Time-varying Data Research Articles

Deep learning-based cell identification and disease diagnosis using spatio-temporal cellular dynamics in compact digital holographic microscopy

We demonstrate a successful deep learning strategy for cell identification and disease diagnosis using spatio-temporal cell information recorded by a digital holographic microscopy system. Shearing digital holographic microscopy is employed using a low-cost, compact, field-portable and 3D-printed microscopy system to record video-rate data of live biological cells with nanometer sensitivity in terms of axial membrane fluctuations, then features are extracted from the reconstructed phase profiles of segmented cells at each time instance for classification. The time-varying data of each extracted feature is input into a recurrent bi-directional long short-term memory (Bi-LSTM) network which learns to classify cells based on their time-varying behavior. Our approach is presented for cell identification between the morphologically similar cases of cow and horse red blood cells. Furthermore, the proposed deep learning strategy is demonstrated as having improved performance over conventional machine learning approaches on a clinically relevant dataset of human red blood cells from healthy individuals and those with sickle cell disease. The results are presented at both the cell and patient levels. To the best of our knowledge, this is the first report of deep learning for spatio-temporal-based cell identification and disease detection using a digital holographic microscopy system.

The natural resources potential assessing aimed at territorial planning using time-varying space data of vegetation index and LST.

This research presents the methodology of mapping a situation of the natural-landscape resources of Khuzestan province in Iran using time-varying space data according to seasonal changes of the indicators of SAVI and LST. The novelty of this research is the combined use of GIS techniques, optical and thermal remote sensing for mapping the situation of the natural-landscape resources aimed at territorial planning. In this research, a classification was carried out for the study area based on the potential of the natural-landscape resources. The Landsat 7 and 8 materials from the periods of 2011-2013 and 2017-2019 were used to perform this research using ERDAS Imagine, ENVI, ArcGIS, and Priznak software. Variations of LST and SAVI were analyzed for 10 seasons of year from April to September. The results showed that both increasing the area of bare soils (wasteland) and residential areas and decreasing the extent of the water resources, forests, and rangeland in Khuzestan province have caused the surface temperature to rise over a period of 9years. Whereas, matching of the Entisols and Inceptisols with areas having high natural resource potential in Khuzestan province was indicated by the results of this study. And also, the Badlands, Rock-outcrops, and Dunelands are the areas with low natural resource potential. It has been shown that integral indicator of the situation of the natural-landscape resources that calculated using the materials of multispectral space imagery is a stable indicator and criterion for land assessing in terms of their suitability for agricultural use. The results of this study can be used to provide a territorial management and planning in accordance with the goals of land use planning and sustainable development, by potential assessment and mapping a status of the natural-landscape resources.

Evaluating environmental performance using data envelopment analysis: The case of European countries

Environmental efficiency is considered to be a critical pillar of sustainable development; consequently, improving environmental performance has aroused attention at both the local and global level. This study evaluated the environmental performance of European Union (EU) countries using the data envelopment analysis (DEA) approach and the global Malmquist-Luenberger index. Although literature using the DEA approach to measure environmental performance exists, this study’s main contribution was considering different types of undesirable outputs and using long-term panel data on EU countries. In particular, the DEA window analysis technique was applied to evaluate the environmental performance of 27 EU countries in cross-sectional and time-varying data during the period 2000–2017. Three DEA models were examined; labor, capital, and energy were used as common inputs with a combination of different outputs, including gross domestic product, carbon dioxide and particulate matter emissions, and waste. The empirical results revealed that the trends in the environmental performance of the entire EU and its individual countries were similar under all examined models. Environmental performance was indeed negatively affected by the financial crisis of 2007–2008; this impact was mainly observed in eastern EU countries. Furthermore, both economic and environmental variables significantly influenced countries’ overall efficiency. The analysis using the global Malmquist-Luenberger index elucidated that overall, EU countries experienced the efficiency improvement during the study period, although fluctuations were observed. These results enable countries and policymakers to understand their environmental performance, identify strengths and weaknesses, and set targets for further improvement based on the current best practices of comparable peers.

