Scalable Model Research Articles

Two-Speed Deep-Learning Ensemble for Classification of Incremental Land-Cover Satellite Image Patches

High-velocity data streams present a challenge to deep learning-based computer vision models due to the resources needed to retrain for new incremental data. This study presents a novel staggered training approach using an ensemble model comprising the following: (i) a resource-intensive high-accuracy vision transformer; and (ii) a fast training, but less accurate, low parameter-count convolutional neural network. The vision transformer provides a scalable and accurate base model. A convolutional neural network (CNN) quickly incorporates new data into the ensemble model. Incremental data are simulated by dividing the very large So2Sat LCZ42 satellite image dataset into four intervals. The CNN is trained every interval and the vision transformer trained every half interval. We call this combination of a complementary ensemble with staggered training a “two-speed” network. The novelty of this approach is in the use of a staggered training schedule that allows the ensemble model to efficiently incorporate new data by retraining the high-speed CNN in advance of the resource-intensive vision transformer, thereby allowing for stable continuous improvement of the ensemble. Additionally, the ensemble models for each data increment out-perform each of the component models, with best accuracy of 65% against a holdout test partition of the RGB version of the So2Sat dataset.

Modeling and control of a photovoltaic-wind hybrid microgrid system using GA-ANFIS

This paper aims to model a PV-Wind hybrid microgrid that incorporates a Battery Energy Storage System (BESS) and design a Genetic Algorithm-Adaptive Neuro-Fuzzy Inference System (GA-ANFIS) controller to regulate its voltage amid power generation variations. Two microgrid models have been developed; a scalable Simulink Case Study Model from underlying mathematical equations and a nested voltage-current loop-based Transfer Function model. The proposed GA-ANFIS controller has been used as a Maximum Power Point Tracking (MPPT) algorithm to optimize the converter outputs and provide voltage regulation. The performance of the GA-ANFIS algorithm was compared with the Search Space Restricted-Perturb and Observe (SSR-P&O) and the Proportional-plus-Integral-plus-Derivative (PID) controllers using a simulation model built in MATLAB/SIMULINK. Results indicated that the GA-ANFIS controller is superior to the SSR-P&O and PID in terms of reduced rise time, settling time, overshoot, and the ability to handle non-linearities in the microgrid. In future work, the GA-ANFIS microgrid control system can be replaced with a three-term hybrid artificial intelligence algorithms controller.

A Standardized Clinical Data Harmonization Pipeline for Scalable AI Application Deployment (FHIR-DHP): Validation and Usability Study

Increasing digitalization in the medical domain gives rise to large amounts of health care data, which has the potential to expand clinical knowledge and transform patient care if leveraged through artificial intelligence (AI). Yet, big data and AI oftentimes cannot unlock their full potential at scale, owing to nonstandardized data formats, lack of technical and semantic data interoperability, and limited cooperation between stakeholders in the health care system. Despite the existence of standardized data formats for the medical domain, such as Fast Healthcare Interoperability Resources (FHIR), their prevalence and usability for AI remain limited. In this paper, we developed a data harmonization pipeline (DHP) for clinical data sets relying on the common FHIR data standard. We validated the performance and usability of our FHIR-DHP with data from the Medical Information Mart for Intensive Care IV database. We present the FHIR-DHP workflow in respect of the transformation of "raw" hospital records into a harmonized, AI-friendly data representation. The pipeline consists of the following 5 key preprocessing steps: querying of data from hospital database, FHIR mapping, syntactic validation, transfer of harmonized data into the patient-model database, and export of data in an AI-friendly format for further medical applications. A detailed example of FHIR-DHP execution was presented for clinical diagnoses records. Our approach enables the scalable and needs-driven data modeling of large and heterogenous clinical data sets. The FHIR-DHP is a pivotal step toward increasing cooperation, interoperability, and quality of patient care in the clinical routine and for medical research.

