Enhanced Temporal Correlation for Universal Lesion Detection

Universal lesion detection(ULD)methods for computed tomography(CT)images play a vital role in the modern clinical medicine and intelligent automation.It is well known that single 2D CT slices lack spatial-temporal characteristics and contextual information compared to 3D CT blocks.However,3D CT blocks necessitate significantly higher hardware resources during the learning phase.Therefore,efficiently exploiting temporal correlation and spatial-temporal features of 2D CT slices is crucial for ULD tasks.In this paper,we propose a ULD network with the enhanced temporal correlation for this purpose,named TCE-Net.The designed TCE module is applied to enrich the discriminate feature representation of multiple sequential CT slices.Besides,we employ multi-scale feature maps to facilitate the localization and detection of lesions in various sizes.Extensive experiments are conducted on the DeepLesion benchmark demonstrate that thismethod achieves 66.84%and 78.18%for FS@0.5 and FS@1.0,respectively,outperforming compared state-of-the-art methods.

Quantitative assessment of the spatio-temporal correlations of seismic events induced by longwall coal mining

Rock failure process as a natural response to mining activities is associated with seismic events, which can pose a potential hazard to mine operators, equipment and infrastructures. Mining-induced seismicity has been found to be internally correlated in both time and space domains as a result of rock fracturing during progressive mining activities. Understanding the spatio-temporal(ST) correlation of mininginduced seismic events is an essential step to use seismic data for further analysis, such as rockburst prediction and caving assessment. However, there are no established methods to perform this critical task. Input parameters used for the prediction of seismic hazards, such as the time window of past data and effective prediction distance, are determined based on site-specific experience without statistical or physical reasons to support. Therefore, the accuracy of current seismic prediction methods is largely constrained, which can only be addressed by quantitively assessing the ST correlations of mininginduced seismicity. In this research, the ST correlation of seismic event energy collected from a study mine is quantitatively analysed using various statistical methods, including autocorrelation function(ACF), semivariogram and Moran’s I analysis. In addition, based on the integrated ST correlation assessment, seismic events are further classified into seven clusters, so as to assess the correlations within individual clusters. The correlation of seismic events is found to be quantitatively assessable, and their correlations may vary throughout the mineral extraction process.

Spatiotemporal Prediction of Urban Traffics Based on Deep GNN

Traffic prediction already plays a significant role in applications like traffic planning and urban management,but it is still difficult to capture the highly non-linear and complicated spatiotemporal correlations of traffic data.As well as to fulfil both long-termand short-termprediction objectives,a better representation of the temporal dependency and global spatial correlation of traffic data is needed.In order to do this,the Spatiotemporal Graph Neural Network(S-GNN)is proposed in this research as amethod for traffic prediction.The S-GNN simultaneously accepts various traffic data as inputs and investigates the non-linear correlations between the variables.In terms of modelling,the road network is initially represented as a spatiotemporal directed graph,with the features of the samples at the time step being captured by a convolution module.In order to assign varying attention weights to various adjacent area nodes of the target node,the adjacent areas information of nodes in the road network is then aggregated using a graph network.The data is output using a fully connected layer at the end.The findings show that S-GNN can improve short-and long-term traffic prediction accuracy to a greater extent;in comparison to the control model,the RMSE of S-GNN is reduced by about 0.571 to 9.288 and the MAE(Mean Absolute Error)by about 0.314 to 7.678.The experimental results on two real datasets,Pe MSD7(M)and PEMS-BAY,also support this claim.

Lockdown during COVID-19 pandemic:A case study from Indian cities shows insignificant effects on persistent property of urban air quality

The influence of reduction in emissions on the inherent temporal characteristics of PMand NOconcentration time series in six urban cities of India is assessed by computing the Hurst exponent using Detrended Fluctuation Analysis(DFA) during the lockdown period(March 24–April 20, 2020) and the corresponding period during the previous two years(i.e., 2018 and 2019). The analysis suggests the anticipated impact of confinement on the PMand NOconcentration in urban cities, causing low concentrations. It is observed that the original PMand NOconcentration time series is persistent but filtering the time series by fitting the autoregressive process of order 1 on the actual time series and subtracting it changes the persistence property significantly. It indicates the presence of linear correlations in the PMand NOconcentrations. Hurst exponent of the PMand NOconcentration during the lockdown period and previous two years shows that the inherent temporal characteristics of the short-term air pollutant concentrations(APCs) time series do not change even after withholding the emissions. The meteorological variations also do not change over the three time periods. The finding helps in developing the prediction models for future policy decisions to improve urban air quality across cities.

