and the ratio of directors representing noncontrolling shareholders to those representing controlling shareholders. Moreover, the extent of board structure adjustments was positively associated with the shareholding percentage of controlling shareholders. These findings provide evidence for covert neutralization and its strategic nature.Practical implicationsThis study indicates the inefficacy of simple legal coercion and suggests that authorities should be aware of the covert countermeasures used by regulated organizations and employ normative approaches to improve the efficacy of regulations. This implication is rather instructive for regulators to effectively promote imposed changes in transitional economies with strong governments.Originality/valueBy differentiating conformity behavior from conformity inclination and considering the inevitable incompleteness of laws, this study challenges the conventional wisdom regarding organizational conformity and shows that organizations can strategically respond to coercive institutional pressure while conforming to it. In contrast to other strategic responses that aim to defy, avoid or resist the implementation (the means) of institutional pressure, covert neutralization aims to neutralize its efficacy (the ends).

Purpose We assessed the potential of resting-state fMRI lag analysis (RSLA) in detecting epileptic activation and in estimating anti-epileptic effects of Levetiracetam (LEV) in Rolandic epilepsy. Methods Forty-three children with Rolandic epilepsy underwent simultaneous EEG-fMRI. They were grouped into LEV vs drug-naive groups according to their medication, and into patients who showed or did not show central-temporal spike (CTS) discharges during scans. We calculated the lag structure of rs-fMRI for all patients and assessed interactions with drug (LEV vs. drug-naive) and CTS status (CTS vs. no-CTS). We furthermore assessed correlations between lag values and number of CTS and medication conditions. Results RSLA analysis indicated earlier intrinsic activations in bilateral Rolandic areas when CTS occurred. More frequent epileptic discharges were correlated with earlier intrinsic activations (r=-0.46, p = 0.03 left Rolandic). Patients with LEV therapy, on the other hand, displayed delayed intrinsic activity in Rolandic areas compared to drug-naive patients Conclusion Our RSLA analysis indicated an association between centro-temporal spikes and earlier hemodynamic activations in epileptogenic regions in Rolandic epilepsy, which were counteracted by LEV treatment. As it allows for the mapping of propagation features of intrinsic activity and drug-effects, our findings suggest potential of lag based analyses in detecting focus localization and estimating treatment effects.

To develop and validate a deep learning model to discriminate transient from persistent subsolid nodules (SSNs) on baseline CT. A cohort of 1414 SSNs, consisting of 319 transient SSNs in 168 individuals and 1095 persistent SSNs in 816 individuals, were identified on chest CT. The cohort was assigned by examination date into a development set of 996 SSNs, a tuning set of 212 SSNs, and a validation set of 206 SSNs. Our model was built by transfer learning, which was transferred from a well-performed deep learning model for pulmonary nodule classification. The performance of the model was compared with that of two experienced radiologists. Each nodule was categorized by Lung CT Screening Reporting and Data System (Lung-RADS) to further evaluate the performance and the potential clinical benefit of the model. Two methods were employed to visualize the learned features. Our model achieved an AUC of 0.926 on the validation set with an accuracy of 0.859, a sensitivity of 0.863, and a specificity of 0.858, and outperformed the radiologists. The model performed the best among Lung-RADS 2 nodules and maintained well performance among Lung-RADS 4 nodules. Feature visualization demonstrated the model’s effectiveness in extracting features from images. The transfer learning model presented good performance on the discrimination between transient and persistent SSNs. A reliable diagnosis on nodule persistence can be achieved at baseline CT; thus, an early diagnosis as well as better patient care is available. • Deep learning can be used for the discrimination between transient and persistent subsolid nodules. • A transfer learning model can achieve good performance when it is transferred from a model with a similar task. • With the assistance of deep learning model, a reliable diagnosis on nodule persistence can be achieved at baseline CT, which can bring a better patient care strategy.

Unmanned aerial vehicles (UAVs) have played an important role in recent high-tech local wars. Seizing air control rights with UAVs will undoubtedly be a popular topic in future military development. Autonomous air combat is complex, antagonistic and mutable, and consequently, the decision-making that depends on unmanned systems is extremely challenging with very little research having been conducted on it. An intelligent air combat learning system inspired by the learning mechanisms of the brain is proposed in this paper. In accordance with research on learning, knowledge and memory, we constructed a cognitive mechanism model of the brain. Based on this model and the inferential abilities of humans, a long short-term hierarchical multi-line learning system is established. Then, the bio-inspired architecture and the basic learning principle of the system are clarified. Taking advantage of the conclusions of studies on information theory, the relationship between the knowledge updating cycle and the system learning performance is analysed. The updating cycle length adjustment problem is transformed into an optimization problem optimization problem, and system performance improvement is guaranteed. Experiments show that the system designed in this paper can acquire confrontation abilities through self-learning without prior rules; the parallel universe mechanism can significantly improve the system's learning speed when the number of parallels is within 40, and the performance of the system improves gradually and continuously. The system can master actions similar to classical tactical manoeuvres such as the high yo-yo and the barrel-roll-attack without prior knowledge. Compared with the Bayesian inference and moving horizon optimization (BI&MHO) method, the learning system proposed in this paper is more flexible in situation assessment and in the prediction of opponents' actions. Although it cannot be deployed quickly, it has a continuous learning ability.

