Root Mean Research Articles

Optimized LSTM Model for Accurate Blood Glucose Prediction in Type-1 Diabetes

Objective: Blood glucose levels in diabetes are influenced by various factors, making prediction challenging when considering many variables. Therefore, this study proposes an optimized LSTM model to improve the prediction accuracy of blood glucose using a single variable (CGM), thereby simplifying diabetes management. Methods: This study has implemented the LSTM model to predict blood glucose levels in type-1 diabetics for a 60-minute prediction horizon. The grid search technique has been implemented to fine-tune the hyperparameters of the LSTM model. Ohio Datasets were used for this study. Root mean squared error (RMSE) and error grid analysis (EGA) were used for the evaluation of model performance. Findings: The proposed study reveals 26.13 ± 3.25 mg/dl predicting accuracy for a 60-minute prediction horizon. Clark Error grid analysis (CEGA) was used to validate clinical acceptance of the proposed model, and the results provided higher than 97% prediction accuracy in the Ohio dataset. Novelty: The patient-specific optimized model has outperformed the conventional LSTM model. The results show a significant improvement in model performance compared to previous work in terms of root mean square error (RMSE). Keywords: Long-short term memory (LSTM), Grid search optimization technique, Continuous Glucose monitors (CGM), Blood Glucose, Type-1 Diabetes

Comparative analysis of different PV technologies under the tropical environments

Abstract In this paper, six different types of solar PV technologies are compared in terms of their performances under tropical conditions, using three years of performance data from a 1.2 MW experimental solar farm. The technologies considered include single-crystalline silicon, polycrystalline silicon, microcrystalline silicon, amorphous silicon, copper indium selenium (CIS), and hetero-junction with intrinsic thin layer (HIT). The field performances of these cells were initially assessed using standard performance indices such as Array Yield, Reference Yield, Capture Loss, Performance Ratio, and Efficiency Ratio. Among the technologies studied, amorphous silicon and HIT-based systems demonstrated better performance, showing higher Performance and Efficiency Ratios, along with lower capture losses. This study also modelled the fluctuations in power production from these panels. Under probabilistic modeling, the ramping behavior of the systems was characterized using the Generalized Logistic Distribution. Based on this analysis, CIS PV systems were found to have minimum power ramps, where as the HIT based systems showed the highest power fluctuations. To predict minute-wise and hourly ramping of the PV systems under varying levels of solar insolation, machine learning methods based on Artificial Neural Networks (ANN), Support Vector Machines (SVM), and k-Nearest Neighbors (kNN) were developed. With a Normalized Root Mean Square Error (NRMSE) of over 96%, these models demonstrated high accuracy in capturing the ramping characteristics of the studied PV systems. The results of this study offer valuable insights into the performance of different PV systems under tropical regions, which can be used in efficiently designing and managing solar PV projects.

Effect of deposition rate on micromorphology analyses and optical parameters in amorphous carbon nickel thin films

In this work, the micromorphology of the Ni @ amorphous carbon films and their relationship to other optical properties using atomic force microscopic (AFM) imaging were examined. The surface topography of these Ni @ amorphous carbon sheets was reported using stereometric analysis and the SPIPTM 6.7.4 software in compliance with ISO 25178-2:2012 and ASME B46.1-2009. It is clear that the Ni @ amorphous carbon films deposited at 180 s had more the root mean square height (Sq) at around of 0.1475 mm and has the greatest value in comparison to other films. The Ni @ amorphous carbon films deposited at 600 s had a more regular surface because their the root mean square height (Sq) had minimum value of 0.1360 mm. The peak energy positions of optical density for the films deposited at 180 s which were the lowest value and about of 1.65 eV, could be due the quasi-metallic mode. The cut-off energy for the skin/shell depth parameters of the Ni @ amorphous carbon films were about 3.5 eV/ or 353 nm. The Ni @ amorphous carbon films deposited at 180 s had the lowest value of the electron - phonon interaction energy, (Ee-p). The small polaron model shown that as input photon energy was increased, the calculated AC conductivity by optical data for the Ni @ amorphous carbon films was decreased.

