Multivariate Linear Regression Method Research Articles

Comparative Study of Frequency Recognition Techniques for Steady-State Visual Evoked Potentials According to the Frequency Harmonics and Stimulus Number.

A key challenge in steady-state visual evoked potential (SSVEP)-based brain-computer interface (BCI) systems is to effectively recognize frequencies within a short time window. To address this challenge, the specific characteristics of the data are needed to select the frequency recognition method. These characteristics include factors, such as the number of stimulation targets and the presence of harmonic frequencies, resulting in optimizing the performance and accuracy of SSVEP-based BCI systems. The current study aimed to examine the effect of data characteristics on frequency recognition accuracy. In this analytical study, five commonly used frequency recognition methods were examined, used to various datasets containing different numbers of frequencies, including sub-data with and without frequency harmonics. The increase in the number of frequencies in the Multivariate Linear Regression (MLR) method has led to a decrease in frequency recognition accuracy by 9%. Additionally, the presence of harmonic frequencies resulted in an 8% decrease in accuracy for the MLR method. Frequency recognition using the MLR method reduces the effect of the number of different frequencies and harmonics of the stimulation frequencies on the frequency recognition accuracy.

A fast and high-fidelity multi-parameter thermal-field prediction system based on CFD and POD coupling: Application to the RPV insulation structure

The thermal performance of the reactor pressure vessel (RPV) insulation structure plays a crucial role in ensuring the operational safety of the reactor. In this study, a fast and high-fidelity prediction system is proposed, which is based on the snapshot ensemble generated by the computational fluid dynamics (CFD) method and incorporates proper orthogonal decomposition (POD) as its core component to predict the heat transfer performance of the RPV insulation structure. The proposed thermal-field prediction system is employed to examine the impacts of air inlet temperature, mass flow rate, and RPV outer wall temperature on the thermal performance parameters: average heat flux qout and temperature Tout of the RPV insulation outer wall, as well as the maximum local temperature Tmax of the concrete pit inner wall. The POD predictive results showcase that the maximum relative deviations of qout, Tout and Tmax for eight sets of off-design cases are 2.25 %, 3.04 %, and 2.37 %, respectively. Meanwhile, the maximum values of qout, Tout and Tmax are 90.13 W·m−2, 44.69 ℃, and 87.51 ℃, respectively. They are all less than their prescribed limits. The field distributions of heat flux and temperature in off-design cases exhibit consistency between the CFD and POD methods. Notably, in comparison to the CFD simulation method, the POD method capturing more than 99.9 % energy demonstrates significant computational time savings for 99.87 %. Moreover, a multivariate linear regression (MLR) model for thermal performance parameters is established based on the POD prediction datasets. The results indicate that the maximum relative deviations between the POD and MLR methods in predicting qout, Tout and Tmax for the 72 sets of test cases are 5.1 %, 7.9 % and 7.1 %, respectively. Meanwhile, 99.99 % CPU time is reduced for MLR model compared with CFD method. Obviously, the predictive system showcases excellent accuracy and efficiency in forecasting.

Theoretical analysis of using multiple borehole heat exchangers for production of heating and cooling energy in shallow geothermal reservoirs with underground water flow

In this work the feasibility of using the double-U borehole heat exchangers (BHE) in multiple arrangements for production of heating and cooling energy has been investigated.Theoretical case studies are given for a few hypothetical lithological cases corresponding to the lithology that can be found in City of Zagreb. The results show interference between the borehole heat exchangers and limits of use of shallow reservoir with respect to slow convection and loss of accumulated energy in the reservoir. Special emphasis is placed on the description of the groundwater flow determined by the parameters of the shallow geothermal reservoir, such as permeability, water saturation, thermal conductivity, specific heat capacity and soil density, topography, precipitation, etc. Simulations have been performed in FeFlow coupled with Python user-scripts for individual control of the BHE’s, and the surface equipment is modelled using RES2GEO. To assess the accuracy of numerical predictions in simulating thermal responses, FeFlow software was utilized to conduct Thermal Response Tests (TRT) for individual Borehole Heat Exchangers (BHEs). These numerical results were then juxtaposed with empirical TRT data collected from a site near Velika Gorica, a town within the City of Zagreb. For each case simulation is done for one year with one hour resolution. The temperature distribution analysis highlights notable differences between scenarios, where the temperature field with lower permeability experiences significant variations, particularly near the BHE. Here, temperatures can soar to 20 °C during cooling periods and plummet to 4 °C in heating seasons. These fluctuations cause overheating and subcooling of the reservoir, impairing the heat pump’s efficiency and necessitating increased electricity consumption. By using Multivariate linear regression and Multi-layer perceptron regressor method, the main parameter for determination of COP of ground source heat pump. The analysis reveals that the majority of errors fall within the range of ± 5 %, although the MLA prediction method exhibits slightly more frequent errors. Additionally, the R2 scores demonstrate strong performance: the MLA procedure achieves 0.95 for heating and 0.99 for cooling, while the MLP procedure attains 0.99 for both heating and cooling phases.

