Extended Multiple Cross-Component Linear Models With Adaptive Thresholding and Overlapped Averaging Beyond VVC

Smart grids are vulnerable to cyber-attacks, which can cause significant damage and huge economic losses. Generally, state estimation (SE) is used to observe the operation of the grid. State estimation of the grid is vulnerable to false data injection attack (FDIA), so diagnosing this type of malicious attack has a major impact on ensuring reliable operation of the power system. In this paper, we present an effective FDIA detection method based on residual recurrent neural network (R2N2) prediction model and adaptive judgment threshold. Specifically, considering the data contains both linear and nonlinear components, the R2N2 model divides the prediction process into two parts: the first part uses the linear model to fit the state data; the second part predicts the nonlinearity of the residuals of the linear prediction model. The adaptive judgment threshold is inferred through fitting the Weibull distribution with the sum of squared errors between the predicted values and observed values. The thorough simulation results demonstrate that our scheme performs better than other prediction based FDIA detection schemes.

The Yellow River Basin (YRB) plays a pivotal role in the water resources management of its region, significantly influenced by the interplay between climate change and human activities, particularly in its upper and middle reaches (UMRYR). This study aims to elucidate the evolving patterns and determinants of runoff within the UMRYR, a matter of considerable importance for the basin’s water resource management, strategy, and distribution. Utilizing the Google Earth Engine (GEE) platform, this research accessed comprehensive datasets including precipitation, drought index, and terrace area, among others, to examine their effects on runoff variations at five gauge stations across the YRB. Terrace data was extracted from Landsat imagery via the Random Forest Model, while annual runoff figures from 1990 to 2020 were sourced from the Sediment Bulletin of China River. Employing the Mann-Kendall test, we assessed the temporal changes in runoff over three decades. In addition, runoff drivers were analyzed by stepwise regression and redundancy analysis, leading to the construction of a multiple linear regression model. The accuracy of predicting annual runoff using the multiple linear model was verified through cross-validation and comparison with the ARIMA time series model. Our findings reveal the efficacy of the random forest algorithm in classifying terraces, achieving an accuracy rate exceeding 0.8. The period from 1990 to 2020 saw a general increase in annual runoff across the five gauging stations in the UMRYR, albeit with variations in the pattern, particularly at the Tangnaihai gauge station which presented the most complex changes. Crucially, three main drivers—summer precipitation (SP), terrace area (TR), and drought index (DI)—were identified as significant predictors in the regression models. The multiple linear regression model outperformed the ARIMA model in forecasting accuracy, underlining the significance of integrating these drivers into runoff prediction models for the UMRYR.

The quality of certain plants is considered to be a key factor affecting the food habitat or migration of some herbivorous species, and, thus, to estimate the spatial and temporal variation of plant quality is crucial for understanding the grazing and migrating behaviours of these herbivores. This study aimed to explore the possibilities of estimating plant protein and phosphorus contents, with the laboratory-based hyperspectral measurements of fresh Carex leaves, which are the main food source of many wintering bird species in Poyang Lake, China. Fifty-four Carex leaf samples were collected, and their hyperspectral reflectance (at 350–2500 nm) and crude protein and phosphorus contents were measured in the laboratory. The successive projections algorithm (SPA) was applied for spectral dimension reduction, and a multiple linear regression model was calibrated to estimate the crude protein and phosphorus contents from the wavelengths selected with the SPA. The model validation results showed that the root mean square errors (RMSEs) of estimation were 2.51% for crude protein and 0.06% for phosphorus. Compared with a multiple linear model with randomly selected inputs and full-spectrum partial least-square regression (PLSR), the multiple linear regression model combined with the SPA method exhibited a significant advantage in terms of accuracy in estimating the crude protein and phosphorus contents of Carex leaves.

