Vertical Slices Research Articles

Fast History Matching and Robust Optimization Using a Novel Physics-Based Data-Driven Flow Network Model: An Application to a Steamflood Sector Model

SummaryFull-fidelity models can be computationally expensive during history matching (HM) and robust optimization as these problems typically require hundreds of simulations. Previously, we have implemented a physics-based data-driven flow network model, general purpose simulator-powered network model (GPSNet), that serves as a surrogate without the need to build the 3D full-fidelity model. In this paper, GPSNet is enhanced to GPSNet-2D to better capture thermal processes, especially gravity drainage. This enhancement successfully enables a rapid HM and robust optimization for a steamflood sector model.In GPSNet-2D, the reservoir is discretized into a series of 2D connections (x-z planes) between well completions. These connections capture the main areal/vertical flow paths, and the grid properties of each connection are interpreted with historical data. Steam segregation and heat loss are included to better represent the subsurface physics. When simulating GPSNet-2D using a commercial simulator, an equivalent 3D Cartesian model is designed where vertical slices correspond to the interwell connection planes. Thereafter, an iterative HM is conducted using the ensemble smoother with multiple data assimilation. The best-matched model is then used for steam injector control optimization.The GPSNet-2D model is first validated through a synthetic steamflood case. HM result shows that the GPSNet-2D model not only aligns closely to field-scale volumetric data but also yields good well-level matches, including bottomhole pressure (BHP)/bottomhole temperature (BHT) and phase rates. Then, GPSNet-2D is successfully applied to steamflood HM and robust optimization for a 36-well sector of a heavy oil field in the San Joaquin Valley in North America. Again, the calibrated GPSNet-2D model demonstrates its capability and reliability by generating accurate field-scale match results and reasonable matches for most of the wells. For robust well control optimization, we select nine realizations from the posterior ensemble and maximize NPV under well constraints simultaneously. The optimal case improves the average NPV by 27% over the reference case. The integration with a commercial simulator in GPSNet-2D provides flexibility to account for complex physics in the thermal recovery processes, such as steam segregation near injectors, fast steam breakthroughs at producers, and heat loss to the overburden/underburden formations. Unlike traditional simulation that relies on a detailed characterization of geological models, the GPSNet-2D model only requires well volumetric production/injection data and its approximate trajectory and can be generated and updated rapidly. In addition, GPSNet-2D also runs much faster (minutes) than a full-fidelity thermal model due to having much fewer gridblocks in the model.To our knowledge, this is the first time a physics-based data-driven network model integrated with a commercial simulator is demonstrated via a field steamflood case. Unlike approaches developed with analytic/empirical solutions or research simulators, the use of a commercial simulator makes it possible to extend flow network modeling to simulate enhanced oil recovery processes more realistically. It serves as an ideal surrogate model for both fast and reliable decision-making in reservoir management.

Novel Three-Dimensional Sarma Method with Vertical Slices for Stability Analysis of Rock Slopes

Novel Three-Dimensional Sarma Method with Vertical Slices for Stability Analysis of Rock Slopes

QUANTITATIVE VARIABILITY OF BURROW PERCENTAGE ESTIMATED FROM 2D VIEWS: EXAMPLE FROM THALASSINOIDES-BEARING STRATA, CENTRAL SAUDI ARABIA