Artificial intelligence probabilities scheme for disease prevention data set construction in intelligent smart healthcare scenario

In the face of an aging population, smart healthcare services are now within reach, thanks to the proliferation of high-speed internet and other forms of digital technology. Data problems in smart healthcare, unfortunately, put artificial intelligence in this area to serious limitations. There are several issues, including a lack of standard samples, noisy data interference, and actual data that is missing. A three-stage AI-based data generating strategy is suggested to handle missing datasets, using a small sample dataset obtained from a smart healthcare program community in a specific city: Step one involves generating the dataset's basic attributes using a tree-based generation strategy that takes the original data distribution into account. Step two involves using the Naive Bayes algorithm to create basic indicators of behavioural capability assessment for the samples. Step three builds on stage two and uses a multivariate linear regression method to create evaluation criteria and indicators of high-level behavioural capability. Six problems involving multiple classifications and two tasks using multiple labels are implemented using various neural network-based training strategies on the obtained data to assess the usefulness of the dataset for downstream tasks. To ensure that the data collected is genuine and useful, the experimental data must be analysed and expert knowledge must be included.

Theoretical Studies on the Insertion Reaction of Polar Olefinic Monomers Mediated by a Scandium Complex

This study aimed to investigate the insertion reaction of the polar monomers mediated by the cationic rare earth metal complex [(C5H5)Sc(NMe2CH2C6H4-o)]+ utilizing a combination of density functional theory (DFT) calculations and multivariate linear regression (MLR) methods. The chain initiation step of the insertion reaction could be described by the poisoning effect and the ease of monomer insertion, which could be represented via the DFT-calculated energy difference between σ- and π-coordination complexes (ΔΔE) and insertion energy barrier (ΔG≠), respectively. The results indicate that ΔΔE and ΔG≠ can be predicted by only several descriptors using multiple linear regression methods, with a root mean squared error (RMSE) of less than 2.5 kcal/mol. Furthermore, the qualitative analysis of the MLR models provided effective information on the key factors governing the insertion reaction chain initiation.

Statistical Downscaling of Remote Sensing Precipitation Estimates Using MODIS Cloud Properties Data over Northeastern Greece

The aim of this study is to spatially downscale the daily precipitation data from the Global Precipitation Measurement (GPM) mission, using the Integrated Multi-satellite Retrievals for GPM (IMERG), utilizing cloud properties from the Moderate Resolution Imaging Spectroradiometer (MODIS) instrument. Cloud optical thickness (COT), cloud effective radius (CER), and cloud water path (CWP) are used to statistically downscale IMERG precipitation estimates from 0.1 to 0.01° spatial resolution, using the Multivariate Linear Regression (MLR) and residual correction methods. The downscaled precipitation estimates were subsequently validated using in situ rain gauge measurements. The residual corrected IMERG downscaled precipitation estimates were found to be more accurate than the downscaled predicted precipitation without the implementation of the residual correction algorithm (up to 37%), with a respective decrease of the Root Mean Square Error (RMSE) (up to 75%), Normalized Root Mean Square Error (NRMSE) (up to 79%), and the Percent Bias (PB) (up to 98%). In addition, the final downscaled product after the MLR method implementation with residual correction was better correlated with the rain gauge observations than the initial IMERG product (up to 20%). Thus, the implementation of the MLR method in conjunction with the residual correction algorithm is an efficient tool for downscaling remote sensing products with a coarse spatial resolution.