Monitoring cow body weight is crucial to support farm management decisions due to its direct relationship with the growth, nutritional status, and health of dairy cows. Cow body weight is a repeated trait, however, the majority of previous body weight prediction research only used data collected at a single point in time. Furthermore, the utility of deep learning-based segmentation for body weight prediction using videos remains unanswered. Therefore, the objectives of this study were to predict cow body weight from repeatedly measured video data, to compare the performance of the thresholding and Mask R-CNN deep learning approaches, to evaluate the predictive ability of body weight regression models, and to promote open science in the animal science community by releasing the source code for video-based body weight prediction. An Intel RealSense D435 camera was installed on Virginia Tech dairy complex to collect top-view videos of 10 lactating Holstein cows twice a day for 28 days after milking sessions and 2 non-lactating, pregnant Jersey cows once a week upon calving. At the same time, the ground truth body weight records were collected using a walk-over weighing system. A total of 40,405 depth images and depth map files were obtained. Three approaches were investigated to segment the cow's body from the background, including single thresholding, adaptive thresholding, and Mask R-CNN. Four image-derived biometric features, such as dorsal length, abdominal width, height, and volume, were estimated from the segmented images and fitted using ordinary least squares, ridge regression, least absolute shrinkage and selection operator, and linear mixed models. Two cross-validation designs, forecasting and leave-three-cows-out, were used to evaluate prediction performance. The Pearson correlation between image-derived biometric features and scale-based body weight ranged from 0.74 to 0.95. On average, the Mask-RCNN approach combined with a linear mixed model resulted in the best prediction coefficient of determination and mean absolute percentage error of 0.98 and 2.03%, respectively, in the forecasting cross-validation. The Mask-RCNN approach was also the best in the leave-three-cows-out cross-validation, followed by adaptive and single thresholding. The prediction coefficients of determination and mean absolute percentage error of the Mask-RCNN coupled with the linear mixed model were 0.90 and 4.70%, respectively. Our results suggest that deep learning-based segmentation improves the prediction performance of cow body weight from longitudinal depth video data.

A new invariant method for representation or description for 3D object is proposed using multiple linear model. This method consists of extracting vector invariant using the multiple linear parameter model applied to the 3D object in question, it's invariant against affine transformation of the object. The concerned 3D objects are transformations of 3D objects by one element of the overall transformation. The set of transformations considered in this work is the general affine group.

Correlation of serum biochemical parameters and saliva pH in healthy individuals

The article is about a problem of mathematical modeling of the natural resource potential of the Perm Territory by 1st and 2nd order finite-difference models. Such models can obtain better forecasts of complex socio-economic processes in comparison with the traditionally used linear multiple regression models. A high quality model of the natural resource potential with forecast possibi¬lities is one of the necessary conditions for the effective management of the natural resources of the region in order to ensure its sustainable economic development. Purpose of work. Aim of this work is work construction of finite-difference models of a natural resource potential complex indicators and an assessment of their prognostic properties. Materials and methods. Our research is based on Perm region statistical data for the period from 2001 to 2018. A multiple linear regression model is used as a comparison base. The natural resource potential complex indicator is calculated as a weighted sum of particular criteria characterizing the natural resources of the region. First and second order finite difference models are obtained by adding autoregressive terms of the first and second orders, respectively, to the multiple linear regression model. An estimation of the unknown parameters of the equations is carried out by a modified least squares method, which preserves the signs of the coefficients with the factors the same as in the original linear model. At the same time, the selection of explanatory factors and the assessment of the quality of the models are carried out based on the accuracy of the predicted values of the studied indicator. The results of the study. Components and factors of the natural resource potential is obtained, and a procedure for constructing finite-difference models is performed for three different time intervals: 2001–2018, 2001–2008, and 2008–2018. These intervals are chooseen because changes in the methodology for generating statistical data nearly 2008. Discussion and conclusions. The number of calculated predicted values was 18, and only in 4 out of 18 cases (22,2%) their quality is worse than forecasts obtained by the linear multiple model. So proposed modification of the multiple linear regression model with the addition of autoregressive terms makes it possible to improve the forecasting quality of the complex indicator of the natural resource potential of the region and, therefore, to make more effective decisions when managing its level.

Netflix is one of the world's leading entertainment services, with millions of subscribers all over the world. To make a prediction of the expected number of future subscribers is a meaningful and valuable research topic. However, there is no authoritative model to help make predictions. Therefore, this study will explore the relationship between possible factors and the number of future subscribers in terms of the data from Netflix over the years. Subsequently, some candidate multiple linear regression models are constructed and the “best” model is selected to help make predictions. The final model shows that the number of subscribers is related to four variables, i.e., a negative relationship with the price of the basic plan, a positive relationship with the price of the standard plan, the number of countries where Netflix is available and the medium level of annual world income. The data on the number of subscribers over the years shows an increase in subscribers every year, as well as the amount of growth varies from year to year. In other words, the increase in the price of the basic plan may lead to a decrease in the number of subscribers, while the increase in the price of the standard plan, in the number of countries where Netflix is available and in the medium level of annual world income may lead to an incline in it. These results shed light on guiding further exploration of having a practical method to predict the number of future subscribers for Netflix.