ABSTRACT This study examines computed tomography (CT) scans of a 15.24-cm diameter and 17.1-cm length core cut from Thalassinoides-bearing strata (Hanifa Formation, central Saudi Arabia) to explore the quantitative variability in burrow percentage when estimated using 2D views. A cropped quadrangular prism of this core allowed 202 2D slices in two vertical orientations and 165 horizontal 2D slices. Thalassinoides volume represents 20.0% (burrow percentage) of the total volume within this cropped quadrangular prism. The estimated burrow percentage by the 2D vertical slices varies from 11.8% to 30.3% with a mean value of 19.9% ± 3.3% and 19.6% ± 3.9%, and coefficients of variation of 17% and 20%, for the two vertical orientations. A wider range and slightly more variability are found when estimated burrow percentage using the horizontal slices (burrow percentage range from 6.1% to 33.1% with a mean value of 19.8% ± 6.5% and coefficient of variation of 33.1%). However, analysis of variance results indicated no statistically significant difference between the distribution of the burrow percentage among vertical and horizontal slices. A test of how randomly selected slices of the CT scan would represent the burrow percentage indicated that even five randomly selected slices could retain a mean comparable to the actual burrow percentage of the CT scan. Based on these results, we suggested a statistical way to quantify the uncertainty associated with estimating the burrow percentage from 2D views, an important step toward a complete understanding of variability in burrow percentage (and bioturbation intensity) when estimated from 2D views.

A new approach to processing and imaging multibeam water column echosounder data: Application to a complex methane seep on the southern Cascadia margin

Acoustic echosounding systems are increasingly used to image water-column backscatter in addition to mapping the seafloor. We have imaged an acoustic flare generated by methane bubbles emanating from a vent sourced at 1840 m water depth offshore northern California using a shipboard Kongsberg EM122. Data include five transits over the flare and approximately 11 h of continuous observation when the ship held station. Shipboard observations showed a strong flare splitting into multiple smaller, intermittent flares at a water depth of 800–1200 m and pronounced temporal variability. We introduce a new approach to processing the data in which we correct the backscatter data for ship motion and bin the data into voxels with dimensions of 20 m in X and Y and 40 m in Z for a transit over the flare and into vertical slices with dimensions of 15 m in X and Z and 4 min in time when the ship was stationary. The processed data indicate that the signal is dominated by bubbles emanating from a source region with a diameter of approximately 40 m located on the southern edge of what is likely a ring of sources with a diameter of approximately 600 m. When the ship was stationary, we were able to track an individual pulse rising at a rate of 8–10 m/min. Our results illustrate the limitations of monitoring temporal variation in gas flux using multibeam echosounders because of the trade-off between imaging the entire flare by averaging over tens of minutes to hours and observing a slice through the flare to capture short-lived pulses of gas expulsion. Nevertheless, because echosounders are widely available, they can continue to provide valuable data on the spatial and temporal distribution of gas emissions on continental margins that can be used to frame hypotheses and plan more comprehensive follow-up experiments.

Spatial-Angular Versatile Convolution for Light Field Reconstruction

Spatial-angular separable convolution (SAS-conv) has been widely used for efficient and effective 4D light field (LF) feature embedding in different tasks, which mimics a 4D convolution by alternatively operating on 2D spatial slices and 2D angular slices. In this paper, we argue that, despite its global intensity modeling capabilities, SAS-conv can only embed local geometry information into the features, resulting in inferior performances in the regions with textures and occlusions. Because the epipolar lines are highly related to the scene depth, we introduce the concept of spatial-angular correlated convolution (SAC-conv). By alternating 2D convolutions on the vertical and horizontal epipolar slices, SAC-conv can embed global and robust geometry information into the features. We verify that SAS-conv and SAC-conv are skilled at different aspects of 4D LF feature embedding through a detailed feature and error analysis. Based on their complementarity, we further combine SAS-conv and SAC-conv by a parallel residual connection, forming a new spatial-angular versatile convolution (SAV-conv) module. We conduct comprehensive experiments on two representative LF reconstruction tasks, i.e., LF angular super-resolution and LF spatial super-resolution. Both the quantitative and qualitative results demonstrate that, without any extra parameters, networks upgraded with our proposed SAV-conv notably outperform those upgraded with SAS-conv and achieve a new state-of-the-art performance. Codes are available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/Joechann0831/SAV_conv</uri> .

Differences in the Buccal Bone Marrow Distance of ≤0.8 mm in the Mandible of Patients Undergoing Sagittal Split Ramus Osteotomy among the Different Skeletal Patterns: A Retrospective Study.