A deep learning framework for virtual continuous glucose monitoring and glucose prediction based on life-log data

While continuous glucose monitoring (CGM) has revolutionized metabolic health management, widespread adoption remains limited by cost constraints and usage burden, often resulting in interrupted monitoring periods. We propose a deep learning framework for glucose level inference that operates independently of prior glucose measurements, utilizing comprehensive life-log data. The model employs a bidirectional Long Short-Term Memory (LSTM) network with an encoder-decoder architecture, incorporating dual attention mechanisms for temporal and feature importance. The system was trained on data from 171 healthy adults, encompassing detailed records of dietary intake, physical activity metrics, and glucose measurements. The encoder’s hidden state as latent representations were analyzed for distributions of patterns of glucose and life-log sequences. The model showed a 19.49 ± 5.42 (mg/dL) in Root Mean Squared Error, 0.43 ± 0.2 in correlation coefficient, and 12.34 ± 3.11 (%) in Mean Absolute Percentage Eror for current glucose level predictions without any information of glucose at the inference step. The distribution of latent representations from the encoder showed the potential differentiation for glucose patterns. The model’s ability to maintain predictive accuracy during periods of CGM unavailability has the potential to support intermittent monitoring scenarios for users.

Investigating the Interaction Mechanism of CAT-BT-Br and Key Residue Mutations for Castration-Resistant Prostate Cancer through Molecular Dynamics Simulation.

Prostate cancer is the second most common cancer in men, second only to lung cancer. Castration-resistant prostate cancer (CRPC) was formerly known as hormone-resistant prostate cancer. The aim of this study is to reveal the effect of key residue mutations on the binding mechanism between catalase (CAT) and benzaldehyde thiourea derivatives (BT-Br), providing theoretical support for the development of novel CAT inhibitors. This article analyzes the structural stability, binding energy and decomposition, hydrogen bonding, etc. of wild-type (WT) and multiple mutations systems. The results showed that, in addition to the R203A mutant, all mutation systems significantly enhanced the binding ability of CAT to BT-Br, and their binding free energy contribution mainly came from van der Waals interactions. Hydrogen bond analysis shows that the hydrogen bond occupancy rate of the WT system is relatively low, while mutations such as V302A have a hydrogen bond occupancy rate as high as 93.05%, indicating a significant enhancement in their binding ability. In addition, mutations have limited impact on the overall stability of proteins, but some mutations such as Y215A and V302A significantly alter the binding site and direction of proteins. The results of principal component analysis (PCA) in other systems are consistent with those of root mean square fluctuation (RMSF) analysis, and the binding site shows little movement. This study not only elucidates the microscopic effects of key residue mutations on the binding mechanism between CAT and BT-Br but also provides new targets and drug design ideas for prostate cancer treatment based on iron death induction strategies.

Arrhythmic burden of long-term postsurgical hypoparathyroidism assessed with 24-hour Holter ECG.

Postsurgical hypoparathyroidism is associated with an increased cardiovascular risk, however an increased arrhythmic risk in this population is controversial. Twenty-two postmenopausal women with postsurgical hypoparathyroidism and 22 healthy postmenopausal women of the same age were enrolled. Each subject underwent blood tests, standard 12-lead ECG, 24-hour Holter ECG and echocardiographic measurements. The exclusion criteria were: previous cardiovascular disease, arrhythmias, diabetes mellitus, kidney failure, use of drugs that may interfere with cardiac conduction. Time since diagnosis of postsurgical hypoparathyroidism was 19.33 ± 8.82 years. As expected, serum calcium and PTH levels were significantly lower in hypoparathyroid patients compared to controls, while phosphorus was higher (all p < 0.05). ECG parameters were within normal values in both groups. A higher number of hypoparathyroid patients had significantly more supraventricular and ventricular premature beats compared to controls (p < 0.05). An index of heart variability, predominantly of parasympathetic activity, such as the root mean square of the difference between successive RR intervals was significantly lower in hypoparathyroid patients compared to controls (21.09 ± 9.84 vs 33.0 ± 17.89 msec, p = 0.009). All the echocardiographic parameters were within the normal limits. The only statistically significant difference between groups was a lower ejection fraction (EF) in the hypoparathyroid group compared to controls (57.5% ± 2.98 vs 64.85% ± 6.09, p < 0.0001). A longer time since diagnosis of hypoparathyroidism was only positively associated with heart rate: B = 0.45, QRS: B = 0.66 and negatively with EF:B = -0.64 (all p < 0.05). 24-hour ECG Holter demonstrated an increase in arrhythmia in postmenopausal women with long-term postsurgical hypoparathyroidism in the absence of long QT or cardiac structural abnormalities.