The bottleneck associated with the validation of the parameters of the entropy model has limited the application of this model to modern functional imaging technologies such as the resting-state functional magnetic resonance imaging (rfMRI). In this study, an optimization algorithm that could choose the parameters of the multiscale entropy (MSE) model was developed, while the optimized effectiveness for localizing the epileptogenic hemisphere was validated through the classification rate with a supervised machine learning method. The rfMRI data of 20 mesial temporal lobe epilepsy patients with positive indicators (the indicators of epileptogenic hemisphere in clinic) in the hippocampal formation on either left or right hemisphere (equally divided into two groups) on the structural MRI were collected and preprocessed. Then, three parameters in the MSE model were statistically optimized by both receiver operating characteristic (ROC) curve and the area under the ROC curve value in the sensitivity analysis, and the intergroup significance of optimized entropy values was utilized to confirm the biomarked brain areas sensitive to the epileptogenic hemisphere. Finally, the optimized entropy values of these biomarked brain areas were regarded as the feature vectors input for a support vector machine to classify the epileptogenic hemisphere, and the classification effectiveness was cross-validated. Nine biomarked brain areas were confirmed by the optimized entropy values, including medial superior frontal gyrus and superior parietal gyrus ( p

With the rapid development of Internet technology, more and more people begin to pay attention to and study the information mining and data analysis in social networks, and the maximization of influence widely used in marketing have gradually become the focus of people's research. Nowadays, most of the researches on the maximization of influence under topic preference in social network information dissemination only tend to judge the influence of nodes on a certain information from the interaction of nodes to the topic content of information. The combination of the topology of social networks and the topic content of information dissemination is not considered. In view of this situation, an influence maximization algorithm under topic preference(IMATP algorithm) is proposed, which evaluates the comprehensive influence of nodes by combining node centrality and theme influence. Experimental results on data sets of different sizes from Twitter show that the algorithm proposed in this paper is closer to the actual propagation situation. And compared with the classical algorithm, this algorithm not only improves the scope of influence, but also improves the running efficiency.

Brain structural covariance network (SCN) can delineate the brain synchronized alterations in a long-range time period. It has been used in the research of cognition or neuropsychiatric disorders. Recently, causal analysis of structural covariance network (CaSCN), winner-take-all and cortex–subcortex covariance network (WTA-CSSCN), and modulation analysis of structural covariance network (MOD-SCN) have expended the technology breadth of SCN. However, the lack of user-friendly software limited the further application of SCN for the research. In this work, we developed the graphical user interface (GUI) toolkit of brain structural covariance connectivity based on MATLAB platform. The software contained the analysis of SCN, CaSCN, MOD-SCN, and WTA-CSSCN. Also, the group comparison and result-showing modules were included in the software. Furthermore, a simple showing of demo dataset was presented in the work. We hope that the toolkit could help the researchers, especially clinical researchers, to do the brain covariance connectivity analysis in further work more easily.

Humans have a fundamental ability that is to learn others’ experience for their own use, while humanoid robots don’t have. Several attempts have been made for specific situations in evolution and study of developmental robots. However, such attempts have provided limitations, e.g. others’ experience learning get overlooked. The present article proposes peers’ experience learning method, which first reviewed the evolution and development of developmental robots as some typical studies revealed, moving from humanlike to developmental. These terms are then reconsidered from humanoid robots’ viewpoint, particularly with the developmental principles: the verification principle and the embodiment principle. Next, a conceptual model of peers’ experience learning is proposed based on the principles, and the simulation results show that robots can “copy” peers’ experience to cognize and develop automatically. Finally, a general discussion and proposals for addressing future issues are given.

Social group behavior analysis has always been the key research direction of sociologists and psychologists. With the rapid development of the Internet of Things and the proposal of deep learning theory, convolutional neural networks are also used in social group research. In the process of social development, group incidents continue to increase, and there are more and more studies on social group behavior analysis. Although the research content and research methods are also richer, the research that combines the Internet of Things, convolutional neural network, and group behavior is more. This article will specifically propose a social group behavior analysis model that combines multitask learning and convolutional neural networks. This paper deeply learns the research of convolutional neural network and group behavior-related theories, makes full use of the advantages of convolutional neural network algorithm and multitask learning mode, and builds a social group behavior analysis model based on multitask learning and convolutional neural network. The experimental results on different data sets are analyzed. The results show that the accuracy rate of the experimental algorithm of convolutional neural network is as high as 95.10%, and it is better than other algorithms in time complexity, which is very suitable for social group behavior analysis.

In recent years, there has been increasing awareness of the occurrence of fatigue fractures. Athletes and soldiers, who engaged in unaccustomed, repetitive or vigorous activities, are potential victims of such a fracture. Due to the slow-growing process of fatigue fracture, the early detection can effectively protect athletes and soldiers from the material bone breakage, which may result in the catastrophe of career retirement. In this paper, we propose a triplet-branch network (TBN) for the accurate fatigue fracture grading, which enables physicians to promptly take appropriate treatments. Particularly, the proposed TBN consists of three branches for representation learning, classifier learning and grade-related prior-knowledge learning, respectively. The former two branches are responsible to tackle the problem of class-imbalanced training data, while the latter one is implemented to embed grade-related prior-knowledge into the framework via an auxiliary ranking task. Extensive experiments have been conducted on our fatigue fracture X-ray image dataset. The experimental results show that our TBN can effectively address the problem of class-imbalanced training samples and achieve a satisfactory accuracy for fatigue fracture grading.