This study investigated the relationship between the thickness of the ramus and skeletal patterns using cone-beam computed tomography (CBCT) for sagittal split ramus osteotomy. Ninety participants were categorized into three skeletal patterns (Class I, Class II, and Class III). The first vertical slice (slice 0) was observed in the intact mandibular canal, and then moved forward to 40 mm (slice 21) with a 2 mm interval. The thickness of buccal bone marrow (B value) was measured. A B value of ≤0.8 mm was considered to be the major risk factor causing the occurrence of postoperative lower lip paresthesia. There were 461 sides with a B value of ≤0.8 mm. There was a significant difference in the skeletal patterns [Class III (198 sides: 15.7%) > Class I (159 sides: 12.6%) > Class II (104 sides: 8.3%)]. Class II participants had significantly larger B values (2.14 to 3.76 mm) and a lower occurrence rate (≤0.8 mm) than those of Class III participants (1.5 to 3 mm) in front of the mandibular foramen (from 6 mm to 20 mm). Class III participants had significantly shorter buccal bone marrow distance and a higher occurrence rate of B values (≤0.8 mm) than Class II.

Orogenic lithosphere and slabs in the greater Alpine area – interpretations based on teleseismic P-wave tomography

Abstract. Based on recent results of AlpArray, we propose a new model of Alpine collision that involves subduction and detachment of thick (∼ 180 km) European lithosphere. Our approach combines teleseismic P-wave tomography and existing local earthquake tomography (LET), allowing us to image the Alpine slabs and their connections with the overlying orogenic lithosphere at an unprecedented resolution. The images call into question the conventional notion that downward-moving lithosphere and slabs comprise only seismically fast lithosphere. We propose that the European lithosphere is heterogeneous, locally containing layered positive and negative Vp anomalies of up to 5 %–6 %. We attribute this layered heterogeneity to seismic anisotropy and/or compositional differences inherited from the Variscan and pre-Variscan orogenic cycles rather than to thermal anomalies. The lithosphere–asthenosphere boundary (LAB) of the European Plate therefore lies below the conventionally defined seismological LAB. In contrast, the lithosphere of the Adriatic Plate is thinner and has a lower boundary approximately at the base of strong positive Vp anomalies at 100–120 km. Horizontal and vertical tomographic slices reveal that beneath the central and western Alps, the European slab dips steeply to the south and southeast and is only locally still attached to the Alpine lithosphere. However, in the eastern Alps and Carpathians, this slab is completely detached from the orogenic crust and dips steeply to the north to northeast. This along-strike change in attachment coincides with an abrupt decrease in Moho depth below the Tauern Window, the Moho being underlain by a pronounced negative Vp anomaly that reaches eastward into the Pannonian Basin area. This negative Vp anomaly is interpreted as representing hot upwelling asthenosphere that heated the overlying crust, allowing it to accommodate Neogene orogen-parallel lateral extrusion and thinning of the ALCAPA tectonic unit (upper plate crustal edifice of Alps and Carpathians) to the east. A European origin of the northward-dipping, detached slab segment beneath the eastern Alps is likely since its down-dip length matches estimated Tertiary shortening in the eastern Alps accommodated by originally south-dipping subduction of European lithosphere. A slab anomaly beneath the Dinarides is of Adriatic origin and dips to the northeast. There is no evidence that this slab dips beneath the Alps. The slab anomaly beneath the Northern Apennines, also of Adriatic origin, hangs subvertically and is detached from the Apenninic orogenic crust and foreland. Except for its northernmost segment where it locally overlies the southern end of the European slab of the Alps, this slab is clearly separated from the latter by a broad zone of low Vp velocities located south of the Alpine slab beneath the Po Basin. Considered as a whole, the slabs of the Alpine chain are interpreted as highly attenuated, largely detached sheets of continental margin and Alpine Tethyan oceanic lithosphere that locally reach down to a slab graveyard in the mantle transition zone (MTZ).