Improved Autuoformer Electricity load forecasting based on model fusion

The development of the power system is undergoing rapid changes. The future direction of power system development is inevitably smart grids. This paper proposes an optimization of the Autoformer model based on model fusion to improve the accuracy of electricity load forecasting. By incorporating the predictive results of traditional statistical multivariate linear regression (MLR) and machine learning method LightGBM into the Autoformer model, we aim to fully leverage the strengths of both traditional statistical and machine learning approaches to enhance the predictive accuracy of the Autoformer model. Experimental results, using mean MAE and MSE as the evaluation metrics, demonstrate that the algorithm proposed in this paper exhibits better performance in the prediction results.

Short-Term Probabilistic Wind Speed Predictions Integrating Multivariate Linear Regression and Generative Adversarial Network Methods

The precise forecasting of wind speeds is critical to lessen the harmful impacts of wind fluctuations on power networks and aid in merging wind energy into the grid system. However, prior research has predominantly focused on point forecasts, often overlooking the uncertainties inherent in the prediction accuracy. For this research, we suggest a new approach for forecasting wind speed intervals (PI). Specifically, the actual wind speed series are initially procured, and the complete ensemble empirical mode decomposition coupled with adaptive noise (CEEMDAN) method decomposes the actual wind speed series into constituent numerous mode functions. Furthermore, a generative adversarial network (GAN) is utilized to achieve the wind speed PI in conjunction with the multivariate linear regression method. To confirm the effectiveness of the suggested model, four datasets are selected. The validation results suggest that this suggested model attains a superior PI accuracy compared with those of numerous benchmark techniques. In the context of PI of dataset 4, the PINAW values show improvements of 68.06% and 32.35% over the CEEMDAN-CNN and VMD-GRU values in single-step forecasting, respectively. In conclusion, the proposed model excels over the counterpart models by exhibiting diminished a PINAW and CWC, while maintaining a similar PICP.

Early-childhood body mass index and its association with the COVID-19 pandemic, containment measures and islet autoimmunity in children with increased risk for type 1 diabetes

Aims/hypothesisThe aim of this study was to determine whether BMI in early childhood was affected by the COVID-19 pandemic and containment measures, and whether it was associated with the risk for islet autoimmunity.MethodsBetween February 2018 and May 2023, data on BMI and islet autoimmunity were collected from 1050 children enrolled in the Primary Oral Insulin Trial, aged from 4.0 months to 5.5 years of age. The start of the COVID-19 pandemic was defined as 18 March 2020, and a stringency index was used to assess the stringency of containment measures. Islet autoimmunity was defined as either the development of persistent confirmed multiple islet autoantibodies, or the development of one or more islet autoantibodies and type 1 diabetes. Multivariate linear mixed-effect, linear and logistic regression methods were applied to assess the effect of the COVID-19 pandemic and the stringency index on early-childhood BMI measurements (BMI as a time-varying variable, BMI at 9 months of age and overweight risk at 9 months of age), and Cox proportional hazard models were used to assess the effect of BMI measurements on islet autoimmunity risk.ResultsThe COVID-19 pandemic was associated with increased time-varying BMI (β = 0.39; 95% CI 0.30, 0.47) and overweight risk at 9 months (β = 0.44; 95% CI 0.03, 0.84). During the COVID-19 pandemic, a higher stringency index was positively associated with time-varying BMI (β = 0.02; 95% CI 0.00, 0.04 per 10 units increase), BMI at 9 months (β = 0.13; 95% CI 0.01, 0.25) and overweight risk at 9 months (β = 0.23; 95% CI 0.03, 0.43). A higher age-corrected BMI and overweight risk at 9 months were associated with increased risk for developing islet autoimmunity up to 5.5 years of age (HR 1.16; 95% CI 1.01, 1.32 and HR 1.68, 95% CI 1.00, 2.82, respectively).Conclusions/interpretationEarly-childhood BMI increased during the COVID-19 pandemic, and was influenced by the level of restrictions during the pandemic. Controlling for the COVID-19 pandemic, elevated BMI during early childhood was associated with increased risk for childhood islet autoimmunity in children with genetic susceptibility to type 1 diabetes.Graphical

Research on Evaluation of Online Education Effectiveness Based on AI Teaching Aid System