Performance enhancement of electrodialysis regeneration of deep eutectic solvent for liquid desiccant air conditioning systems

Despite the promising energy-saving potential of the electrodialysis regeneration technique for liquid desiccant regeneration in air conditioning systems, there are still issues that need to be resolved in electrodialysis separation systems, such as the high initial cost owing to the complexity of the cell gaps, over-limiting current, and electro-osmotic flow effects. This study utilized stereolithography (SLA) 3-D printing technology to manufacture a low-cost scalable model of an electrodialysis system. An electrodialysis system with a cell gap of 5 mm and serpentine flow feed spacers was investigated to regenerate deep eutectic solvent (DES) developed from choline chloride and ethylene glycol as a bio-friendly liquid desiccant for air conditioning systems. Ammonium chloride electrode rinsing solution with about 250 mS/cm electric conductivity was utilized to enhance the conductive properties of the system. The optimum parametric operating conditions for the system are found to be around 0.0058 m/s for the circulation flow rate, 0–4 wt.% for the concentration difference of the diluted and regenerated solutions' cells, and 6–8 V voltage supply energy input. In 60 minutes of operation, the current efficiency value of the constructed model was discovered to be as high as 65.82%. Matching the performance threshold of the commercial electrodialysis equipment described in the literature for regenerating lithium chloride conventional liquid desiccant.

Scalable growth models for time‐series multispectral images

AbstractVegetation indices (VIs) are produced as a combination of different reflectance bands that are captured by multispectral images (MSIs). These indices, such as normalized difference vegetation index (NDVI), are reported to be proxy indicators of photosynthetic activity, plant canopy biomass, and leaf area index. To determine the utility of using VI derived from MSI to model plant growth, random regression (RR) models with linear splines and different orders of Legendre polynomials were applied to data collected (years 2019 and 2020) as part of the Genome‐to‐Fields initiative. Growth curves of maize (Zea mays L.) hybrids were modeled using both NDVI and cumulative NDVI (cNDVI) phenotypes. Due to the difference in MSI recording dates, and sparse overlap in hybrids between years, all the analyses were nested within a year. Results indicate that RR models using Legendre polynomials provide a robust and scalable method for modeling growth curves using phenotypes extracted from MSI; however, RR models using linear splines showed inconsistent convergence. Growth curves estimated using NDVI and cNDVI showed low‐to‐moderate heritability (0.11–0.44) and a range of genetic correlations (−0.15 to 0.97) with grain yield. This study demonstrates the utility of MSI for modeling genetic growth trends, with the best modeling results obtained when using Legendre polynomials and cNDVI.

Tie-Line Power Transferred: Data Security Using Block Chain Technology

In this paper, the authors have used Block Chain technology for tie-line power transferred data in respect to trustworthy monitoring, non-tamper proof and traceability. Nepanagar–Dharni tie-line between MP Electricity Board and Maharashtra Electricity Board is considered. Also, an example of a solar power plant connected to the grid and a consumer is considered to show the security concern using Block Chain Technology. The intrinsic features of decentralized block chain technology enable the grid owner, utility and consumer to maintain the privacy and security of the data set and Image sharing of tie-line power transferred data and Image. The smart contract can register and authorize the grid corporation authority to access the tie-line power transferred during the period and make payments to the concern in compliance with the regulatory and involved parties’ consent policy. Security’s role as a trust model is used to sort out the discrepancy in the exchange for power monitored by the phase measurement unit. Distributed and scalable data models based on the block chain are implemented using a heterogeneous set of MYSQL database management systems, hosted on the AWS cloud. If the power is transferred to a tie line connected to the Railway traction system, the data and Image of the train and its status can be secured with block chain.