BackgroundBlood pressure (BP) variability is associated with target organ damage in hypertension and diabetes. The 24 h ambulatory blood pressure monitoring (24 h-ABPM) has been proposed as an evaluation for BP variability using several indexes [standard deviation (SD) of mean BP, coefficient of variation (CV), BP variation over time (time-rate index)].MethodsWe evaluated the association between BP variability measured by 24 h-ABPM indexes and echocardiographic variables in a cross-sectional study in 305 diabetic-hypertensive patients.ResultsTwo groups were defined by the median (0.55 mmHg/min) of time-rate systolic BP (SBP) index and classified as low or high variability. Age was 57.3 ± 6.2 years, 196 (64.3 %) were female. Diabetes duration was 10.0 (5.0–16.2) years, HbA1c was 8.2 ± 1.9 %. Baseline clinical characteristics were similar between low (n = 148) and high (n = 157) variability groups. Office SBP and systolic 24 h-ABPM were higher in the high variability group (139.9 mmHg vs 146.0 mmHg, P = 0.006; 128.3 mmHg vs 132.9 mmHg, P = 0.019, respectively). Time-rate index, SD and CV of SBP, were higher in high variability group (P < 0.001; P < 0.001 and P = 0.003, respectively). Time-rate index was not independently associated with the echocardiography’s variables in multiple linear model when adjusting for age, 24 h-ABPM, diabetes duration and HbA1c. The multiple linear regression model revealed that the significant and independent determinants for septum thickness, relative wall thickness and posterior wall thickness (parameters of left ventricular hypertrophy) were: age (p = 0.025; p = 0.010; p = 0.032, respectively) and 24 h-SBP (p < 0.001 in the three parameters).ConclusionBP variability estimated by 24 h-ABPM is not independently associated with echocardiographic parameters in diabetic-hypertensive patients.

Artificial Neural Networks are proposed to model and predict electricity consumption of China. Multilayer perceptron with backpropagation training algorithm is used as the neural network topology. Tangent-sigmoid and pure-linear transfer functions are selected in the hidden and output layer processing elements, respectively. Energy demand is modeled as a function of economic indicators such as population, gross national product, imports and exports. Performance comparison among the ANN and multiple linear models is made based on absolute mean square error. The ANN model also has much better RMSE for the forecasted energy demands than multiple linear regressions. Energy demand of China is predicted until 2050 using data from 1990 to 2008 along with other economic indicators. The results show that energy demands would peak during 2026-2035 and decrease gradually. This decreasing trend of energy demand appears very unique because all independent variables are increasing. It is estimated that the ANN model considered the high-nonlinearity of the data, which is not easily detectable using multiple linear regression model.

This study examines gender-based differences in financial behavior, focusing on various bank transactions and demographic factors using a large dataset from Kaggle, consisting of over one million entries. By applying multiple linear regression models and permutation feature importance rankings, the study explores how different variables, such as age, geographic location, and transactional timing, influence banking patterns across genders. The findings reveal that age is a significant predictor of transaction behavior for males, while it is less impactful for females. Geographic location, particularly “Location_West” and “Location_Other,” plays a crucial role for males but has minimal influence on females. For females, transactional timing, indicated by “TransactionHour,” shows more importance in predicting banking behaviors. In contrast, gender itself does not significantly affect transaction outcomes when controlling for other variables. Overall, the study highlights the importance of demographic and contextual factors, such as age and geography, over inherent genderbased differences. These insights provide valuable guidance for financial institutions aiming to tailor their services more effectively to meet the needs of diverse customer segments. The results emphasize the need for data-driven approaches to better understand gender-specific financial behavior and improve service offerings.