Abstract In the age of technology, the application of intelligent assistive systems in teaching is getting more and more attention. After systematically describing the framework of the AI teaching assistance system and the process of online education effect evaluation, the study selects the course of an AI teaching assistance platform as a research case, collects the students’ learning behavior data, and applies the K-means-based clustering algorithm to divide them into clusters to obtain the learner portrait. The association rule algorithm based on Apriori is used to initially mine the connection between online learning behavior and educational effects. Furthermore, the multivariate linear regression method is applied to determine specific associations between online learning behaviors and educational effects to complete the assessment of online educational effects. Through the analysis, four types of learners were summarized: general (37.89%), negative (6.84%), interactive (9.48%), and positive (45.79%). Significant correlations between students’ online learning behaviors and educational outcomes were obtained, with regression equations between them. The effectiveness of online education is greatly influenced by students’ time commitment and learning motivation. Subsequent teaching can start from these two aspects to promote the improvement of the quality of teaching in online education.

Enhancement or suppression: a double-edged sword? Differential association of digital literacy with subjective health of older adult-evidence from China.

The emergence of an aging society and the digital age makes healthy aging a hot topic in Chinese society. This paper explores the associations between digital literacy and the subjective health of older adult individuals in PR China, offering insights that May assist policymakers and service providers in developing strategies and interventions suited to the digital era, potentially enhancing the healthy aging process for this demographic in China. This study utilized data from the China Longitudinal Aging Social Survey. Initially, demographic variables of 2086 individuals in the sample were analyzed. Subjective health differences among different populations and correlations between core variables were examined. Subsequently, multivariate linear regression and chain mediation methods were utilized to examine the relationships and potential pathways among the three dimensions of digital literacy and the subjective health of older adult individuals. (1) The subjective health status of older adult individuals in China was generally favorable, with an average score of 3.406 ± 0.764. (2) There was no direct correlation observed between the frequency of digital information use and the subjective health of the older adult (b = -0.032, p > 0.1). Digital entertainment information (b = 0.294, p > 0.1) did not show a significant effect, whereas life management information (b = 0.437, p < 0.01) demonstrated a positive association. Similarly, the use of smart healthcare devices (b = 0.842, p < 0.001) indicated a positive association (3) The frequency of digital information use indirectly enhanced the subjective health of the older adult through life management digital information and the use of smart healthcare devices, but had no indirect effect through entertainment and leisure digital information. Digital literacy is significantly correlated with the subjective health of the older adult, especially when they acquire life management information and utilize smart healthcare devices. However, a potential negative relationship is suggested between digital entertainment information and the subjective health of older adult individuals. Therefore, digital infrastructure should have prioritized the provision of high-quality, age-friendly digital applications for the older adult. This approach could have better harnessed the potential of digitalization to enhance health and well-being in older adults.

A remote sensing-based strategy for mapping potentially toxic elements of soils: Temporal-spatial-spectral covariates combined with random forest

The selection of predictor variables is a crucial issue in building a digital mapping model of potentially toxic elements (PTEs) in soil. Traditionally, the predictor variables for mapping models of soil PTEs have been chosen from sets of spatial parameters or spectral parameters derived from geographical environmental data. However, the enrichment of soil PTEs exhibits significant variations in both spatial and temporal dimensions, with the temporal dimension often being overlooked in the selection of predictor variables for digital mapping models. This limitation hampers the robustness and generalizability of the models. Therefore, multi-source geographical data were used in this study to determine three temporal indices for characterizing the enrichment process of soil PTEs in temporal dimensions, and additionally to construct the temporal-spatial-spectral (TSS) covariate combinations. The random forest (RF) algorithm was used to map soil PTEs at a regional scale. Results showed that: (1) When using spatial parameters or spectral parameters as predictor variables and measured Pb content as the dependent variable, the values of the model performance indicator RPIQ (ratio of performance to inter-quartile range) were 2.66 and 2.27, respectively. After incorporating the temporal parameters into the model input, values of RPIQ for the RF model reached 3.55 (using spatial-temporal covariates) and 3.21 (using spectral-temporal covariates), representing performance improvements of 33.46% and 41.41%, respectively. (2) The RF model constructed with the temporal-spatial-spectral covariates achieved satisfactory mapping accuracy (R2 = 0.85; RMSE = 0.80 mg kg−1; RPIQ = 4.09). (3) The soil Pb content in the western and northeastern regions was relatively high, while the remaining areas exhibited lower Pb levels, mainly due to industrial activities. (4) The mapping results of Pb obtained in this study were superior to other mapping methods, such as ordinary kriging, artificial neural networks, and multivariate linear regression methods. The soil PTE mapping technique employed in this study that combined TSS covariates with the RF provided an effective methodological approach for preventing soil pollution, controlling environmental risk, and improving soil management.