Active learning is an important method to efficiently reduce the labeled sample size in large-scale machine learning. As the large amount of data presents the nature of imbalance, the study of active learning methods for corresponding unbalanced datasets highlights important practical significance. There is a problem with the current algorithm. The initialization of active learning becomes difficult with the increase of the unbalanced degree, which causes more labels waste. This paper proposes a one-class support vector machine active learning (OCSVM-AL) method for this problem. Firstly, we obtain the structure information of the sample set by k-means clustering. Then, the OCSVM-AL is used on each cluster. For each cluster, OCSVM-AL iteratively marks the sample from the cluster and adds the marked sample to the marked sample pool. A one-class support vector machine model is trained by using majority class samples in the marked sample pool. The trained model is used to predict the unlabeled samples in this cluster, and the labels of unlabeled samples is obtained. By the results of each cluster, all sample labels are obtained. Simulation experiments proved that this method is an effective active learning method and can significantly reduce the labeled sample size.

Purpose To propose a practical strategy for the clinical application of deep learning algorithm, i.e., Hierarchical-Ordered Network-ORiented Strategy (HONORS), and a new approach to pulmonary nodule classification in various clinical scenarios, i.e., Filter-Guided Pyramid NETwork (FGP-NET). Materials and methods We developed and validated FGP-NET on a collection of 2106 pulmonary nodules on computed tomography images which combined screened and clinically detected nodules, and performed external test (n = 341). The area under the curves (AUCs) of FGP-NET were assessed. A comparison study with a group of 126 skilled radiologists was conducted. On top of FGP-NET, we built up our HONORS which was composed of two solutions. In the Human Free Solution, we used the high sensitivity operating point for screened nodules, but the high specificity operating point for clinically detected nodules. In the Human-Machine Coupling Solution, we used the Youden point. Results FGP-NET achieved AUCs of 0.969 and 0.847 for internal and external test. The AUCs of the subsets of the external test set ranged from 0.890 to 0.942. The average sensitivity and specificity of the 126 radiologists were 72.2 ± 15.1 % and 71.7 ± 15.5 %, respectively, while a higher sensitivity (93.3 %) but a relatively inferior specificity (64.0 %) were achieved by FGP-NET. HONORS-guided FGP-NET identified benign nodules with high sensitivity (sensitivity,95.5 %; specificity, 72.5 %) in the screened nodules, and identified malignant nodules with high specificity (sensitivity, 31.0 %; specificity, 97.5 %) in the clinically detected nodules. These nodules could be reliably diagnosed without any intervention from radiologists, via the Human Free Solution. The remaining ambiguous nodules were diagnosed with high performance, which however required manual confirmation by radiologists, via the Human-Machine Coupling Solution. Conclusions FGP-NET performed comparably to skilled radiologists in terms of diagnosing pulmonary nodules. HONORS, due to its high performance, might reliably contribute a second opinion, aiding in optimizing the clinical workflow.

Background: The rs405509 polymorphism ofthe apolipoprotein E (APOE) promoter is related to Alzheimer’sdisease (AD). The T/T allele of rs405509 is known to decrease the transcription of the APOE gene and lead to impairments in specific brain structural networks with aging; thus, it is an important risk factor for AD. However, it remains unknown whether rs405509 affects brain functional connectivity (FC) in aging. Methods: We investigated the effect of the rs405509 genotype (T/T vs. G-allele) on age-related brain FC using functional magnetic resonance imaging. Forty-five elderly TT carriers and 45 elderly G-allele carriers were scanned during a working memory (WM) task. Results: We found that TT carriers showed an accelerated age-related increase in functional activation in the left postcentral gyrus compared with G-allele carriers. Furthermore, the FC between the left postcentral gyrus and some key regions during WM performance, including the right caudal and superior frontal sulcus (SFS), was differentially modulated by age across rs405509 genotype groups. Conclusions: These results demonstrate that the rs405509 T/T allele of APOE causes an age-related brain functional decline in nondemented elderly people, which may be beneficial for understanding the neural mechanisms of rs405509-related cognitive aging and AD pathogenesis.

A stacked voltage domain SRAM is proposed where arrays are split into two sets (top and bottom) with their supplies connected in series. System supply current is reused by top and bottom sets, and supply voltage is divided among the two sets of arrays, enabling seamless integration of very low voltage SRAM retention in a larger system with a nominal supply, without need for an efficiency-reducing LDO. An array swapping approach provides stable access to arbitrary banks within one system clock cycle. A comprehensive sizing strategy (W&L) is employed to optimally balance hold stability and bitcell size. Integrated in an IoT imaging system in 40nm CMOS, the proposed 8.9Mb SRAM achieves 1.03pW/bit leakage, a >100× reduction over conventional SRAM in the same technology.