Predicting Occupant Annoyance in Acoustic-Thermal Compound Environments

With heavy trucks being more widely used in the logistics industry, more and more lorry drivers are frequently exposed to the acoustic-thermal dynamically coupled cockpit environment for a long time. The comfort in the cockpit directly affects driving safety and occupational health. However, the existing research lacks a multi-parameter fusion prediction method for occupant annoyance in this scenario. In this paper, we studied the effect of an acoustic-thermal composite environment on the annoyance level of truck occupants and predicted the annoyance level of the human body by combining environmental parameters and physiological parameters. A total of 20 adult males participated in the subjective annoyance evaluation test, and 60 sets of sample data were obtained under four working conditions by collecting environmental parameters and monitoring physiological parameters, and the effect of acoustic-thermal composite environments was explored using statistical analysis in combination with the subjects’ annoyance polls. The results showed that the human physiological parameters were significantly correlated with the thermal environment, and the correlation coefficient between PMV value and skin temperature was r1 = 0.99, with p &lt; 0.05. The subjective annoyance level was more sensitive to the thermal environment than noise. The correlation coefficient between PMV and annoyance level was r2 = 0.931, and the correlation coefficient between the noise parameter roughness R and annoyance level was r3 = 0.545. The results of this study were based on the screened predictor variables, the annoyance prediction model using the random forest algorithm showed high accuracy on the test set (R2 = 0.941, root mean square error RMSE = 0.259, mean absolute error MAE = 0.201). The study showed that the annoyance prediction model incorporating environmental and physiological parameters could estimate subjects’ annoyance more accurately.

A two-stage forecasting model using random forest subset-based feature selection and BiGRU with attention mechanism: Application to stock indices.

The heteroscedastic and volatile characteristics of stock price data have attracted the interest of researchers from various disciplines, particularly in the realm of price forecasting. The stock market's non-stationary and volatile nature, driven by complex interrelationships among financial assets, economic developments, and market participants, poses significant challenges for accurate forecasting. This research aims to develop a robust forecasting model to improve the accuracy and reliability of stock price predictions using machine learning. A two-stage forecasting model is introduced. First, a random forest subset-based (RFS) feature selection with repeated [Formula: see text]-fold cross-validation selects the best subset of features from eight predictors: highest price, lowest price, closing price, volume, change, price change ratio, and amplitude. These features are then used as input in a bidirectional gated recurrent unit with an attention mechanism (BiGRU-AM) model to forecast daily opening prices of ten stock indices. The proposed model exhibits superior forecasting performance across ten stock indices when compared to twelve benchmarks, evaluated using root mean squared error (RMSE), mean absolute error (MAE), and the coefficient of determination, [Formula: see text]. The improved prediction accuracy enables financial professionals to make more reliable investment decisions, reducing risks and increasing profits.

CEG-0598, a novel dual inhibitor of EGFR and C5aR demonstrates in vitro anticancer and antimetastatic activity in prostate cancer cells

BackgroundThe EGFR is abundantly expressed in prostate cancer (PC). The anaphylatoxin C5a induces leukocyte migration via the C5a receptor (C5aR) by releasing matrix metalloproteinases (MMP) to favor metastasis in the tumor microenvironment. This work aims to selectively inhibit the EGFR and C5aR in PC cells to abort cell growth/ proliferation and metastasis.MethodsFor lead identification, high-throughput virtual screening (HTVS) of the ChemBridge library was followed by protein–ligand interaction profilers, GROMACS, and GMX-MMPBSA techniques. LNCaP and PC3 cells were used to validate in vitro efficacy.ResultsHTVS identified CEG-0598 with favorable binding affinities of − 10.2 kcal/mol and − 13.5 kcal/mol towards EGFR and C5aR respectively. Molecular dynamic simulations demonstrated stable binding interactions for CEG-0598 with Root Mean Square Deviation values around 0.06 nm. The ΔG binding calculation was − 50.29, and − 51.64 for EGFR and C5aR respectively. ADME supported favorable small molecule characteristics and selective inhibition profiles. Kinome-wide off-target virtual screening predicted EGFR to have above-average docking scores. CEG-0598 inhibited EGFR and C5aR activities with IC50 values of 145.8 nM and 55.51 nM respectively. The compound effectively controlled the proliferation of LNCaP and PC3cells with GI50 values of 156.1 nM, and 112.2 nM respectively. CEG-0598 prompted dose-responsive apoptosis in the PC cells and decreased the tarns endothelial migration of both PC cells. Treatment with CEG-0598 reduced the C5a-induced MMP activity in the LNCaP and PC3cells.ConclusionCEG-0598 is a selective EGFR/C5a dual inhibitor that downregulates MMP activity to control proliferation, migration and induce apoptosis, in PC cells warranting further preclinical developments.