Haery: A Hadoop Based Query System on Accumulative and High-Dimensional Data Model for Big Data

Column-oriented stores, known for their scalability and flexibility, are a common NoSQL database implementation and are increasingly used in big data management. In column-oriented stores, a “full-scan” query strategy is inefficient and the search space can be reduced if data is well partitioned or indexed; however, there is no pre-defined schema for building and maintaining partitions and indexes at lower cost. We leverage an accumulative and high-dimensional data model, a sophisticated linearization algorithm, and an efficient query algorithm, to solve the challenge of how a pre-defined and well-partitioned data model can be applied to flexible and time-varied key-value data. We adapt a high-dimensional array as the data model to partition the key-value data without additional storage and massive calculation; improve the Z-order linearization algorithm, which map multidimensional data to one dimension while preserving locality of the data points, for flexibility; efficiently build an expansion mechanism for the data model to support time-varied data. The result is Haery, a column-oriented store, based on a distributed file system and computing framework. In experiments, Haery is compared with Hive, HBase, Cassandra, MongoDB, PostgresXL, and HyperDex in terms of query performance. With results indicating Haery on average performs 4.57x, 4.23x, 3.55x, 1.79x, 1.82x, and 120.6x faster, respectively.

Materialized View Selection Using Self-Adaptive Perturbation Operator-Based Particle Swarm Optimization

A data warehouse is a central repository of time-variant and non-volatile data integrated from disparate data sources with the purpose of transforming data to information to support data analysis. Decision support applications access data warehouses to derive information using online analytical processing. The response time of analytical queries against speedily growing size of the data warehouse is substantially large. View materialization is an effective approach to decrease the response time for analytical queries and expedite the decision-making process in relational implementations of data warehouses. Selecting a suitable subset of views that deceases the response time of analytical queries and also fit within available storage space for materialization is a crucial research concern in the context of a data warehouse design. This problem, referred to as view selection, is shown to be NP-Hard. Swarm intelligence have been widely and successfully used to solve such problems. In this paper, a discrete variant of particle swarm optimization algorithm, i.e. self-adaptive perturbation operator based particle swarm optimization (SPOPSO), has been adapted to solve the view selection problem. Accordingly, SPOPSO-based view selection algorithm (SPOPSOVSA) is proposed. SPOPSOVSA selects the Top-K views in a multidimensional lattice framework. Further, the proposed algorithm is shown to perform better than the view selection algorithm HRUA.

2019 IEEE Scientific Visualization Contest Winner: Visual Analysis of Structure Formation in Cosmic Evolution.

Simulations of cosmic evolution are a means to explain the formation of the universe as we see it today. The resulting data of such simulations comprise numerous physical quantities, which turns their analysis into a complex task. Here, we analyze such high-dimensional and time-varying particle data using various visualization techniques from the fields of particle visualization, flow visualization, volume visualization, and information visualization. Our approach employs specialized filters to extract and highlight the development of so-called active galactic nuclei and filament structures formed by the particles. Additionally, we calculate X-ray emission of the evolving structures in a preprocessing step to complement visual analysis. Our approach is integrated into a single visual analytics framework to allow for analysis of star formation at interactive frame rates. Finally, we lay out the methodological aspects of our work that led to success at the 2019 IEEE SciVis Contest.

Time-varying volume visualization: a survey

Time-varying volume data is often generated from scientific simulations in a variety of application domains, such as computational fluid dynamics, combustion science, and computational cosmology. Data visualization plays an important role in analyzing the dynamics and evolution of phenomena hidden in the data. Over the last two decades, a substantial amount of visualization techniques have been proposed in this research area. In this paper, we systematically review the recent literature on data visualization and visual analytics for time-varying scalar volume data. We first collect a corpus of relevant technical and application papers in visualization journals and conferences from 2008 to 2019. Based on this corpus, we classify these techniques into three aspects, including feature tracking, evolution visualization, and rendering, and then detaily describe relevant techniques in these three aspects. Finally, we conclude this survey with emerging trends and future challenges in time-varying volume visualization.