Generalized tonic–clonic seizures (GTCS) are the severest and most remarkable clinical expressions of human epilepsy. Cortical, subcortical, and cerebellar structures, organized with different network patterns, underlying the pathophysiological substrates of genetic associated epilepsy with GTCS (GE‐GTCS) and focal epilepsy associated with focal to bilateral tonic–clonic seizure (FE‐FBTS). Structural covariance analysis can delineate the features of epilepsy network related with long‐term effects from seizure. Morphometric MRI data of 111 patients with GE‐GTCS, 111 patients with FE‐FBTS and 111 healthy controls were studied. Cortico‐striato‐thalao‐cerebellar networks of structural covariance within the gray matter were constructed using a Winner‐take‐all strategy with five cortical parcellations. Comparisons of structural covariance networks were conducted using permutation tests, and module effects of disease duration on networks were conducted using GLM model. Both patient groups showed increased connectivity of structural covariance relative to controls, mainly within the striatum and thalamus, and mostly correlated with the frontal, motor, and somatosensory cortices. Connectivity changes increased as a function of epilepsy durations. FE‐FBTS showed more intensive and extensive gray matter changes with volumetric loss and connectivity increment than GE‐GTCS. Our findings implicated cortico‐striato‐thalamo‐cerebellar network changes at a large temporal scale in GTCS, with FE‐FBTS showing more severe network disruption. The study contributed novel imaging evidence for understanding the different epilepsy syndromes associated with generalized seizures., In this work, we used a winner‐take‐all strategy‐based structural covariance connectivity analysis, to delineate the cortico‐striato‐thalamo‐cerebellar networks in two syndromes of generalized epilepsy. Our findings implicated cortico‐striato‐thalamo‐cerebellar network changes at a large temporal scale in GTCS, with FE‐FBTS showing more severe network disruption. The study contributes novel imaging evidence for understanding the different epilepsy syndromes associated with generalized seizures.

As a country with rich culture and a long history, China has a large number of cultural relics. The current management system for entry and exit of cultural relics is inefficient and untrusted, which contains a lot of offline operations. What's worse, the cultural relic data is distributed geographically and can be tampered with. In order to solve these problems, this paper proposes a blockchain-based system for managing entry and exit of cultural relics, to improve data traceability, data immutability and workflow efficiency. We use blockchain technology to monitor and trace the whole process of entry and exit of cultural relic. Smart contracts are designed to maintain an immutable access control list and grant privileges to authorized entities. We also combined the blockchain storage with distributed file system to facilitate the scalability and efficiency of our system. We provide the design aspects of system architecture and our technological concerns in system implementation. We also evaluate the computational cost, data transferring cost and storage cost to show that our system is efficient and practical. Our aim is to replace the traditional management system for entry and exit of cultural relics with a blockchain-based one.

The potential pattern changes in brain micro-structure can be used for the brain development assessment in children and adolescents by MRI scans. In this paper, we propose a highly accurate and efficient end-to-end brain age estimation network (BAENET) on T1-weighted MRI images. On the network, 3D skipping and outlier constraint loss are designed to accommodate deeper network and increase the robustness. Besides, we incorporate the neuroimaging domain knowledge into stratified sampling for better generalization ability for datasets with different age distributions, and gender learning for more gender-specific features during modeling. We verify the effectiveness of the proposed method on the public ABIDE2 and ADHD200 benchmark, consisting of 382 and 378 normal children scans respectively. Our BAENET achieves MAE of 1.11 and 1.16, significantly outperforming the best reported methods by 5.1 % and 9.4%.

Background and PurposeLimited optimization was clinically applicable for reducing missed diagnosis, misdiagnosis and inter-reader variability in pulmonary nodule diagnosis. We aimed to propose a deep learning-based algorithm and a practical strategy to better stratify the risk of pulmonary nodules, thus reducing medical errors and optimizing the clinical workflow.Materials and MethodsA total of 2,348 pulmonary nodules (1,215 with lung cancer) containing screened nodules from National Lung Cancer Screening Trial (NLST) and incidentally detected nodules from Jinling Hospital (JLH) were used to train and evaluate a deep learning algorithm, Filter-guided pyramid network (FGP-NET). Internal and external test of FGP-NET were performed on two independent datasets (n=542). The performance of FGP-NET at Youden point which maximizing the Youden index was compared with 126 board-certificated radiologists. We further proposed Hierarchical Ordered Network ORiented Strategy (HONORS), which manipulates the emphasis either on sensitivity or specificity to target risk-stratified clinical scenarios, directly making decisions for some patients.ResultsFGP-NET achieved a high area under the curve (AUC) of 0.969 and 0.855 for internal and external testing, and was comparable or even outperformed the radiologists when considering sensitivity. HONORS-guided FGP-NET identified benign nodules with a high sensitivity (95.5%) in the screening scenario, and demonstrated satisfactory performance for the rest ambiguous nodules with 0.945 of AUC by the Youden point. FGP-NET also detected lung cancer with a high specificity of 94.5% in routine diagnostic scenario; an AUC of 0.809 was achieved for the rest nodules.ConclusionThe combination of HONORS and FGP-NET provides well-organized stratification for pulmonary nodules and also offers the potential for reducing medical errors.HighlightsPulmonary nodules were managed for both screening and diagnostic scenariosProposal of a hierarchical strategy for targeting risk-stratified clinical scenariosA large scale Human-deep learning contest for reliable performance evaluation