Assessing and validating machine learning-enhanced imputation of admission American Spinal Injury Association Impairment Scale grades for spinal cord injury.

The American Spinal Injury Association Impairment Scale (AIS) assigned at patient admission is an important predictor of outcomes following spinal cord injury (SCI). However, nearly 80% of records in the Spinal Cord Injury Model Systems (SCIMS) database-a multicenter prospective database of patients with SCI-lack admission AIS grades. Accurate imputation of this missing data could enable more robust analyses and insights into SCI recovery. This study aims to develop and validate methods for imputing missing admission AIS data in the SCIMS database. The study included 16,062 patients with SCI from the publicly available SCIMS database (1988-2020). Five machine learning algorithms-random forest (RF), linear discriminant analysis, K-nearest neighbors, naive Bayes, and support vector machine-were compared using performance metrics (accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and multiclass area under the receiver operating characteristic curve) using five-fold cross-validation on a training subset of 6054 patients with complete AIS admission grades. The model with the highest performance was trained on all 16,062 patients. The imputed AIS grades were validated by predicting discharge functional independence measure (FIM) scores (range 13-91) with simple and multiple linear regression models on a 1:1 propensity score-matched cohort (n = 5828). Model performance was compared using differences in root mean square error (∆RMSE) with bootstrapped 95% confidence intervals (CIs). The full cohort contained a representative distribution of AIS grades (45% grade A, 13% grade B, 18% grade C, and 24% grade D), and the propensity score-matched cohort characteristics were well balanced. The RF algorithm demonstrated the highest validation accuracy (81.7%). Predictive models showed no significant differences between models using true versus imputed AIS grades, with 95% CIs for ∆RMSE of -0.60 to 0.47 for simple regression and -0.63 to 0.46 for multiple regression models. The coefficients of AIS grades also did not significantly differ between models with true versus imputed values. A data-driven approach to imputation resulted in a robust method for imputing admission AIS grades that demonstrated clinical validity in the SCIMS database. This approach extends the utility of this longitudinal database and may provide a framework for other SCI databases.

Scaling of pressure fluctuation in turbulent internal flows

Previous studies on the scaling of pressure fluctuations in wall-bounded turbulent flows have typically employed the same frameworks as those used for mean flow, with inner scaling based on frictional velocity and viscous length scales, and outer scaling relying on boundary layer thickness or displacement thickness. These traditional scales primarily reflect the characteristics of the mean streamwise velocity profile and momentum balance. In this work, we propose novel scaling frameworks for pressure fluctuations in turbulent channel and pipe flows, derived from the Poisson equation for pressure fluctuations. Applying the scaling patch approach, we analyse the rapid and slow terms in the Poisson equation, and introduce new scaling for pressure fluctuation variance in both the inner and outer regions. These new scales are designed to better capture the influence of Reynolds stresses by incorporating their peak values. Additionally, we establish a strong correlation between the root mean square (r.m.s.) of pressure fluctuations and the Reynolds shear stress, resulting in an empirical equation that accurately predicts their ratio. This equation provides a practical method for estimating the r.m.s. of pressure fluctuations in the flow, which remains challenging to measure in experimental investigations.

Advanced Time Series Forecasting for CO₂ Emissions: Insights for Sustainable Climate Policies

To address the global issue of climate change and create focused mitigation plans, accurate CO2 emissions forecasting is essential. Using CO2 emissions data from 1990 to 2023, this study assesses the predicting performance of five sophisticated models: Random Forest (RF), XGBoost, Support Vector Regression (SVR), Long Short-Term Memory networks (LSTM), and ARIMA To give a thorough evaluation of the models’ performance, measures including Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE) are used. To guarantee dependable model implementation, preprocessing procedures are carried out, such as feature engineering and stationarity tests. Machine learning models outperform ARIMA in identifying complex patterns and long-term associations, but ARIMA does better with data that exhibits strong linear trends. These results provide important information about how well the model fits various forecasting scenarios, which helps develop data-driven carbon reduction programs. Predictive modeling should be incorporated into sustainable climate policy to encourage the adoption of low-carbon technologies and proactive decision-making. Achieving long-term environmental sustainability requires strengthening carbon trading systems, encouraging clean energy investments, and enacting stronger emission laws. In line with international climate goals, suggestions for lowering CO2 emissions include switching to renewable energy, increasing energy efficiency, and putting afforestation initiatives into action.