Sequential-Mining-Based Vulnerable Branches Identification for the Transmission Network Under Continuous Load Redistribution Attacks

Load redistribution (LR) attacks represent a major cyber threat to the power grid that can result in branch overloads and even cascading failures in the transmission network by injecting false data to load bus power injection and line power flow measurements. While the mechanism and defense strategies have been well studied for single-occurrence (i.e., discrete) LR attacks, this paper focuses on a yet unexplored category of LR attacks, namely, the continuous LR (CLR) attack, which continuously injects time-varying false data to one or more bus and branch measurements within the system for a certain period of time. To cope with the potentially ever-changing targeted branches (i.e., the branches under attack) and system vulnerability features caused by CLR attacks, this paper takes a different route from the traditional LR defense and mitigation approaches. We propose a novel correlation chain model to capture the vulnerability correlations among branches under the CLR attack. Then, a dynamic sequential-mining algorithm (DSMA) with top-k structure is proposed based on the specific nature of the CLR attack, to mine the historical SCADA measurement database, extract vulnerable branch sequences, and identify the most suitable branches to defend preventively. Simulation studies are conducted on IEEE RTS-96 system to evaluate the effectiveness of the proposed approach in terms of increasing attack costs for the attackers and reducing the potential impacts of CLR attacks.

Green resource allocation and energy management in heterogeneous small cell networks powered by hybrid energy

In heterogeneous networks (HetNets), how to improve spectrum efficiency is a crucial issue. Meanwhile increased energy consumption inspires network operators to deploy renewable energy sources as assistance to traditional electricity. Based on above aspects, we allow base stations (BSs) to share their licensed spectrum resource with each other and adjust transmission power to adapt to the renewable energy level. Considering the sharing fairness among BSs, we formulate a multi-person bargaining problem as a stochastic optimization problem. We divide the optimization problem into three parts: data rate control, resource allocation and energy management. An online dynamic control algorithm is proposed to control admission rate and resource allocation to maximize the transmission and sharing profits with the least grid energy consumption. Simulation results investigate the time-varying data control and energy management of BSs and demonstrate the effectiveness of the proposed scheme.

An Adaptive Data Placement Architecture in Multicloud Environments

Cloud service providers (CSPs) can offer infinite storage space with cheaper maintenance cost compared to the traditional storage mode. Users tend to store their data in geographical and diverse CSPs so as to avoid vendor lock-in. Static data placement has been widely studied in recent works. However, the data access pattern is often time-varying and users may pay more cost if static placement is adopted during the data lifetime. Therefore, it is a pending problem and challenge of how to dynamically store users’ data under time-varying data access pattern. To this end, we propose ADPA, an adaptive data placement architecture that can adjust the data placement scheme based on the time-varying data access pattern and subject for minimizing the total cost and maximizing the data availability. The proposed architecture includes two main components: data retrieval frequency prediction module based on LSTM and data placement optimization module based on Q-learning. The performance of ADPA is evaluated through several experimental scenarios using NASA-HTTP workload and cloud providers information.

Bayesian estimation of spatially varying soil parameters with spatiotemporal monitoring data

The characterization of in situ ground conditions is essential for geotechnical practice. The probabilistic estimation of soil parameters can be achieved via updating with monitoring data within the Bayesian framework. The estimation of spatially varying soil parameters is seldom undertaken with time-variant monitoring data. In this study, an efficient Bayesian method is presented for the estimation of spatially varied saturated hydraulic conductivity ks of unsaturated soil slope with spatiotemporal monitoring data. The computationally cheap surrogate model of the adaptive sparse polynomial chaos expansion method is adopted to approximate the transient numerical model. Markov chain Monte Carlo method is used for the probabilistic estimation of basic random variables. Based on the hypothetical cases, the effects of monitoring frequency and stage are studied. The errors and the uncertainties of the estimated ks fields are increased with the decreasing monitoring frequency. Bayesian estimation of spatial variability is more accurate when using the later stage of monitoring data. The estimated method is further verified with a real case study by the comparison of borehole data, dynamic probe test (DPT) data, and field monitoring data. The distribution of the soil types acquired from boreholes is reflected in the estimated ks. The estimated field of ks has a certain agreement with the borehole log and DPT measurements and can reproduce the spatial variability of the site to an acceptable degree.