Preliminary studies have shown the feasibility of deep learning (DL)-based super-resolution (SR) technique for reconstructing thick-slice/gap diagnostic MR images into high-resolution isotropic data, which would be of great significance for brain research field if the vast amount of diagnostic MRI data could be successively put into brain morphometric study. However, less evidence has addressed the practicability of the strategy, because lack of a large-sample available real data for constructing DL model. In this work, we employed a large cohort (n = 2052) of peculiar data with both low through-plane resolution diagnostic and high-resolution isotropic brain MR images from identical subjects. By leveraging a series of SR approaches, including a proposed novel DL algorithm of Structure Constrained Super Resolution Network (SCSRN), the diagnostic images were transformed to high-resolution isotropic data to meet the criteria of brain research in voxel-based and surface-based morphometric analyses. We comprehensively assessed image quality and the practicability of the reconstructed data in a variety of morphometric analysis scenarios. We further compared the performance of SR approaches to the ground truth high-resolution isotropic data. The results showed (i) DL-based SR algorithms generally improve the quality of diagnostic images and render morphometric analysis more accurate, especially, with the most superior performance of the novel approach of SCSRN. (ii) Accuracies vary across brain structures and methods, and (iii) performance increases were higher for voxel than for surface based approaches. This study supports that DL-based image super-resolution potentially recycle huge amount of routine diagnostic brain MRI deposited in sleeping state, and turning them into useful data for neurometric research.

To investigate the morphological changes of cerebral cortex correlating with anti-seizure medication in Childhood Epilepsy with Centrotemporal Spikes (CECTS), and their relationships with seizure control. This study included a total of 188 children, including 62 patients with CECTS taking anti-seizure drugs, 56 patients with drug-naive, and 70 healthy controls. A portion of cases were also followed-up for longitudinal analysis. Cortical morphological parameters were quantitatively measured by applying surface-based morphometry analysis to high-resolution three-dimension T1 weighted images. Among the three groups, the morphological indices were compared to quantify any cortical changes affected by seizures and medication. The relationships among anti-seizure medication, seizure controls and cortical morphometry were investigated using causal mediator analysis. The Rolandic cortex of the drug-naive patients showed abnormal cortical thickness by comparing with that of healthy controls, and thinning by comparing with that of patients with medication. The cortical thickness in the Rolandic regions was negatively correlated with duration of medication and duration of seizure-free. Longitudinal analysis further demonstrated that the thickness of Rolandic cortex thinned in post-medication state relative to the pre-medication state. Mediation analysis revealed that morphological alteration of the Rolandic cortex might act as a mediator in the path of anti-seizure medication on seizure control. Our findings highlighted that anti-seizure medication was associated with regression of abnormal increment of cortical thickness in the Rolandic regions in CECTS. The neuroanatomical alteration might be a mediating factor in the process of seizure control by anti-seizure medication.

With photoplethysmograph (PPG) sensors showing increasing potential in wearable health monitoring, the challenging problem of motion artifact (MA) removal during intensive exercise has become a popular research topic. In this study, a novel method that combines heart rate frequency (HRF) estimation and notch filtering is proposed. The proposed method applies a cascaded adaptive noise cancellation (ANC) based on the least mean squares (LMS)-Newton algorithm for preliminary motion artifacts reduction, and further adopts special heart rate frequency tracking and correction schemes for accurate HRF estimation. Finally, notch filters are employed to restore the PPG signal with estimated HRF based on its quasi-periodicity. On an open source data set that features intensive running exercise, the proposed method achieves a competitive mean average absolute error (AAE) result of 0.92 bpm for HR estimation. The practical experiments are carried out with the PPG evaluation platform developed by ourselves. Under three different intensive motion patterns, a 0.89 bpm average AAE result is achieved with the average correlation coefficient between recovered PPG signal and reference PPG signal reaching 0.86. The experimental results demonstrate the effectiveness of the proposed method for accurate HR estimation and robust MA removal in PPG during intensive exercise.