Machine Learning Algorithms to Distinguish Myocardial Perfusion SPECT Polar Maps.

Myocardial perfusion imaging (MPI) plays an important role in patients with suspected and documented coronary artery disease (CAD). Machine Learning (ML) algorithms have been developed for many medical applications with excellent performance. This study used ML algorithms to discern normal and abnormal gated Single Photon Emission Computed Tomography (SPECT) images. We analyzed one thousand and seven polar maps from a database of patients referred to a university hospital for clinically indicated MPI between January 2016 and December 2018. These studies were reported and evaluated by two different expert readers. The image features were extracted from a specific type of polar map segmentation based on horizontal and vertical slices. A senior expert reading was the comparator (gold standard). We used cross-validation to divide the dataset into training and testing subsets, using data augmentation in the training set, and evaluated 04 ML models. All models had accuracy >90% and area under the receiver operating characteristics curve (AUC) >0.80 except for Adaptive Boosting (AUC = 0.77), while all precision and sensitivity obtained were >96 and 92%, respectively. Random Forest had the best performance (AUC: 0.853; accuracy: 0,938; precision: 0.968; sensitivity: 0.963). ML algorithms performed very well in image classification. These models were capable of distinguishing polar maps remarkably into normal and abnormal.

Layer-Averaged Water Temperature Sensing in a Lake by Acoustic Tomography with a Focus on the Inversion Stratification Mechanism.

Continuous sensing of water parameters is of great importance to fluid dynamic progress study in oceans, coastal areas and inland waters. The acoustic tomography technique can perform water temperature field measurements horizontally and vertically using sound wave travel information. The layer-averaged water temperature can also be measured with the acoustic tomography method. However, investigations focusing on the stratified mechanism, which consists of stratification form and its influence on inversion error, are seldom performed. In this study, an acoustic tomography experiment was carried out in a reservoir along two vertical slices to observe the layer-averaged water temperature. Specifically, multi-path sound travel information is identified through ray tracing using high-precision topography data obtained via a ship-mounted ADCP during the experiment. Vertical slices between sound stations are divided into different layers to study layer division inversion methods in different preset types. The inversion method is used to calculate the average water temperature and inversion temperature error of every layer. Different layer methods are studied with a comparison of results. The layer division principle studied in this paper can be used for layer-averaged water temperature sensing with multi-path sound transmission information.

A Numerical Analysis of Six Physics‐Dynamics Coupling Schemes for Atmospheric Models

AbstractSix strategies to couple the dynamical core with physical parameterizations in atmospheric models are analyzed from a numerical perspective. Thanks to a suitably designed theoretical framework featuring a high level of abstraction, the truncation error analysis and the linear stability study are carried out under weak assumptions. Indeed, second‐order conditions are derived which are not influenced either by the specific formulation of the governing equations, nor by the number of parameterizations, nor by the structural design and implementation details of the time‐stepping methods. The theoretical findings are verified on two idealized test beds. Particularly, a hydrostatic model in isentropic coordinates is used for vertical slice simulations of a moist airflow past an isolated mountain. Self‐convergence tests show that the sensitivity of the prognostic variables to the coupling scheme may vary. For those variables (e.g., momentum) whose evolution is mainly driven by the dry dynamics, the truncation error associated with the dynamical core dominates and hides the error due to the coupling. In contrast, the coupling error of moist variables (e.g., the precipitation rate) emerges gradually as the spatio‐temporal resolution increases. Eventually, each coupling scheme tends toward the formal order of accuracy, upon a careful treatment of the grid cell condensation. Indeed, the well‐established saturation adjustment may cap the convergence rate to first order. A prognostic formulation of the condensation and evaporation process is derived from first principles. This solution is shown effective to alleviate the convergence issues in our experiments. Potential implications for a complete forecasting system are discussed.