265-OR: Identifying Adults at Risk of Unintentional Severe Hypoglycemia in Hospital Using Artificial Intelligence (RUSHH-AI)

Background: Machine learning carries great promise to improve healthcare delivery. Clinical outcomes that are routinely and objectively measured, and have serious consequences that can be prevented, are ideal targets for prediction and intervention. Hypoglycemia, defined as a blood glucose less than 3.9 mmol/L (70 mg/dL), meets these criteria. The purpose of this study was to predict hypoglycemia using artificial intelligence models in patients hospitalized to general internal medicine (GIM) and cardiovascular surgery (CV) at a tertiary-care teaching hospital in Toronto, Ontario. Methods: Models were built using routinely-collected clinical data from the hospital’s electronic health record. Models were trained using data from Jan 2013-Apr 2017, tested using data from Apr 2017-Mar 2018, and validated using held-out test data from Apr 2018-Mar 2019. Three models were generated using supervised machine learning: LASSO regression, gradient boosted trees, and a recurrent neural network. Each model included baseline patient data and time-varying data. Natural language processing was used to incorporate text data from physician and nursing notes. Results: We included 8492 GIM admissions and 8044 CV admissions. The average age of patients was 68 years, 35% were women, the baseline creatinine was 90 μmol/L (1.0mg/dL) and the baseline A1C was 7%. Hypoglycemia occurred in 15% of GIM admissions and 13% of CV admissions. The area under the curve for the model in the held-out validation set was approximately 0.80 on the GIM ward and 0.82 on the CV ward. When the threshold for hypoglycemia was lowered to 2.9 mmol/L (52 mg/dL), similar results were observed. Among the patients at the highest decile of risk, the positive predictive value was approximately 50% and the sensitivity was 99%. Conclusion: Using natural language processing and machine learning we were able to accurately identify patients at high risk of hypoglycemia in hospital. Disclosure M. Fralick: None. D. Dai: None. C. Pou-Prom: None. A.A. Verma: None. M. Mamdani: None.

Dissociable mechanisms govern when and how strongly reward attributes affect decisions.

Theories and computational models of decision-making usually focus on how strongly different attributes are weighted in choice, for example, as a function of their importance or salience to the decision-maker. However, when different attributes affect the decision process is a question that has received far less attention. Here, we investigated whether the timing of attribute consideration has a unique influence on decision-making by using a time-varying drift diffusion model and data from four separate experiments. Experimental manipulations of attention and neural activity demonstrated that we can dissociate the processes that determine the relative weighting strength and timing of attribute consideration. Thus, the processes determining either the weighting strengths or the timing of attributes in decision-making can independently adapt to changes in the environment or goals. Quantifying these separate influences of timing and weighting on choice improves our understanding and predictions of individual differences in decision behaviour.

Survival analysis of failures based on Hawkes process with Weibull base intensity

In this paper, we construct a Hawkes process with time-varying base intensity to model the sequence of failure, i.e., failure events of the compressor station, and we combine survival analysis and point process model on various failure events of the compressor station based on Hawkes process. To our best knowledge, until now, nearly all relevant literature of the Hawkes point processes assumes that the base intensity of the conditional intensity function is time-invariant. This assumption is apparently too harsh to be verified. For example, in the practical application, including financial analysis, reliability analysis, survival analysis and social network analysis, the truth variation of the base intensity of the failure occurrence over time is not constant. The constant base intensity will not reflect the base intensity trend of the failure occurring over time. Thus, in order to solve this problem, in this paper, we propose a new time-varying base intensity, e.g. which is treated as obeying Weibull distribution. First, we introduce the base intensity into a Hawkes process that obeys the Weibull distribution, and then we propose an effective learning algorithm based on the maximum likelihood estimator. Experiments on the constant base intensity synthetic data, time-varying base intensity synthetic data, and real-world data show that our method can learn the triggering patterns of the Hawkes processes and the time-varying base intensity simultaneously and robustly. Experiments on real-world data also reveal the Granger causality of different types of failures and the base probability of failure varying over time. We put forward some suggestions for practical production based on the experimental results.