Multilayer self‐organized aggregation control for large‐scale unmanned aerial vehicle swarms

SummaryThe cooperative control of large‐scale unmanned aerial vehicle (UAV) swarms to establish a complex configuration is one of the most challenging goals. The poor scalability of the network topology, inability to self‐organize the desired configuration and the separation of swarm configuration are the main obstructions that hinder the attainment of the goal. This paper studies the self‐organized aggregation control of large‐scale UAVs by designing a scalable distributed network architecture. We present a novel centerless and scalable multilayer graph model to solve the nonextensible problem of the traditional network topology. A large number of UAVs are organized with several interconnected and globally covered layers, and each layer consists of many undifferentiated clusters. Afterwards, a self‐organized aggregation control law using the potential function method is designed for UAVs to realize autonomous configuration and configuration swarming under the consideration of collision and obstacle avoidance.

Conversion and SISO Equivalence of Impedance Model of Single-Phase Converter in Electric Multiple Units

The direct-quadrature (DQ) model, mirror-frequency impedance (MFI) model, and harmonic state-space (HSS) model are three main modeling methods for single-phase power electronic converter systems. These models are established under three different frameworks: the dq-frame, the mirror-frequency frame, and the extended harmonic domain frame. The purpose of this article is to demonstrate how to create a unified modeling framework and to provide an unambiguous relationship between the impedances in these three domains. The unified model framework facilitates flexible and scalable modeling by making different models to be easily interconnected with other systems. In addition, to simplify the stability analysis and impedance measurement process, the three modeling methods are reduced to first-order single-input and single-output (SISO) equivalent models, respectively, based on the model reduction technique. Finally, the impedance measurement and stability analysis results of electric multiple units (EMUs) accessed vehicle-grid system demonstrate the accuracy of the conversion and SISO equivalence technique.

Leveraging Electronic Health Record Data for Timely Chronic Disease Surveillance: The Multi-State EHR-Based Network for Disease Surveillance.

Electronic health record (EHR) data can potentially make chronic disease surveillance more timely, actionable, and sustainable. Although use of EHR data can address numerous limitations of traditional surveillance methods, timely surveillance data with broad population coverage require scalable systems. This report describes implementation, challenges, and lessons learned from the Multi-State EHR-Based Network for Disease Surveillance (MENDS) to help inform how others work with EHR data to develop distributed networks for surveillance. Funded by the Centers for Disease Control and Prevention (CDC), MENDS is a data modernization demonstration project that aims to develop a timely national chronic disease sentinel surveillance system using EHR data. It facilitates partnerships between data contributors (health information exchanges, other data aggregators) and data users (state and local health departments). MENDS uses query and visualization software to track local emerging trends. The program also uses statistical and geospatial methods to generate prevalence estimates of chronic disease risk measures at the national and local levels. Resulting data products are designed to inform public health practice and improve the health of the population. MENDS includes 5 partner sites that leverage EHR data from 91 health system and clinic partners and represents approximately 10 million patients across the United States. Key areas of implementation include governance, partnerships, technical infrastructure and support, chronic disease algorithms and validation, weighting and modeling, and workforce education for public health data users. MENDS presents a scalable distributed network model for implementing national chronic disease surveillance that leverages EHR data. Priorities as MENDS matures include producing prevalence estimates at various geographic and subpopulation levels, developing enhanced data sharing and interoperability capacity using international data standards, scaling the network to improve coverage nationally and among underrepresented geographic areas and subpopulations, and expanding surveillance of additional chronic disease measures and social determinants of health.

Scalable Charge-Based Compact Model for Drain Current in Fin-Shaped GaN HEMTs

We present a novel charge-based compact model for drain current in fin-shaped GaN high-electron-mobility transistors (Fin-HEMTs). A positive threshold voltage shift is observed in Fin-HEMTs due to better gate control of the channel. The combined effect of top and side gates on the channel charge density is studied with the help of detailed TCAD simulation. For the first time, a closed-form expression for 2-D electron gas (2-DEG) density in Fin-HEMTs is derived. The fin width dependence of the 2-DEG current is modeled in a unique way by considering the modified gate capacitance and threshold voltage in Fin-HEMTs. The total drain current is modeled as the sum of 2-DEG and double-gate field effect transistor (FET) currents. In addition, the total gate capacitance including the side-gate capacitance for fin structure is also formulated. The model is generic and can be used for different fin geometries using the proposed equivalent circuits as demonstrated. A fin-width-dependent thermal resistance model is proposed and included in the drain current model using a thermal subcircuit. The proposed model is completely analytical and does not require any iteration or fitting parameters in contrast to the available models discussed in the literature. Our model shows excellent agreement with the state-of-the-art experimental data.