Modified pseudo-dynamic analysis of slope considering logarithmic spiral failure surface with numerical solution

ABSTRACT A methodology for the analysis of soil slope made up of c-ϕ soil using a modified pseudo-dynamic approach is tried to develop here. In this study, the slope is divided into a number of vertical slices and the failure surface of the slope is assumed to be logarithmic spiral. The suggested modified pseudo-dynamic approach satisfies the zero-stress boundary condition at the free ground surface and considers the damping properties of the materials. Results of the present analysis are presented in tabular form. The effects of the variation of different parameters like horizontal and vertical seismic acceleration, slope angle, soil friction angle, damping ratio, frequency ratio, cohesion and surcharge on the FOS are shown graphically. Consequently, required reinforcement strength is evaluated to ensure the safety of the slope under seismic loading conditions. The results obtained from the present method are compared with the results of the available literature and also a numerical validation of the model is given using PLAXIS 2D.

How Others See Us: Anthropologists, WikiLeaks, and the Vertical Slice

Abstract Drawing on Laura Nader’s concept of the vertical slice, this article reviews the hundreds of instances where the work of anthropologists, or anthropologists themselves appear in the leaked US State Department documents known as the “Manning Cables” published by WikiLeaks. The analysis of these documents shows anthropologists engaging with the US government in various ways, including in advisory capacities or bringing cultural or political knowledge from peripheral geographical regions to the core. Ethical, political, and disciplinary dimensions of these interactions are discussed, and Nader’s conception of the vertical slice is used to distinguish political dimensions of these anthropological engagements with state power.

Antennas Up! Laura Nader’s Undergraduate Lecture Courses as Public Anthropology

Abstract While teaching lecture courses at the University of California, Berkeley, Laura Nader taught generations of students to raise their anthropological antennae. This article uses an autoethnographic approach to describe the author’s exposure to anthropology at Berkeley in the nineteen-nineties, gesturing towards the way undergraduate lecture courses play an important but largely underrecognized role in fostering public anthropology. Nader’s lecture courses were particularly effective at this because their focus on pushing students to question dogma and analyze controlling processes offered students a sense of how anthropology could foster critical public discourse. Nader stressed the importance of asking good questions designed to challenge assumptions, finding the right methods to answer those questions, and paying attention to pathways of power. While always questioning received wisdom, ideological assumptions, and Western categories of knowledge, Nader continued to stress the importance of developing straightforward, highly-accessible concepts that captured the attention of students—like Harmony Ideology, trustanoia, controlling processes, and the vertical slice.

Studying Up: Four Modalities, Two Challenges

Abstract In her 1973 article “up the anthropologist” Laura Nader called on anthropologists to engage in critical studies of the relationship between powerful institutions and the broader society, using a “vertical slice” approach. But Nader worried that participant observation was hard in the context of studying up, and yet it has been presented as definitive of anthropology’s methodology. This article discusses four methodological strategies for studying up in the light of this concern: insider ethnography; covert ethnography; remote ethnography; and adapted participant observation. The first two have intellectual or ethical liabilities. The last is increasingly normalized. Going forward, anthropologists studying up face two obstacles: first, the increasingly totalizing hold of corporate and government workplaces over their employees, even when they are not at work; and, second, university institutional review boards (irb s) concerned to avoid conflictual or critical research.