Unsupervised machine learning via Hidden Markov Models for accurate clustering of plant stress levels based on imaged chlorophyll fluorescence profiles & their rate of change in time

As far back as 1931, studies have shown that Chlorophyll a fluorescence (ChlF) is a useful tool in plant stress detection. Early and accurate detection of plant stress is invaluable in enabling appropriate and timely intervention. One of the major limitations of past work on ChIF-based plant stress identification is that, often, only a small number of commonly used inflection points, locally oriented within the ChlF transient curve, are utilized to calculate a single index value. These singular values offer limited insight into stressor level or stressor type. In this work, we present an unsupervised method for plant stress classification (identification) that utilizes global (versus local) time-varying ChlF signal data obtained via plant video imaging. The contributions of this work are multi-fold: (a) We classify via clustering these time-varying-intensity signals using unsupervised learning via Hidden Markov Models (HMMs). We show how: (b) the proposed globally based feature selection of a plant’s entire ChlF signal profile, via low-pass filtering, can, in some scenarios, improve classification accuracy of plant stress; (c) the rate-of-change-in-time of the plant’s ChlF intensity time-varying profile, as an additional global feature selection, can further improve the plant stress classification accuracy in some scenarios; (d) quantification can improve the proposed HMM classification method in certain scenarios; (e) HMMs allow more variability in categorizing data via clustering than other raw data distance based metrics. (f) We explore the ergodic and Bakis models for HMM state transition matrix initialization. (g) In addition we propose a new method for initialization of the HMM state transition matrix: state information based initial probability assignment (SIPA) and compare it to the often used heuristic initial probability assignment (HIPA) method. (h) We show how using the Bayesian Information Criterion (BIC) as a performance metric allows a good state number selection (using the quantified data with the proposed state transition matrix initialization methods). (i) We present a method to automatically determine the number of existing classes in the data set. Note this is in contrast to existing unsupervised learning methods, where the number of existing classes in the data set is, in general, guessed a priori, and often incorrectly. (j) In conclusion we show experimental results to illustrate the potential and value of the proposed methods to enable more specific and accurate classification of plant stressor levels in some scenarios, and (k) in some instances, stressor types.

Non-stationary Gaussian process regression applied in validation of vehicle dynamics models

This work compares methods to compute confidence bands in a validation task of a vehicle single-track model. The confidence bands are computed from time series by naïve method, Gaussian process regression and heteroscedastic and non-stationary Gaussian process regression. The simulation model considers the epistemic uncertainty of the vehicle mass parameter by Latin hypercube sampling. The validation procedure compares all stochastically simulated time series of the vehicle yaw rate with the confidence band of the reference data. The model is marked as valid if the yaw rate for each time step is within the confidence band of the reference data. The data was challenging due to noise and time-varying variance and smoothness. Due to required data pre-processing and the high sensitivity to noise in the reference data, the naïve method has generated unusable confidence bands and cannot be recommended for similar validation tasks. Gaussian process regression solved the problem of noise sensitivity, but was not able to model the time-varying length scale of the reference data. Therefore, heteroscedastic and non-stationary Gaussian process regression is proposed to calculate accurate confidence bands of time-varying and noisy reference data for the validation of dynamic models by a confidence band approach.

Evaluation of the profiles of CB1 cannabinoid receptor signalling bias using joint kinetic modelling.