Dynamic Characteristics of Meteorological Drought and Its Impact on Vegetation in an Arid and Semi-Arid Region

Under the background of global climate warming, meteorological drought disasters have become increasingly frequent. Different vegetation types exhibit varying responses to drought, thus, exploring the heterogeneity of the impact of meteorological drought on vegetation is particularly important. In this study, we focused on Inner Mongolia (IM) as the research area and employed Standardized Precipitation Evapotranspiration Index (SPEI) and Vegetation Health Index (VHI) as meteorological drought and vegetation indices, respectively. The Breaks for Additive Seasons and Trend algorithm (BFAST) was utilized to reveal the dynamic characteristics of both meteorological drought and vegetation changes. Additionally, the Pixel-Based Trend Identification Method (PTIM) was employed to identify the trends of meteorological drought and vegetation during spring, summer, autumn, and the growing season. Subsequently, we analyzed the correlation between meteorological drought and vegetation growth. Finally, the response of vegetation growth to various climate factors was explored using the standardized multivariate linear regression method. The results indicated that: (1) During the study period, both SPEI and VHI exhibited a type of interrupted decrease. The meteorological drought was aggravated and the vegetation growth was decreased. (2) Deserts and grasslands exhibited higher sensitivity to meteorological drought compared to forests. The strongest correlation between SPEI-3 and VHI was observed in desert and grassland regions. In forest areas, the strongest correlation was found between SPEI-6 and VHI. (3) The r between severity of meteorological drought and status of vegetation growth was 0.898 (p &lt; 0.01). Vegetation exhibits a more pronounced response to short-term meteorological drought events. (4) Evapotranspiration is the primary climatic driving factor in the IM. The findings of this study provide valuable insights for the rational utilization of water resources, the formulation of effective irrigation and replenishment policies, and the mitigation of the adverse impacts of meteorological drought disasters on vegetation growth in the IM.

A systematic study of the performance of machine learning models on analyzing the association between semen quality and environmental pollutants

Human exposure to Phthalates, a family of chemicals primarily used to enhance the flexibility and durability of plastics, could lead to a decline in semen quality. Extensive studies have been performed to investigate the associations between semen quality and exposure to environmental pollutants, such as phthalates. However, these early studies mainly focus on using conventional statistical methods, such as simple and efficient multi-variable linear regression methods, to perform the analysis, which may not be effective in analyzing these complex multi-variable associations. Herein, we perform a systematic study of the performance of different machine learning methods in analyzing these associations. We will use data from a cohort of 1070 Chinese males from Hubei province who provided repeated urine samples to measure phthalate metabolites. In addition, phthalate metabolites in semen are also evaluated as a biomarker to give a more direct metric. We also incorporate patient demographics and administered medications into the analysis. Overall, six machine learning models, including linear and non-linear models, are implemented to analyze associations among thirty-one features and five metrics of the quality of the semen. The performance of the models is evaluated based on root-mean-square deviation through 10-fold cross-validation. Our investigations show that the performance of different models is varied when employed to study different metrics that represent the quality of the semen. Therefore, a systematic study of the patients’ data with various machine learning models is essential in improving the quantitative analysis in discovering the critical environmental pollutants that dictate the quality of semen. We hope this study could provide guidance of employing machine learning models in the future investigation of the impact of various pollutants on semen quality.

Multivariate piecewise linear regression model to predict radiosensitivity using the association with the genome-wide copy number variation.