Background Repetitive transcranial magnetic stimulation (rTMS) can modulate cortical excitability, and may be beneficial for motor recovery after stroke. However, the neuroplasticity effects of rTMS have not been thoroughly investigated in the early stage after stroke. Objective To comprehensively assess the effects of high- and low-frequency repetitive transcranial magnetic stimulations on motor recovery in early stroke patients, using a randomized controlled trial based on clinical, neurophysiological and functional imaging assessments. Methods Sixty hospitalized, first-ever ischemic stroke patients (within 2 weeks after stroke) with motor deficits were randomly allocated to receive, in addition to standard physical therapy, five consecutive sessions of either: (1) High-frequency (HF) rTMS at 10 Hz over the ipsilesional primary motor cortex (M1); (2) Low-frequency (LF) rTMS at 1 Hz over the contralesional M1; (3) sham rTMS. The primary outcome measure was a motor impairment score (Upper Extremity Fugl-Meyer) evaluated at baseline, after rTMS intervention, and at 3-month follow-up. Cortical excitability and functional magnetic resonance imaging (fMRI) data were obtained within 24 h before and after rTMS intervention. Analyses of variance were conducted to compare the recovery effects among the three rTMS groups, assessed using clinical, neurophysiological and fMRI tests. Results Motor improvement was significantly larger in the two rTMS groups than in the control group. The HF-rTMS group showed significantly increased cortical excitability and motor-evoked fMRI activation in ipsilesional motor areas, whereas the LF-rTMS group had significantly decreased cortical excitability and motor-evoked fMRI activation in contralesional motor areas. Activity in ipsilesional motor cortex significantly correlated with motor function, after intervention as well as at 3-month follow-up. Conclusion HF- and LF-rTMS can both improve motor function by modulating motor cortical activation in the early phase of stroke., Highlights • rTMS is an effective neurorehabilitative strategy for enhancing motor recovery in the early stage after stroke. • Effects of high- versus low-frequency rTMS on motor regions reflect different neuroplastic mechanisms supporting motor recovery. • Functional reorganization of the ipsilesional M1 after stroke plays a critical role in long-term motor recovery.

Humans have a fundamental ability, that is, to share vision among each other to fulfill common goals, which cooperative UAVs do not have. The difficulties mainly lie in the homologous mathematical description of humans and elusive experimental practice. This paper proposed a parallel multiview splicing on clouds, which first review both theory and practice studies in UAVs. These terms are then reconsidered from humans’ vision sharing. Next, a conceptual model of parallel multiview splicing on clouds is proposed and the mathematical deduction if fulfilled. Furthermore, an experimental cooperative UAVs platform is built to practically implement the algorithms. Both the simulated and practiced results have validated the feasibility of our method. Finally, a general discussion and proposals for addressing future issues are given.

This study proposes a new method for motion artifact (MA) removal in Photoplethysmography (PPG) signal that combines accurate heart rate (HR) frequency estimation and notch filtering. The method applies LMS-Newton adaptive filtering to reduce motion artifact noise and uses a novel HR correction stage for accurate HR frequency estimation. Notch filters are used to recover clean PPG signal from HR frequency and second harmonic frequency of PPG. On a widely used dataset of 12 recordings, our method achieves an averaged HR error of 0. 92bpm and a Pearson correlation of 0.997. Experimental results further show that our method can recover the PPG waveform with clear dicrotic peaks even for strongly MA-corrupted PPG signal.

The objective was to study the modulation effects of levetiracetam on the fMRI activation/deactivation patterns associated with centrotemporal spikes (CTS) in Rolandic epilepsy. Forty patients with Rolandic epilepsy, including levetiracetam-medicated patients (n = 20) and drug-naive patients (n = 20), were studied. Single and sequential hemodynamic response functions-based EEG-fMRI analysis was performed to detect dynamic activation/deactivation associated with CTS. Comparisons of spatiotemporal features of activation/deactivation were performed between the two groups. Both the groups (CTS were detected in 12 cases of levetiracetam-medicated group, and 11 cases of drug-naive group) showed CTS-associated activation in the Rolandic cortex, whereas activation strength, time-to-peak delay, and overall activation were diminished in the levetiracetam-medicated group. Moreover, the drug-naive group showed deactivation in the regions engaged in higher cognition networks compared with the levetiracetam-medicated group. Levetiracetam inhibits CTS-associated activation intensity and alters the temporal pattern of this activation in the epileptogenic regions, and it also affects the brain deactivation related to higher cognition networks. The findings sheds a light on the pharmocological mechanism of levetiracetam therapy on Rolandic epilepsy.

Our aim was to investigate regional difference in brain activities in response to antiepileptic drug (AED) medications in benign epilepsy with central-temporal spikes (BECTS) using resting-state functional magnetic resonance imaging (fMRI). Fifty-seven patients with BECTS underwent resting-state fMRI scans after receiving either valproic acid (VPA) (n = 15), levetiracetam (LEV) (n = 21), or no medication (n = 21). fMRI regional homogeneity (ReHo) parameter among the three groups of patients were compared and were correlated with total doses of AED in the two medicated groups. Compared with patients on no-medication, patients receiving either VPA or LEV showed decreased ReHo in the central-temporal region, frontal cortex, and thalamus. In particular, the VPA group showed greater ReHo decrease in the thalamus and milder in cortices and caudate heads compared with the LEV group. In addition, the VPA group demonstrated a negative correlation between ReHo values in the central-temporal region and medication dose. Both VPA and LEV inhibit resting-state neural activity in the central-temporal region, which is the main epileptogenic focus of BECTS. VPA reduced brain activity in the cortical epileptogenic regions and thalamus evenly, whereas LEV reduced brain activity predominantly in the cortices. Interestingly, VPA showed a cumulative effect on inhibiting brain activity in the epileptogenic regions in BECTS. • Regional differences in brain activity in response to different AEDs in BECTS. • AEDs inhibit resting-state neural activity in epileptogenic and subcortical regions in BECTS. • Valproic acid effect on the cortical epileptogenic regions and thalamus evenly. • Levetiracetam effect seen predominantly in cortices. • Valproic acid has a cumulative effect on inhibiting brain activity in epileptogenic regions.