Towards Scalable Model Checking of Reflective Systems via Labeled Transition Systems

Reflection is a technique that enables a system to inspect or change its structure and/or behavior at runtime. It is a key enabler of many techniques for developing systems that have to function despite rapidly changing requirements and environments. A crucial issue in developing reflective systems is to ensure the correctness of their behaviors, because object-level behaviors are affected by metalevel behaviors. In this paper, we present an extended labeled transition system (LTS), which we call a metalevel LTS (MLTS), that supports data representation of another LTS for use in modeling a reflective tower. We show that two of the existing state reduction techniques for an LTS ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">symmetry reduction</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">divergence-sensitive stutter bisimulation</i> ) are also applicable to an MLTS. Then, we introduce two strategies for implementing an MLTS model in Promela, thereby enabling verification with the SPIN model checker. We also present case studies of applying MLTSs to two reflection applications: self-adaptation of a reconnaissance robot system, and dynamic evolution of an Internet-of-things (IoT) system. The case studies demonstrate the applicability of our approach and its scalability improvement through the state reduction techniques.

The interagency care team: A new model to integrate social and medical care for older adults in primary care.

To integrate management of social drivers of health with complex clinical needs of older adults, we connected patients aged 60 and above from primary care practices with a nurse practitioner (NP) led Interagency Care Team (ICT) of geriatrics providers and community partners via electronic consult. The NP conducted a geriatric assessment via telephone, then the team met to determine recommendations. Thirteen primary care practices referred 123 patients (median age=76) who had high rates of emergency department use and hospitalization (28.9% and 17.4% respectively). Issues commonly identified included medication management (84%), personal safety (72%), disease management (69%), food insecurity (63%), and cognitive decline (53%). Referring providers expressed heightened awareness of older adults' social needs and high satisfaction with the program. The ICT is a scalable model of care that connects older adults with complex care needs to geriatrics expertise and community services through partnerships with primary care providers.

Using jointly calibrated fine-scale drain models across Denmark to assess the influence of physical variables on spatial drain flow patterns

Study regionDenmark Study focusTile drainage, widespread across agricultural areas in Denmark, significantly impacts the hydrological cycle. Tile drain flow dynamics and their spatial patterns are crucial for water managers to address water quality and quantity issues. However, these processes are challenging to simulate accurately. In this study, we developed 10 m resolution drain models in MIKE SHE for 26 drain sites across Denmark to predict drain flow spatial and temporal dynamics. Joint calibration of all drain models was conducted by evaluating PBIAS and KGE of simulated and observed drain flow. Subsequently, we performed a correlation analysis between physical parameters and spatial patterns of simulated drain fraction (DF, ratio of drain flow to recharge per grid cell). New hydrological insightsThe jointly calibrated models achieved average PBIAS and KGE of − 6.7% and 0.53 respectively, for drain flow predictions across 26 drain sites. The correlation of DF with topographical variables was highest on a national scale and on most of the drain sites, including Gyldenholm1–4, Norsminde1–11, and other Jylland drain sites. Lillebæk drain sites showed a high correlation with average clay content in 0–30 cm layer of soil. Thus, in addition to developing scalable fine-resolution drain models for Denmark, this study also identifies the control parameters for spatial patterns of drain flows across Denmark.