The search for biomarkers to predict radiosensitivity is important not only to individualize radiotherapy of cancer patients but also to forecast radiation exposure risks. The aim of this study was to devise a machine-learning method to stratify radiosensitivity and to investigate its association with genome-wide copy number variations (CNVs) as markers of sensitivity to ionizing radiation. We used the Affymetrix CytoScan HD microarrays to survey common CNVs in 129 fibroblast cell strains. Radiosensitivity was measured by the surviving fraction at 2 Gy (SF2). We applied a dynamic programming (DP) algorithm to create a piecewise (segmented) multivariate linear regression model predicting SF2 and to identify SF2 segment-related distinctive CNVs. SF2 ranged between 0.1384 and 0.4860 (mean=0.3273 The DP algorithm provided optimal segmentation by defining batches of radio-sensitive (RS), normally-sensitive (NS), and radio-resistant (RR) responders. The weighted mean relative errors (MRE) decreased with increasing the segments' number. The borders of the utmost segments have stabilized after partitioning SF2 into 5 subranges. The 5-segment model associated C-3SFBP marker with the most-RS and C-7IUVU marker with the most-RR cell strains. Both markers were mapped to gene regions (MCC and SLC1A6, respectively). In addition, C-3SFBP marker is also located in enhancer and multiple binding motifs. Moreover, for most CNVs significantly correlated with SF2, the radiosensitivity increased with the copy-number decrease.In conclusion, the DP-based piecewise multivariate linear regression method helps narrow the set of CNV markers from the whole radiosensitivity range to the smaller intervals of interest. Notably, SF2 partitioning not only improves the SF2 estimation but also provides distinctive markers. Ultimately, segment-related markers can be used, potentially with tissues' specific factors or other clinical data, to identify radiotherapy patients who are most RS and require reduced doses to avoid complications and the most RR eligible for dose escalation to improve outcomes.

Aquaculture 4.0: hybrid neural network multivariate water quality parameters forecasting model

This study examined the efficiency of hybrid deep neural network and multivariate water quality forecasting model in aquaculture ecosystem. Accurate forecasting of critical water quality parameters can allow for timely identification of possible problem areas and enable decision-makers to take pre-emptive remedial actions that can significantly improve water quality management in aquaculture industry. A novel hybrid deep learning neural network multivariate water quality parameters forecasting model is developed with the aid of ensemble empirical mode decomposition (EEMD) method, deep learning long-short term memory (LSTM) neural network (NN), and multivariate linear regression (MLR) method. The presented water quality forecasting model (shortened as EEMD–MLR–LSTM NN model) is developed using multivariate time-series water quality sensor data collected from Loch Duart company, a Salmon offshore aquaculture farm based around Scourie, northwest Scotland. The performance of the novel hybrid water quality forecasting model is validated by comparing the forecast result with measured water quality parameters data and the real Phytoplankton data count from the aquaculture farm. The forecast accuracy of the results suggests that the novel hybrid water quality forecasting model can be used as a valuable support tool for water quality management in aquaculture industries.

A novel correlation models for predicting stress intensity factor of semi-elliptical crack in welded pipelines under cyclic internal pressure based on Bézier base XIGA

Pipelines are strategic infrastructures that are used for energy transfer in the world. Cracks in the residual stress field are the common defects in welded pipelines. These defects with cyclic internal pressure cause crack growth and decrease the fatigue life of pipelines. In this study, a novel correlation model is proposed to predict the stress intensity factor (SIF) of the semi-elliptical crack front in welded steel pipeline subjected to cyclic internal pressure. For this purpose, the Bézier extraction base extended isogeometric method is applied for numerical analysis and preparing a database of results based on the thermo elastic-plastic equation. Taguchi standard L25 orthogonal array was used as the analysis design to create the simulations. New correlations were proposed for SIF prediction using the multivariate linear regression method. These correlations were obtained based on a parametric study that included three parameters. These parameters are dimensionless crack geometry (a/c), pipe geometry (D/t), and loading (P/Pb) or (σres/E). The R-Square (R2) values of the presented model for the welding residual stress and cyclic internal pressure were obtained as 0.989 and 0.992, respectively. The results of FEM analysis with similar conditions were compared to the results of XIGA and correlation models to validate the presented method. The maximum difference between them is approximately 3.71% for the cyclic internal pressure, and 2.94% for the welding residual stress. RMSD for the cyclic internal pressure model is 0.49, and for the welding residual stress model is 0.51. Therefore, the proposed correlation models have acceptable accuracy in predicting the SIF and fatigue crack propagation rate of the semi-elliptical crack in the steel pipes.

A Comparative Analysis of the Full and Short Versions of the Social Responsiveness Scale in Estimating an Established Autism Risk Factor Association in ECHO: Do we Get the Same Estimates?