Background: The advancement in pediatric cancer treatment led to an increased five-year survival. Symptoms develop due to cancer and its treatment during the treatment phase, but they can persist even after treatment completion. Cancer patients experience multiple interrelated symptoms known as symptom clusters. Compared with a single symptom, symptom clusters synergistically affect quality of life (QoL). Limited studies in pediatric cancer have assessed the association between symptom clusters and quality of life, but none were assessed after the treatment completion. Objective: To find the association between symptom clusters and QoL among pediatric cancer survivors after treatment completion. Method: We conducted a cross-sectional study of 136 pediatric cancer survivors aged 10-25 who completed treatment within eight years from 2 major hospitals in Nebraska. We used a survey to collect data during their hospital visit, the validated Memorial Symptom Assessment Scale tool to collect information on symptoms, and the PedsQoL instrument to measure the quality of life. We used exploratory factor analysis to derive symptom clusters. We conducted a multiple linear regression model to find the association between symptom clusters and quality of life at a significance level with a p-value of less than 0.05 and an adjusted beta coefficient. Results: The mean age was 14.91, the mean year since diagnosis and treatment completion was 4.66 and 3.05, respectively, and the mean number of symptoms was 6.06. Difficulty concentrating (48.53%), lack of energy (45.59%), and feeling drowsy (45.59%) were the most common symptoms. Around 80% of participants experienced more than one symptom. The mean overall quality of life score was 80.84, the psychosocial score was 79.84, and the physical score was 82.72 out of 100. The exploratory analysis derived four symptom clusters: Cluster 1: dry mouth, lack of energy, feeling drowsy, feeling irritable, and difficulty sleeping; Cluster 2: feeling nervous, worrying, feeling sad, and difficulty concentrating; Cluster 3: nausea and lack of appetite; and Cluster 4: pain, cough, numbness in hands/feet, dizziness and sweats. The multiple linear models were adjusted for possible confounders and sociodemographic and clinical characteristics. Cluster 1 (β= -4.74, 95% CI: 7.32, 2.15) and Cluster 2 (β= -3.39, 95% CI: 5.80, -0.98) were associated with lower overall quality of life. Cluster 1 (β= -3.96, 95% CI: -7.47, -0.44) was associated with lower physical quality of life. Cluster 1 (β= -5.15, 95% CI: -7.83, -2.47) and Cluster 2 (β= -4.64, 95% CI: -7.13, -2.14) were associated with lower psychosocial quality of life. Conclusion: Even after treatment completion, symptom clusters are prevalent in pediatric cancer survivors. Among the four clusters identified, Cluster 1 and Cluster 2, which include mainly psychological symptoms, affect all dimensions of quality of life. Citation Format: Krishtee Napit, Don Coulter, Katrina Cordts, Daisy Dai, Evi Farazi, Shinobu Watanabe-Galloway. Impact of symptom clusters on quality of life in pediatric cancer survivors after treatment completion [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 706.

Control of a solution copolymerization reactor using multi-model predictive control

Whether human immunodeficiency virus (HIV) seroconverters have been presenting with progressively lower CD4 cell counts over the course of the HIV epidemic is controversial. Additional data on whether HIV might have become more virulent on a population level (measured by post-seroconversion CD4 cell counts) may provide important insights regarding HIV pathogenesis. To determine whether post-seroconversion CD4 cell counts have changed over time, we evaluated 2174 HIV seroconverters as part of a large cohort study during the period 1985-2007. Participants were documented antiretroviral-naive HIV seroconverters who had a CD4 cell count measured within 6 months after receiving a diagnosis of HIV infection. Multiple linear regression models were used to assess trends in initial CD4 cell counts. The mean initial CD4 cell count decreased during the study period from 632 cells/mm(3) in 1985-1990 to 553 cells/mm(3) in 1991-1995, 493 cells/mm(3) in 1996-2001, and 514 cells/mm(3) in 2002-2007. During those periods, the percentages of seroconverters with an initial CD4 cell count <350 cells/mm(3) were 12%, 21%, 26%, and 25%, respectively. In the multiple linear model, the mean decrease in CD4 cell count from 1985-1990 was 65 cells/mm(3) in 1991-1995 (P < .001)), 107 cells/mm(3) in 1996-2001 (P < .001), and 102 cells/mm(3) in 2002-2007 (P < .001). Similar trends occurred with regard to CD4 cell percentage and total lymphocyte count. Similar decreases in initial CD4 cell counts were observed among African American and white persons during the epidemic. A significant decrease in initial CD4 cell counts among HIV seroconverters in the United States has occurred during the HIV epidemic. These data provide an important clinical correlate to suggestions that HIV may have adapted to the host, resulting in a more virulent infection.

Associations between polybrominated diphenyl ethers (PBDEs) levels in adipose tissues and blood lipids in women of Shantou, China