Complexity and Self-Organized Criticality of Solid Earth System(Ⅰ)

The author puts forward the proposition of Complexity and Self Organized Criticality of Solid Earth System in the light of: (1) the science of complexity studies the mechanisms of emergence of complexity and is the science of the 21st century, (2) the study of complexity of the earth system would be one of the growing points occupying a strategic position in the development of geosciences in the 21st century. By the proposition we try to cogitate from a new viewpoint the ancient yet ever new solid earth system. The author abstracts the fundamental problem of the solid earth system from the essence of the generalized geological systems and processes which reads: the complexity and self organized criticality of the global nature, structure and dynamical behavior of the whole solid earth system emerging from the multiple coupling and superposition of non linear interactions among the multicomponents of the earths material and the multiple generalized geological (geological, geophysical, and geochemical) processes . Starting from this cognizance the author proposes eight major themes and the methodology of researches on the complexity and self organized criticality of the solid earth system.

Information Driven Data Gathering for Energy Efficient Wireless Sensor Network

Large scale dense Wireless Sensor Networks (WSNs) have been progressively employed for different classes of applications for the resolve of precise monitoring. As a result of high density of nodes, both spatially and temporally correlated information can be detected by several nodes. Hence, energy can be saved which is a major aspect of these networks. Moreover, by using these advantages of correlations, communication and data exchange can be reduced. In this paper, a novel algorithm that selects the data based on their contextual importance is proposed. The data, which are contextually important, are only transmitted to the upper layer and the remains are ignored. In this way, the proposed method achieves significant data reduction and in turn improves the energy conservation of data gathering.

Temporal sequential correlation analysis on compoundaction potentials of the auditory nerve versus the stimulating AM-sound envelope

Compound action potentials of the auditory nerve in response to amplitude modulating tones were recorded in guinea pigs with electrode implanted to the exit of the internal auditory meatus and temporal sequential correlation between the responses and the modulators was studied in a paradigm of systematically changing acoustic parameters. Three kinds of modulators were used. continuous or burst sinusoids of fixed frequency (in the range of 40 Hz-5 kHz), short bursts of sinusoids with changing frequency and short segments of speech signal. Ranges of parametric variation were 500 Hz-20 kHz for carrier frequency, 5%-95% for modulation depth and 20 dB-90 dB SPL for intensity. For continuous or burst sinusoidal modulators of fixed frequencies, the correlation coefficient (r) remained quite high in most parametric conditions, ranging from 0.80 to 0.95. It became smaller mainly in instances of decreased response amplitude on account of unfavourable parameters. For burst modulators of changing frequency, r varied around 0.66-0.86. When segments of speech signal served as the modulators, significant correlation (r around 0.50 ) also existed, indicating the validity of the timing mode of information encoding for speech sound at the cochlear nerve level. Some theoretical and technical points in studying the timing mechanism of audition is discussed.

Self-consistent tomography of temporally correlated errors

The error model of a quantum computer is essential for optimizing quantum algorithms to minimize the impact of errors using quantum error correction or error mitigation.Noise with temporal correlations,e.g.low-frequency noise and context-dependent noise,is common in quantum computation devices and sometimes even significant.However,conventional tomography methods have not been developed for obtaining an error model describing temporal correlations.In this paper,we propose self-consistent tomography protocols to obtain a model of temporally correlated errors,and we demonstrate that our protocols are efficient for low-frequency noise and context-dependent noise.

Physics-guided Deep Learning for Power System State Estimation

In the past decade,dramatic progress has been made in the field of machine learning.This paper explores the possibility of applying deep learning in power system state estimation.Traditionally,physics-based models are used including weighted least square(WLS)or weighted least absolute value(WLAV).These models typically consider a single snapshot of the system without capturing temporal correlations of system states.In this paper,a physics-guided deep learning(PGDL)method is proposed.Specifically,inspired by autoencoders,deep neural networks(DNNs)are used to learn the temporal correlations.The estimated system states from DNNs are then checked against physics laws by running through a set of power flow equations.Hence,the proposed PGDL is both data-driven and physics-guided.The accuracy and robustness of the proposed PGDL method are compared with traditional methods in standard IEEE cases.Simulations show promising results and the applicability is further discussed.