Lactation Curve Modelling and Genetic Parameters Estimation in Murciano-Granadina Goats.

The present study aims to determine the best non-linear model for describing lactation curves and estimating genetic parameters for the lactation curve traits in the Murciano-Granadina goats in Iran. We compared five mathematical models including the Cappio-Borlino (CB), Cobby and Le Du (CD), Narushin-Takma (NT), Wilmink (WL), and Wood (WD) to characterise the lactation curve in the first and second lactations of Murciano-Granadina does. The dataset consisted of 36,958 and 23,319 milk yield test-day records from 4964 first-parity and 3335 s-parity Murciano-Granadina does, respectively. These records were collected from 2017 to 2024 in a private dairy farm, located in Ghale-Ganj city, Kerman province, southern area of Iran. In both lactation periods, the WD model showed the lowest values for root mean squares of prediction error (RMSE) and Akaike's information criterion (AIC), as well as the highest adjusted coefficient of determination ( ) among the evaluated models. Additionally, positive autocorrelations were observed among the residuals for all the models considered, with the lowest positive autocorrelation obtained under the WD model. Therefore, WD was identified as the best model to characterise the lactation curve of the Murciano-Granadina does in the first and second lactation periods. Consequently, we computed the individual lactation curve traits for does in the ith parity (where i = 1 for the first parity and i = 2 for the second parity), including peak time (PTi), peak milk yield (PYi), and lactation persistency (LPi), using the parameters derived from the WD model. A multivariate animal model utilising a Bayesian approach was employed to estimate the genetic parameters of the lactation curve traits. The posterior means for heritability estimates were 0.07, 0.13, 0.05, 0.05, 0.11, and 0.08 for PT1, PY1, LP1, PT2, PY2, and LP2, respectively. In the first parity, genetic correlations among the lactation curve traits were positive estimates of 0.28, 0.96, and 0.25 for PT1-PY1, PT1-LP1, and PY1-LP1, respectively. In the second parity, the corresponding genetic correlation estimates were 0.88, 0.89, and 0.59 for PT2-PY2, PT2-LP2 and PY2-LP2, respectively. It can be concluded that the low heritability estimates for the investigated lactation curve traits suggest these traits are mainly affected by non-additive genetic and environmental effects. Consequently, direct genetic selection may not effectively modify the shape of the Murciano-Granadina lactation curve. The positive genetic correlation estimates among the traits examined within each parity, as well as among the same traits across the parities, suggest that selecting one trait will also enhance the other traits.

GNSS Precipitable Water Vapor Prediction for Hong Kong Based on ICEEMDAN-SE-LSTM-ARIMA Hybrid Model

Accurate prediction of Global Navigation Satellite System-derived precipitable water vapor (GNSS-PWV), which is a crucial indicator for climate change monitoring, holds significant scientific value for climate disaster prevention and mitigation. In the study of GNSS-PWV prediction, the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) algorithm within a decomposition–integration framework effectively addresses the non-stationarity and complexity of PWV sequences, enhancing prediction accuracy. However, residual noise and pseudo-modes from decomposition can distort signals, reducing the predictor system’s reliability. Additionally, independent modeling of all decomposed components decreases computational efficiency. To address these challenges, this paper proposes a hybrid model combining the improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN), autoregressive integrated moving average (ARIMA), and long short-term memory (LSTM) networks. Enhanced by local mean optimization and adaptive noise regulation, the ICEEMDAN algorithm effectively suppresses pseudo-modes and minimizes residual noise, enabling its decomposed intrinsic mode functions (IMFs) to more accurately capture the multi-scale features of GNSS-PWV. Sample entropy (SE) is used to quantify the complexity of IMFs, and components with similar entropy values are reconstructed into the following three sub-sequences: high-frequency, low-frequency, and trend. This process significantly reduces modeling complexity and improves computational efficiency. We propose different modeling strategies tailored to the dynamics of various subsequences. For the nonlinear and non-stationary high-frequency components, the LSTM network is used to effectively capture their complex patterns. The LSTM’s gating mechanism and memory cell design proficiently address the long-term dependency issue. For the stationary and weakly nonlinear low-frequency and trend components, linear patterns are extracted using ARIMA. Differencing eliminates trends and moving average operations capture random fluctuations, effectively addressing periodicity and trends in the time series. Finally, the prediction results of the three components are linearly combined to obtain the final prediction value. To validate the model performance, experiments were conducted using measured GNSS-PWV data from several stations in Hong Kong. The results demonstrate that the proposed model reduces the root mean square error by 56.81%, 37.91%, and 13.58% at the 1 h scale compared to the LSTM, EMD-LSTM, and ICEEMDAN-SE-LSTM benchmark models, respectively. Furthermore, it exhibits strong robustness in cross-month forecasts (accounting for seasonal influences) and multi-step predictions over the 1–6 h period. By improving the accuracy and efficiency of PWV predictions, this model provides reliable technical support for the real-time monitoring and early warning of extreme weather events in Hong Kong while offering a universal methodological reference for multi-scale modeling of geophysical parameters.