Faults resembling channel-like features

We analyzed a synthetic transfer zone and its associated fault planes and relay ramp in Penobscot, a potential offshore field in the Scotian Basin. Transfer zones are structural areas where one fault dies out and another fault begins, forming a relay ramp in the middle. They can be categorized as divergent, convergent, and synthetic transfer zones depending on the relative location and dipping directions of the faults. These zones not only play an important role in fluid migration but also help interpreters delineate secondary features such as fractures, splay shears, and Riedel faults. Commonly, those faults would branch into smaller splays and the relay ramp can get “breached” with connecting faults with the increase of slip. The study area in the Scotian Basin is characterized by two major listric normal faults dipping in the same direction giving rise to a synthetic transfer zone. These faults are clearly visible on seismic attributes, including curvature and coherence slices extracted along the top of the Cretaceous Petrel Formation. However, when analyzing the seismic attributes along the overlying Wyandot Formation’s top, we observe channel-like features, which run parallel as well as at an angle to these faults. However, when we performed further analysis using seismic amplitude’s vertical slices, interpreted horizons, and seismic attributes, we found that these features are not channels. We divided the features into two types: The first type is parallel to the main faults and can be associated with the grabens formed by synthetic and antithetic secondary faults (northeast–southwest); the second type is related to the polygonal faulting associated with differential compaction and gravitational loading of the Wyandot Chalk Formation. Apart from the two lineations, there are north-northeast–south-southwest-oriented lineations, which are an impression of basement faulting, and north-northwest–south-southeast-oriented lineations representing the acquisition footprint. Geological feature: Synthetic and antithetic faults in a synthetic transfer zone Seismic appearance: Channel-like feature on seismic attribute horizon slices Alternative interpretations: Channels Features with similar appearance: Steep slope channels Formation: Wyandot Age: Late Cretaceous Location: Penobscot, Nova Scotia, Canada Seismic data: Open-source SEG Analysis tools: Coherence and curvature attributes

Geophysical investigations to search for the remains of sister Chiara Isabella D'Amato

This paper reports the results of the integrated geophysical surveys performed inside the Monastery of St Chiara in Nardó in southern Italy. Ground-penetrating radar (GPR) and electrical resistivity tomography (ERT) investigations were carried out to find the tomb of Sister Chiara D’Amato, whose location has been lost over the centuries. Sister Chiara Isabella D'Amato was a charismatic and holy figure. She died in 1693. She was buried inside the monastery but her body has never been found. The data acquisition was performed along with a series of closely spaced lines for GPR and using a non-standard array for ERT. Data were processed and visualised as two-dimensional vertical sections and depth slices or three-dimensional volumes (GPR and ERT) to allow an integrated interpretation of the geophysical results. The analysis of the geophysical data sets revealed a series of anomalies that could be ascribed possible archaeological structures, probably related to the earliest ages of the sacred building as well as other anomalies (bedrock, fractures) of presumable natural origin. In particular, one geophysical anomaly was suspected of being connected to burial and consequently further investigated with the use of a video endoscope. The results reveal the presence of a void but it has not yet been clarified whether it is the burial of Sister Chiara or not as it is awaiting the excavation.

Continuous Sensing of Water Temperature in a Reservoir with Grid Inversion Method Based on Acoustic Tomography System

The continuous sensing of water parameters is of great importance to the study of dynamic processes in the ocean, coastal areas, and inland waters. Conventional fixed-point and ship-based observing systems cannot provide sufficient sampling of rapidly varying processes, especially for small-scale phenomena. Acoustic tomography can achieve the sensing of water parameter variations over time by continuously using sound wave propagation information. A multi-station acoustic tomography experiment was carried out in a reservoir with three sound stations for water temperature observation. Specifically, multi-path propagation sound waves were identified with ray tracing using high-precision topography data obtained with ship-mounted ADCP. A new grid inverse method is proposed in this paper for water temperature profiling along a vertical slice. The progression of water temperature variation in three vertical slices between acoustic stations was mapped by solving an inverse problem. The reliability and adaptability of the grid method developed in this research are verified by comparison with layer-averaged water temperature results. The grid method can be further developed for the 3D mapping of water parameters over time, especially in small-scale water areas, where sufficient multi-path propagation sound waves can be obtained.