Considered the high dynamic and complexity of air combat and the limitations of existing technologies, inspired by knowledge structure of human and the biological neural mechanism of learning, this paper designs a brain like air to air combat system with autonomous learning ability. Firstly, the air combat confrontation ability was divided into the ability of action value assessment which belongs to the category of declarative knowledge and the decision-making ability which belongs to the category of procedural knowledge. The system structure of alternate learning and the sub-regional learning principle using error as learning motivation was designed with reference to the biological neural mechanism of learning process. The simulation experiment proves that the learning system designed in this paper can achieve self-confrontation through self-learning without human prior knowledge and calibrated data and can learn the maneuver tactics similar to the human pilot’s classical maneuver without human guidance.

This study presents a robust heart rate monitoring algorithm using photoplethysmography (PPG) signal during physical exercise. The proposed method combines two stage: motion artifact removal and frequency refinement. The cascaded normalized least mean square adaptive filter is used to attenuate the noise introduced by motion artifacts in the PPG signal. A phase vocoder technique is used to refine the frequency calculated by Fourier Transform, from which the heart rate is finally tracked. On a publicly available database of twelve PPG recordings, the proposed technique obtains an average absolute error (AAE) of 1.08 beat per minute (BPM). Person correlation coefficient of 0.997 is achieved between true heart rate and estimated heart rate. In contrast to other available approaches, the proposed method has merely one parameter to tune in spectral peak tracking step for heart rate estimation.

Single satellite passive location has many advantages and practical applications, but it also has limitations such as the positioning is not accurate and the sample points of track are relatively sparse. Knowing the shortcomings of such single satellite positioning measurements, this paper gives the simulation of a three-point passive location method based on joint Time Difference of Arrival (TDOA) and Frequency Difference of Arrival (FDOA) and analyzes the Geometric Dilution of Precision (GDOP) performances based on positioning error. The simulation results show that this method is able to calculate the position of radiation source with TDOA and FDOA measured at three points. GDOP performances show the relative velocity and position between radiation source and satellite can affect positioning accuracy in the range of TDOA and FDOA estimation error.

Purpose. This study was aimed at evaluating the motor cortical excitability and connectivity underlying the neural mechanism of motor deficit in acute stroke by the combination of functional magnetic resonance imaging (fMRI) and electrophysiological measures. Methods. Twenty-five patients with motor deficit after acute ischemic stroke were involved. General linear model and dynamic causal model analyses were applied to fMRI data for detecting motor-related activation and effective connectivity of the motor cortices. Motor cortical excitability was determined as a resting motor threshold (RMT) of motor evoked potential detected by transcranial magnetic stimulation (TMS). fMRI results were correlated with cortical excitability and upper extremity Fugl-Meyer assessment scores, respectively. Results. Greater fMRI activation likelihood and motor cortical excitability in the ipsilesional primary motor area (M1) region were associated with better motor performance. During hand movements, the inhibitory connectivity from the contralesional to the ipsilesional M1 was correlated with the degree of motor impairment. Furthermore, ipsilesional motor cortex excitability was correlated with an enhancement of promoting connectivity in ipsilesional M1 or a reduction of interhemispheric inhibition in contralesional M1. Conclusions. The study suggested that a dysfunction of the ipsilesional M1 and abnormal interhemispheric interactions might underlie the motor disability in acute ischemic stroke. Modifying the excitability of the motor cortex and correcting the abnormal motor network connectivity associated with the motor deficit might be the therapeutic target in early neurorehabilitation for stroke patients.

BIM is regarded as a new problem-solving method in the field of architecture construction. With the continuous development and progress of BIM, it also has a series of impacts on the field of project cost. In this paper, through literature research, expert interviews, questionnaires, data analysis, and other ways, the current front-line AEC (architecture, Engineering & Construction) practitioners were investigated. This paper investigates the understanding of BIM in the field of construction cost and the current development obstacles and makes a cross-analysis of the development obstacles of various industries. It is found that technology is the most important obstacle in technology, economy, society and other factors, and how to solve the problem of model standardization and accuracy is the most important part. In the cross research of technical problems and professions, we found that there are many deficiencies in the modeling function of current BIM software, and there are also some problems in the transformation of two-dimensional and three-dimensional data, which are considered as obstacles of the whole industry. The research results of this paper can be used as a practical reference for future research, trying to provide technical and management solutions for the implementation and development of BIM in the construction cost of our country.

With the national low-altitude airspace open, small UAVs are widely used in urban space. But there are dense obstacles and complex electromagnetic waves, producing serious challenges for UAVs' safety. In the region UAV loses track, it should have the ability to fly automatically and to escape from this region. By imitating the human-beings' escaping thoughts, five flying principles(exploring principle, optimal path principle, dead zone escaping principle, neighbor gather principle and roominess tending principle) are proposed based on the combination of breadth first algorithm, exploring random tree structure, so a humanoid algorithm is constructed on the basis of the five principles. The contrastive simulation results have verified that the proposed method in this paper can not only avoid the problem in traditional algorithms that obstacles simplifying is not applicable to dense environment, but also greatly boost exploring efficiency by applying exploring random tree structure to breadth first algorithm.