Prior work developed a shortened 16-item version of the Social Responsiveness Scale (SRS), a quantitative measure of social communication and autism spectrum disorder (ASD)-related traits. However, its properties for use in risk factor estimation have not been fully tested compared to the full SRS. We compared the associations between gestational age (previously established risk factor for ASD) and the 65-item "full" and 16-item "short" versions of the SRS to test the shortened version's ability to capture associations in epidemiologic analyses of ASD risk factors. We used data from participants in the Environmental influences on Child Health Outcomes (ECHO) Program (n = 2,760). SRS scores were collected via maternal/caregiver report when children were aged 2.5-18 years. We compared estimates of associations between gestational age and preterm birth between the full and short SRS using multivariable linear regression, quantile regression, and prediction methods. Overall, associations based on full and short SRS scores were highly comparable. For example, we observed positive associations between preterm birth with both full ([Formula: see text]=2.8; 95% CI [1.7, 4.0]) and short ([Formula: see text]=2.9; 95% CI [1.6, 4.3]) SRS scores. Quantile regression analyses indicated similar direction and magnitude of associations across the distribution of SRS scores between gestational age with both short and full SRS scores. The comparability in estimates obtained for full and short SRS scores with an "established" ASD risk factor suggests ability of the shortened SRS in assessing associations with potential ASD-related risk factors and has implications for large-scale research studies seeking to reduce participant burden.

Source Aware Deep Learning Framework for Hand Kinematic Reconstruction Using EEG Signal.

The ability to reconstruct the kinematic parameters of hand movement using noninvasive electroencephalography (EEG) is essential for strength and endurance augmentation using exoskeleton/exosuit. For system development, the conventional classification-based brain-computer interface (BCI) controls external devices by providing discrete control signals to the actuator. A continuous kinematic reconstruction from EEG signal is better suited for practical BCI applications. The state-of-the-art multivariable linear regression (mLR) method provides a continuous estimate of hand kinematics, achieving a maximum correlation of up to 0.67 between the measured and the estimated hand trajectory. In this work, three novel source aware deep learning models are proposed for motion trajectory prediction (MTP). In particular, multilayer perceptron (MLP), convolutional neural network-long short-term memory (CNN-LSTM), and wavelet packet decomposition (WPD) for CNN-LSTM are presented. In addition, novelty in the work includes the utilization of brain source localization (BSL) [using standardized low-resolution brain electromagnetic tomography (sLORETA)] for the reliable decoding of motor intention. The information is utilized for channel selection and accurate EEG time segment selection. The performance of the proposed models is compared with the traditionally utilized mLR technique on the reach, grasp, and lift (GAL) dataset. The effectiveness of the proposed framework is established using the Pearson correlation coefficient (PCC) and trajectory analysis. A significant improvement in the correlation coefficient is observed when compared with the state-of-the-art mLR model. Our work bridges the gap between the control and the actuator block, enabling real-time BCI implementation.

Exchanging screen for non-screen sitting time or physical activity might attenuate depression and anxiety: A cross-sectional isotemporal analysis during early pandemics in South America.

To examine the theoretical substitutions of screen exposure, non-screen sitting time, moderate and vigorous physical activity with depressive and anxiety symptoms in South American adults during the COVID-19 pandemic. A cross-sectional study during the first months of the COVID-19 pandemic with data from 1981 adults from Chile, Argentina, and Brazil. Depressive and anxiety symptoms were assessed using the Beck Depression and Anxiety Inventories. Participants also reported physical activity, sitting time, screen exposure, sociodemographic, and tobacco use data. Isotemporal substitution models were created using multivariable linear regression methods. Vigorous physical activity, moderate physical activity, and screen exposure were independently associated with depression and anxiety symptoms. In adjusted isotemporal substitution models, replacing 10 min/day of either screen exposure or non-screen sitting time with any intensity of physical activity was associated with lower levels of depressive symptoms. Improvements in anxiety symptoms were found when reallocating either screen exposure or non-screen sitting time to moderate physical activity. Furthermore, replacing 10 min/day of screen exposure with non-screen sitting time was beneficially associated with anxiety (B = -0.033; 95 % CI = -0.059, -0.006) and depression (B = -0.026; 95 % CI = -0.050, -0.002). Replacement of screen exposure with any intensity of physical activity or non-screen sitting time could improve mental health symptoms. Strategies aiming to reduce depressive and anxiety symptoms highlight physical activity promotion. However, future interventions should explore specific sedentary behaviors as some will relate positively while others negatively.