A Novel Method for the Stability Assessment of Soil Slopes with Multi-layers Based on the Upper Bound Limit Analysis

The discrete technique combined with the upper bound approach is applied to analyze the stability of soil slopes with multiple layers in this paper. Based on the geometrical relationship derived from the normality requirement, an algorithm is proposed to determine the coordinates of discrete points on the failure surface. Specifically, these points are generated from the initial point to the terminal point. After completion of the discrete sliding surface, the sliding block is sliced by the vertical slice technique. Equations with integral form for rates of internal energy dissipation and external work are then derived. The scheme for calculation of the safety factor is presented by means of the strength reduction method and the optimization algorithm. Comparison analysis based on four examples under various conditions demonstrates the validity of the proposed method in addressing stability problem of slopes with multiple layers.

Tensor learning of pointwise mutual information from EHR data for early prediction of sepsis

Early detection of sepsis can facilitate early clinical intervention with effective treatment and may reduce sepsis mortality rates. In view of this, machine learning-based automated diagnosis of sepsis using easily recordable physiological data can be more promising as compared to the gold standard rule-based clinical criteria in current practice. This study aims to develop such a machine learning framework that demonstrates the quantification of heterogeneity within the tabular electronic health records (EHR) data of clinical covariates to capture both linear relationships and nonlinear correlation for the early prediction of sepsis. Here, the statistics of pairwise association for each hour-covariate pair within the EHR data for every 6-hours window-duration with selected 24 covariates is described using pointwise mutual information (PMI) matrix. This matrix gives the heterogeneity of data as a two-dimensional map. Such matrices are fused horizontally along the z-axis as vertical slices in the xy plane to form a 3-way tensor for each record with the corresponding Length of Stay (L). Tensor factorization of such fused tensor for every record is performed using Tucker decomposition, and only the core tensors are retained later, excluding the 3 unitary matrices to provide the latent feature set for the prediction of sepsis onset. A five-fold cross-validation scheme is employed wherein the obtained 120 latent features from the reshaped core tensor, are fed to Light Gradient Boosting Machine Learning models (LightGBM) for binary classification, further alleviating the involved class imbalance. The machine-learning framework is designed via Bayesian optimization, yielding an average normalized utility score of 0.4519 as defined by challenge organizers and area under the receiver operating characteristic curve (AUROC) of 0.8621 on publicly available PhysioNet/Computing in Cardiology Challenge 2019 training data.The proposed tensor decomposition of 3-way fused tensor formulated using PMI matrices leverages higher-order temporal interactions between the pairwise associations among the clinical values for early prediction of sepsis. This is validated with improved risk prediction power for every hour of admission to the ICU in terms of utility score, AUROC, and F1 score. The results obtained show a significant improvement particularly in terms of utility score of ~1.5−2% under a 5-fold cross-validation scheme on entire training data as compared to a top entrant research study that participated in the challenge.

The three-dimensional ionospheric electron density imaging in Japan using the approximate Kalman filter algorithm

The three-dimensional ionospheric electron density (Ne) over Japan is retrieved by assimilating the ground-based total electron content (TEC) datasets into the International Reference Ionosphere (IRI) background model. To reduce the computational burdens encountered in the full Kalman filter (KF) algorithm, the simplified KF (SKF), a time-independent data assimilation scheme with no recursion, is adopted. The spatial resolution is 1°, 1°, and 30 km in latitude, longitude, and altitude, and the temporal resolution is 1 h. Regarding the quite geomagnetic condition, seasonal variations in the ionosphere were investigated firstly. The results show that the reconstructed Ne could well capture the seasonal ionospheric characteristics and semiannual anomaly. Subsequently, NmF2 (F2 layer peak electron density) values during April 3–9, 2013 (high solar activity) and 2018 (low solar activity) were reproduced and simulated by the SKF and IRI model, respectively. In contrast with the ionosonde data, the results indicated the reconstructed NmF2 was apt to capture the underlying trends of diurnal and annual variations more accurately. For the efficiency under the disturbed geomagnetic condition, we took a great geomagnetic storm that happened on August 26, 2018 as an example. The reproduced results also provided better imaging for Ne distributions by analyzing the vertical slices and profiles provided by SKF and IRI. It is suggested that the SKF algorithm is promising for ionospheric researches and practical applications in Japan.