High-precision large-aperture single-frame interferometric surface profile measurement method based on deep learning

Abstract Large-aperture optical components are of paramount importance in domains such as integrated circuit, photolithography, aerospace and inertial confinement fusion. While, the measurement of their surface profiles predominantly relies on the phase-shifting approach, which involves collecting multiple interferograms and imposes stringent demands on environmental stability. Such issues significantly hinder its capability of achieving real-time and dynamic high-precision measurement. In this paper, we demonstrate a high-precision large-aperture single-frame interferometric surface profile measurement (LA-SFISPM) method based on deep learning. The interferogram is matched to the phase by training the data that measured by the small-aperture. And the consistency of the surface features of the small and large aperture is enhanced through contrast learning and the feature distribution alignment. Hence, high-precision phase reconstruction of large-aperture optical components is achieved without the phase shifter. Experimental results show that for the tested mirror with Φ=820 mm, the surface profile obtained from LA-SFISPM is subtracted point-by-point from the ground truth which resulting in a maximum single-point error is 4.56 nm. Meanwhile, the peak-to-valley (PV) value is 0.0758λ, and the simple repeatability of root mean square (SR-RMS) value is 0.00025λ which are in good agreement with the measured results obtained by ZYGO. Especially, a significant reduction in measurement time (reduced by a factor of 48 times) is realized in comparison with the traditional phase-shifting method. Our demonstrated method provides an efficient, rapid and accurate way to obtain the surface profile of optical components with different diameters without phase-shifting approach which is highly desired in large-aperture interferometric measurement systems.

How to Better Use Canopy Height in Soybean Biomass Estimation

Soybean, a globally important food and oil crop, requires accurate estimation of above-ground biomass (AGB) to optimize management and prevent yield loss. Despite the availability of various remote sensing methods, systematic research on effectively integrating canopy height (CH) and spectral information for improved AGB estimation remains insufficient. This study addresses this gap using drone data. Three CH utilization approaches were tested: (1) simple combination of CH and spectral vegetation indices (VIs), (2) fusion of CH and VI, and (3) integration of CH, VI, and growing-degree days (GDDs). The results indicate that adding CH always enhances AGB estimation which is based only on VIs, with the fusion approach outperforming simple combination. Incorporating GDD further improved AGB estimation for highly accurate CH data, with the best model achieving a root mean square error (RMSE) of 87.52 ± 5.88 g/m2 and a mean relative error (MRE) of 28.59 ± 1.99%. However, for the multispectral data with low CH accuracy, the VIs + GDD fusion (RMSE = 92.94 ± 6.84 g/m2, MRE = 30.08 ± 2.29%) surpassed CH + VIs + GDD (RMSE = 97.99 ± 6.71 g/m2, MRE = 31.41 ± 2.56%). The findings highlight the role of CH accuracy in AGB estimation and validate the value of growth-stage information in robust modeling. Future research should prioritize the refining of CH prediction and the optimization of composite variable construction to promote the application of this approach in agricultural monitoring.