Biased agonism describes the ability of ligands to differentially regulate multiple signalling pathways when coupled to a single receptor. Signalling is affected by rapid agonist-induced receptor internalisation. Hence, the conventional use of equilibrium models may not be optimal, because (i) receptor numbers vary with time and, in addition, (ii) some pathways may show non-monotonic profiles over time. Data were available from internalisation, cAMP inhibition and phosphorylation of ERK (pERK) of the cannabinoid-1 (CB1 ) receptor using a concentration series of six CB1 ligands (CP55,940, WIN55,212-2, anandamide, 2-arachidonylglycerol, Δ9 -tetrahydrocannabinol and BAY59,3074). The joint kinetic model of CB1 signalling was developed to simultaneously describe the time-dependent activities in three signalling pathways. Based on the insights from the kinetic model, fingerprint profiles of CB1 ligand bias were constructed and visualised. A joint kinetic model was able to capture the signalling profiles across all pathways for the CB1 receptor simultaneously for a system that was not at equilibrium. WIN55,212-2 had a similar pattern as 2-arachidonylglycerol (reference). The other agonists displayed bias towards internalisation compared to cAMP inhibition. However, only Δ9 -tetrahydrocannabinol and BAY59,3074 demonstrated bias in the pERK-cAMP pathway comparison. Furthermore, all the agonists exhibited little preference between internalisation and pERK. This is the first joint kinetic assessment of biased agonism at a GPCR (e.g. CB1 receptor) under non-equilibrium conditions. Kinetic modelling is a natural method to handle time-varying data when traditional equilibria are not present and enables quantification of ligand bias.

IT2-GSETSK: An evolving interval Type-II TSK fuzzy neural system for online modeling of noisy data

As a core part of a fuzzy neural system, the rule base antecedents and consequents may carry uncertainties because they are trained using noisy data. So, handling the uncertain rule base is an important need in some specific problems such as noisy non-dynamic problems which leads a better data modeling. As a solution, Interval Type-II (IT2) version of GSETSK (Generic Self-Evolving Takagi-Sugeno-Kang), namely IT2-GSETSK, is presented in this paper. This solution uses IT2 membership functions for handling uncertainties, plus having Type-I (GSETSK) capabilities. In this way IT2-GSETSK is a fully-online model able to handle data streams and cope with time-variant data. It also provides up-to-date, relevant and compact rule base that is easily interpretable by human. The IT2-GSETSK is tested over several applications including medical, environmental and financial predictions, which show satisfactory performance of IT2-GSETSK. Moreover, it is observed that while GSETSK performs well enough for dynamic problems with less noise, noisy non-dynamic problems benefit significantly from IT2-GSETSK.

Global Glacial Isostatic Adjustment Constrained by GPS Measurements: Spherical Harmonic Analyses of Uplifts and Geopotential Variations

In addition to studies of sea level change and mantle rheology, reliable Glacial Isostatic Adjustment (GIA) models are necessary as a background model to correct the widely used Gravity Recovery and Climate Experiment (GRACE) monthly gravity solutions to determine subsecular, nonviscous variations. Based on spherical harmonic analyses, we developed a method using degree-dependent weighting to assimilate the Global Positioning System (GPS) derived crustal uplift rates into GIA model predictions, in which the good global pattern of GIA model predictions and better local resolution of GPS solutions are both retained. Some systematic errors in global GPS uplift rates were also corrected during the spherical harmonic analyses. Further, we used the refined GIA uplift rates to infer the GIA-induced rates of Stokes coefficients (complete to degree/order 120) relying on the accurate relationship between GIA vertical surface deformation and gravitational potential changes. The results show notable improvements relative to GIA model outputs, and may serve as a GIA-correction model for GRACE time-variable gravity data.

New Results on Joint Channel and Impulsive Noise Estimation and Tracking in Underwater Acoustic OFDM Systems

Impulsive noise can greatly affect the performance of underwater acoustic (UA) orthogonal frequency-division multiplexing (OFDM) systems. In this paper, by utilizing the sparsity of the UA channel impulse response and impulsive noise, we first propose a novel sparse Bayesian learning (SBL) based expectation maximization (EM) algorithm for joint channel estimation and impulsive noise mitigation in UA OFDM systems. Secondly, considering that the UA channel and impulsive noise are fast time-varying, we develop a new approach which combines the SBL with the forward-backward Kalman filtering to track the UA channel and impulsive noise. To further improve the system performance, we utilize the information available on data subcarriers for joint time-varying channel estimation and data detection, based on the SBL algorithm and the Kalman filter. The performance of our proposed algorithms is verified through both numerical simulations and by data collected during a UA communication experiment conducted in the estuary of the Swan River, Perth, Australia. The results demonstrate that compared with existing approaches, the proposed algorithms achieve a better system bit-error-rate and frame-error-rate performance.