Collision avoidance is becoming a severe safe challenge for the small UAVs flying among urban buildings. For flight in an unknown urban area, real-time safe route planning method is needed urgently. This paper just focuses on this problem. Based on the basic principle of Laguerre diagram, the Laguerre route is firstly induced for a single prepared flight aisle between two buildings. And then the safest Laguerre route is found by analyzing the collision avoidance safe error of routes. Furthermore, a real-time safe route planning method based on Laguerre diagram can be built for the collision avoidance of quad-rotor UAV in urban areas. The presented method has been shown that its route safe property is fine by the simulation results, and it can be applied to make a quad-rotor UAV to fly through an unknown urban area.

Recent electrophysiological evidence suggests that the analysis of structure and the processing of meaning may differ across languages and across types of materials being processed. In this study we used event-related potentials (ERPs) to examine the interplay between structural analysis and meaning processing in Chinese idioms. Our results revealed that N400 effects reflect graded semantic distances in our experimental conditions involving synonyms, semantic violations, and combined semantic and syntactic violations. The P600 effects were uniform across these experimental conditions. There was no difference between the semantic only and the combined violation conditions with regard to either the N400 or the P600 component. These patterns suggest that in Chinese, unlike in other languages, meaning integration does not depend on the intactness of structural information. They also suggest, consistently with some previous studies, that P600 is not only an index of syntactic processes but may be a more general index of the processing of linguistic or perceptual well-formedness in structure in highly constraining context.

One-photon and two-photon spectroscopic properties of a novel two-photon dye doped polymethylmethacrylate (PMMA) were reported. High-density three-dimensional (3D) optical data storage was achieved by photobleaching two-photon fluorescence in the dye-doped PMMA film with data storage storage density as high as 14 Gbits/cm3 theoretically. Photobleaching decay kinetics at different excitation intensity reveal that the mechanism of two-photon photobleaching process concerns the action of three photons. (© 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

This paper designs a Hyperlipemia Information Analysis System, which can realize hyperlipemia document classification and information analysis. In document indexing, we propose an improved approach, called Term Frequency, Inverted Document Frequency and Inverted Entropy (TFIDFIE), to compute term weights in document indexing. In addition, an improved immune algorithm proposed by us is used in this system, which called Clonal Selection Algorithm Based on Antibody Density (CSABAD). According to the clonal selection principle and density control mechanism, only those cells that have higher affinity and lower density are selected to proliferate. The system obtains better classification performance. In further work, we will research the feature selection and data mining for hyperlipemia.

Medical document categorization is the process of automatically assigning one or more predefined category labels to medical documents. Document indexing plays a very important role in the process of classification. This paper proposes an improved method of computing term weights which is called tfidfie (term frequency, inverted document frequency and inverted entropy). In comparison with the tfidf (term frequency and inverted document frequency) function, the tfidfie function adds an information entropy factor, H, which represents the distribution of documents in the training set in which the term occurs. Then, we discuss the effects of training set in medical document categorization. An imbalanced training set decreases the performance of classifier. Considering the characteristics of medical documents, the medical classifiers are constructed by the methods of Naive Bayes and Rocchio respectively. The experiment results show that tfidfie improves the classification performance and Naive Bayes outperforms Rocchio.

This paper proposes a new method of text categorization called the clonal selection algorithm based on antibody density (CSABAD). In this method, antigen, B cell and antibody are respectively corresponded with training texts, a possible individual of classifier and the affinity between the individual and training texts. B cells consist of general cells, fresh cells and memory cells. General cells are used to store various individuals, fresh cells are used to replace the degraded general cells, and memory cells are used to record the best individuals. According to the clonal selection principle and density control mechanism, only those cells that have higher affinity and lower density are selected to proliferate. The ultimate classifier is composed with many memory cells. Considering the characters of medical information, we realize a medical text classifier based on CSABAD, and tests the system on OHSUMED data set. The experiment results show that it can obtain the better classification performance.

The nature of immune algorithm is to distinguish between self and non-self. On the basis of immune algorithm, we propose a new method of text categorization called the Clonal Selection Algorithm Based on Antibody Density (CSABAD). According to the clonal selection principle and density control mechanism, only those cells that have higher affinity and lower density are selected to proliferate. The ultimate classifier is composed with many memory cells. Because CSABAD is a two-class algorithm, we discuss the main approaches how to apply it to the multi-class problem, such as one-vs-one (OVO), one-vs-rest (OVR) and directed acyclic graph (DAG). Our experiments show that one-vs-one obtains the best classification performance on the 20_newsgroups dataset. In addition, the experiment results also show that it significantly outperforms Rocchio and Naive Bayes.

We propose an ultra-low power (ULP) Image Signal Processor (ISP) that performs on-the-fly in-processing frame (de)compression and hierarchical event recognition to exploit the temporal and spatial sparsity in an image sequence to achieve a 16× imaging system energy gain. The ISP is fabricated in 40 nm CMOS and consumes only 170 μW at 5 fps for neural network-based intruder detection and 192× compressed image recording.