Localization algorithm for 3D four-wheel alignment based on monocular vision

Abstract A novel localization algorithm for a four-wheel alignment is present. As our algorithm is based on monocular vision, the problem that a traditional 3D four-wheel alignment needs to unify the global coordinate system and easy to cause error accumulation is not existing. In accordance with the principle of monocular reconstruction in computer vision, the representation of equivalent reference plane for each wheel in its corresponding camera coordinate system is calculated out. Based on these new definitions, related angle parameters of the vehicle wheel are finally determined. Simulations and analysis show the feasibility and robustness of our algorithm described in this paper. Experiment results show that the proposed algorithm can obtain precise results and the root mean square error is about 0.005 degrees. Compared with the traditional localization algorithm of a 3D four-wheel alignment, the measurement precision of our proposed algorithm is similar, but does not need global coordinate system calibration, so the operation is simple and efficient. Furthermore, as the wheel measurement is relatively independent, it is not easy to cause error accumulation.

Triplanar Point Cloud Reconstruction of Head Skin Surface from Computed Tomography Images in Markerless Image-Guided Surgery

Accurate preoperative image processing in markerless image-guided surgeries is an important task. However, preoperative planning highly depends on the quality of medical imaging data. In this study, a novel algorithm for outer skin layer extraction from head computed tomography (CT) scans is presented and evaluated. Axial, sagittal, and coronal slices are processed separately to generate spatial data. Each slice is binarized using manually defined Hounsfield unit (HU) range thresholding to create binary images from which valid contours are extracted. The individual points of each contour are then projected into three-dimensional (3D) space using slice spacing and origin information, resulting in uniplanar point clouds. These point clouds are then fused through geometric addition into a single enriched triplanar point cloud. A two-step downsampling process is applied, first at the uniplanar level and then after merging, using a voxel size of 1 mm. Across two independent datasets with a total of 83 individuals, the merged cloud approach yielded an average of 11.61% more unique points compared to the axial cloud. The validity of the triplanar point cloud reconstruction was confirmed by a root mean square (RMS) registration error of 0.848 ± 0.035 mm relative to the ground truth models. These results establish the proposed algorithm as robust and accurate across different CT scanners and acquisition parameters, supporting its potential integration into patient registration for markerless image-guided surgeries.

Coordinated Control of Autonomous Electric Vehicles With Lateral and Longitudinal Control Using a Hybrid Approach

ABSTRACTThe rise of Autonomous Electric Vehicles (AEVs) has presented formidable challenges in the automotive sector, demanding advanced sensor technology, intricate control systems, and sophisticated decision‐making algorithms. Due to the inherently nonlinear dynamics and uncertainties associated with these vehicles, conventional control methods fall short of providing robust solutions. This study proposes a hybrid approach for coordinated longitudinal and lateral control in autonomous driving scenarios. Addressing lateral and longitudinal control, the research integrates road geometry and lateral dynamics considerations. Utilizing a Proportional Integral Derivative (PID) controller with Fire Hawk Optimizer (FHO) algorithm. This study optimizes controller gains for Nonlinear longitudinal dynamics, ensuring reliable speed tracking. Additionally, a Linear Parameter Varied‐Models Predictive Controller (LPV‐MPC) addresses the challenges related to time‐varying longitudinal speeds and distance impact on vehicle lateral stability. Implementation in the matrix laboratory demonstrates the approach's superiority in terms of speed, precision, stability, trajectory tracking, and achieving a minimal lateral error of 0.0526 and mean error, mean absolute error and root mean squared error of 0.193, 0.087 and 0.108 respectively.

Short-term wind power prediction based on adaptive Stacking integrated learning model

Abstract Aiming at the problem that when a single combined model uses multiple features for short-term wind power prediction with a small number of training samples, it is difficult to capture the characteristics of time series data, resulting in a large generalization error of the prediction model. This paper proposes a short-term wind power prediction method based on an adaptive Stacking integrated learning model. This ensemble prediction method selects four basic models with different working principles, namely Random Forest (RF), Transformer, Bidirectional Long Short-Term Memory Network (BiLSTM), and Gated Temporal Convolutional Network (GTCN). Since the meta-learner of Stacking ensemble learning cannot fuse each basic model differentially, this limits the advantage of the fusion of model prediction results to a certain extent, leading to the accumulation of errors. In this paper, by introducing an adaptive dynamic attention mechanism, weights are assigned to the preliminary prediction results of each basic model to form weighted input features. Finally, the weighted input features of each basic model are sent to the meta-learner for ensemble training, and the prediction results are mapped and fused to obtain the final wind power prediction result. According to the actual power generation data of a certain offshore wind farm in Fujian Province, taking the data in December as an example, the Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and correlation coefficient ( ) of the proposed model are 97.13, 65.77, and 0.91 respectively. The comparison results with multiple models show that the proposed method has higher prediction accuracy.&amp;#xD;