Towards POI-based large-scale land use modeling: spatial scale, semantic granularity, and geographic context

ABSTRACT The combination of spatial distribution, semantic characteristics, and sometimes temporal dynamics of POIs inside a geographic region can capture its unique land use characteristics. Most previous studies on POI-based land use modeling research focused on one geographic region and select one spatial scale and semantic granularity for land use characterization. There is a lack of understanding on the impact of spatial scale, semantic granularity, and geographic context on POI-based land use modeling, particularly large-scale land use modeling. In this study, we developed a scalable POI-based land use modeling framework and examined the impact of these three factors on POI-based land use characterization using data from three geographic regions. We developed a unified semantic representation framework for POI semantics that can help fuse heterogeneous POI data sources. Then, by combining POIs with a neural network language model, we developed a spatially explicit approach to learn the embedding representation of POIs and AOIs. We trained multiple supervised classifiers using AOI embeddings as input features to predict AOI land use at different semantic granularities. The classification performance of different land use classes was analyzed and compared across three geographic regions to identify the semantic representativeness of POI-based AOI embedding and the impact of geographic context.

A Cre-LoxP-based approach for combinatorial chromosome rearrangements in human HAP1 cells.

Alterations of human karyotype caused by chromosomal rearrangements are often associated with considerable phenotypic effects. Studying molecular mechanisms underlying these effects requires an efficient and scalable experimental model. Here, we propose a Cre-LoxP-based approach for the generation of combinatorial diversity of chromosomal rearrangements. We demonstrate that using the developed system, both intra- and inter-chromosomal rearrangements can be induced in the human haploid HAP1 cells, although the latter is significantly less effective. The obtained genetically modified HAP1 cell line can be used to dissect genomic effects associated with intra-chromosomal structural variations.

MiMiCPy: An Efficient Toolkit for MiMiC-Based QM/MM Simulations.

MiMiC is a highly flexible, extremely scalable multiscale modeling framework. It couples the CPMD (quantum mechanics, QM) and GROMACS (molecular mechanics, MM) codes. The code requires preparing separate input files for the two programs with a selection of the QM region. This can be a tedious procedure prone to human error, especially when dealing with large QM regions. Here, we present MiMiCPy, a user-friendly tool that automatizes the preparation of MiMiC input files. It is written in Python 3 with an object-oriented approach. The main subcommand PrepQM can be used to generate MiMiC inputs directly from the command line or through a PyMOL/VMD plugin for visually selecting the QM region. Many other subcommands are also provided for debugging and fixing MiMiC input files. MiMiCPy is designed with a modular structure that allows seamless extensions to new program formats depending on the requirements of MiMiC.

Statistical multi-level shape models for scalable modeling of multi-organ anatomies.

Statistical shape modeling is an indispensable tool in the quantitative analysis of anatomies. Particle-based shape modeling (PSM) is a state-of-the-art approach that enables the learning of population-level shape representation from medical imaging data (e.g., CT, MRI) and the associated 3D models of anatomy generated from them. PSM optimizes the placement of a dense set of landmarks (i.e., correspondence points) on a given shape cohort. PSM supports multi-organ modeling as a particular case of the conventional single-organ framework via a global statistical model, where multi-structure anatomy is considered as a single structure. However, global multi-organ models are not scalable for many organs, induce anatomical inconsistencies, and result in entangled shape statistics where modes of shape variation reflect both within- and between-organ variations. Hence, there is a need for an efficient modeling approach that can capture the inter-organ relations (i.e., pose variations) of the complex anatomy while simultaneously optimizing the morphological changes of each organ and capturing the population-level statistics. This paper leverages the PSM approach and proposes a new approach for correspondence-point optimization of multiple organs that overcomes these limitations. The central idea of multilevel component analysis, is that the shape statistics consists of two mutually orthogonal subspaces: the within-organ subspace and the between-organ subspace. We formulate the correspondence optimization objective using this generative model. We evaluate the proposed method using synthetic shape data and clinical data for articulated joint structures of the spine, foot and ankle, and hip joint.

A Systematic Literature Review on Urban Resilience Enabled with Asset and Disaster Risk Management Approaches and GIS-Based Decision Support Tools

Urban Resilience (UR) enables cities and communities to optimally withstand disruptions and recover to their pre-disruption state. There is an increasing number of interdisciplinary studies focusing on conceptual frameworks and/or tools seeking to enable more efficient decision-making processes that lead to higher levels of UR. This paper presents a systematic review of 68 Scopus-indexed journal papers published between 2011 and 2022 that focus on UR. The papers covered in this study fit three categories: literature reviews, conceptual models, and analytical models. The results of the review show that the major areas of discussion in UR publications include climate change, disaster risk assessment and management, Geographic Information Systems (GIS), urban and transportation infrastructure, decision making and disaster management, community and disaster resilience, and green infrastructure and sustainable development. The main research gaps identified include: a lack of a common resilience definition and multidisciplinary analysis, a need for a unified scalable and adoptable UR model, margin for an increased application of GIS-based multidimensional tools, stochastic analysis of virtual cities, and scenario simulations to support decision making processes. The systematic literature review undertaken in this paper suggests that these identified gaps can be addressed with the aid of asset and disaster risk management methods combined with GIS-based decision-making tools towards significantly improving UR.

Enabling Informed Decision Making in the Absence of Detailed Nutrition Labels: A Model to Estimate the Added Sugar Content of Foods.

Obesity and diabetes have emerged as an increasing threat to public health, and the consumption of added sugar can contribute to their development. Though nutritional content information can positively influence consumption behavior, added sugar is not currently required to be disclosed in all countries. However, a growing proportion of the world's population has access to mobile devices, which allow for the development of digital solutions to support health-related decisions and behaviors. To test whether advances in computational science can be leveraged to develop an accurate and scalable model to estimate the added sugar content of foods based on their nutrient profile, we collected comprehensive nutritional information, including information on added sugar content, for 69,769 foods. Eighty percent of this data was used to train a gradient boosted tree model to estimate added sugar content, while 20% of it was held out to assess the predictive accuracy of the model. The performance of the resulting model showed 93.25% explained variance per default portion size (84.32% per 100 kcal). The mean absolute error of the estimate was 0.84 g per default portion size (0.81 g per 100 kcal). This model can therefore be used to deliver accurate estimates of added sugar through digital devices in countries where the information is not disclosed on packaged foods, thus enabling consumers to be aware of the added sugar content of a wide variety of foods.

Video-based marker-free tracking and multi-scale analysis of mouse locomotor activity and behavioral aspects in an open field arena: A perspective approach to the quantification of complex gait disturbances associated with Alzheimer's disease.

Complex gait disturbances represent one of the prominent manifestations of various neurophysiological conditions, including widespread neurodegenerative disorders such as Alzheimer's and Parkinson's diseases. Therefore, instrumental measurement techniques and automatic computerized analysis appears essential for the differential diagnostics, as well as for the assessment of treatment effectiveness from experimental animal models to clinical settings. Here we present a marker-free instrumental approach to the analysis of gait disturbances in animal models. Our approach is based on the analysis of video recordings obtained with a camera placed underneath an open field arena with transparent floor using the DeeperCut algorithm capable of online tracking of individual animal body parts, such as the snout, the paws and the tail. The extracted trajectories of animal body parts are next analyzed using an original computerized methodology that relies upon a generalized scalable model based on fractional Brownian motion with parameters identified by detrended partial cross-correlation analysis. We have shown that in a mouse model representative movement patterns are characterized by two asymptotic regimes characterized by integrated 1/f noise at small scales and nearly random displacements at large scales separated by a single crossover. More detailed analysis of gait disturbances revealed that the detrended cross-correlations between the movements of the snout, paws and tail relative to the animal body midpoint exhibit statistically significant discrepancies in the Alzheimer's disease mouse model compared to the control group at scales around the location of the crossover. We expect that the proposed approach, due to its universality, robustness and clear physical interpretation, is a promising direction for the design of applied analysis tools for the diagnostics of various gait disturbances and behavioral aspects in animal models. We further believe that the suggested mathematical models could be relevant as a complementary tool in clinical diagnostics of various neurophysiological conditions associated with movement disorders.