Search Results (40)

Introduction: Recurrence after pelvic exenteration remains a significant concern in patients with locally advanced rectal cancer (LARC). Therefore, there is a need for improved non-invasive predictive tools to aid in patient selection. Radiomics, which extracts quantitative imaging features, may help identify patients at greater risk of recurrence. This study aimed to develop and validate a radiomics-based nomogram using pre-treatment MRI to predict postoperative recurrence risk in LARC. Methods: The largest multicenter retrospective radiomics analysis of 191 patients with pathologically confirmed LARC treated at fourteen centres (2016–2018) was performed. All patients received neoadjuvant chemoradiotherapy followed by curative-intent exenterative surgery. Manual tumour segmentation was performed on pre-treatment T2-weighted MRI. Feature selection employed LASSO regression with 5-fold cross-validation across 1000 bootstrap samples. The most frequently selected features were used to construct a logistic regression model via stepwise backward selection. Model performance was assessed using ROC analysis, calibration plots, decision curve analysis, and internal validation with 1000 bootstraps. A nomogram was generated to enable individualized recurrence risk estimation. Results: Postoperative recurrence occurred in 51% (n = 98) of cases. Five radiomic features reflecting tumour heterogeneity, morphology, and texture were included in the final model. In multivariable analysis, all selected features were significantly associated with recurrence, with odds ratios ranging from 0.63 to 1.64. The model achieved an optimism-adjusted AUC of 0.70, indicating fair discrimination. Calibration plots showed good agreement between predicted and observed recurrence probabilities. Decision curve analysis confirmed clinical utility across relevant thresholds. A clinically interpretable nomogram was developed based on the final model. Conclusions: A radiomics-based model using preoperative MRI can predict recurrence in LARC. The derived nomogram provides a practical tool for preoperative risk assessment. Prospective validation is necessary. Full article

Coastal wetlands are critical ecosystems that require effective monitoring to support conservation and restoration efforts. This study evaluates the use of small unmanned aerial systems (sUAS) and multispectral imagery to estimate aboveground biomass (AGB) in tidal marshes, comparing models calibrated with destructive versus non-destructive in situ sampling methods. Imagery was collected over South Carolina’s North Inlet-Winyah Bay National Estuarine Research Reserve, and vegetation indices (VIs) were derived from sUAS imagery to model biomass. Stepwise linear regression was used to develop and validate models based on both sampling approaches. Destructive sampling models, particularly those using the Normalized Difference Vegetation Index (NDVI) and Difference Vegetation Index (DVI), achieved the lowest root mean square error (RMSE) values (as low as 70.91 g/m²), indicating higher predictive accuracy. Non-destructive models, while less accurate (minimum RMSE of 214.86 g/m²), demonstrated higher R² values (0.44 and 0.61), suggesting the potential for broader application with further refinement. These findings highlight the trade-offs between ecological impact and model performance, and support the viability of non-destructive methods for biomass estimation in sensitive wetland environments. Future work should explore machine learning approaches and improved temporal alignment of data collection to enhance model robustness. Full article

This study presents a novel approach combining near-infrared (NIR) spectroscopy with multivariate calibration to develop simplified yet robust regression models for evaluating the quality of various edible oils. Using a reduced number of NIR wavelengths selected via the stepwise decorrelation method (SELECT) and ordinary least squares (OLS) regression, the models quantify pigments (carotenoids and chlorophyll), antioxidant activity, and key sensory attributes (rancid, fruity green, fruity ripe, bitter, and pungent) in nine extra virgin olive oil (EVOO) varieties. The dataset also includes low-quality olive oils (e.g., refined and pomace oils, supplemented or not with hydroxytyrosol) and sunflower oils, both before and after deep-frying. SELECT improves model performance by identifying key wavelengths—up to 30 out of 700—and achieves high correlation coefficients (R = 0.86–0.96) with low standard errors. The number of latent variables ranges from 26 to 30, demonstrating adaptability to different oil properties. The best models yield low leave-one-out (LOO) prediction errors, confirming their accuracy (e.g., 1.36 mg/kg for carotenoids and 0.88 for rancidity). These results demonstrate that SELECT–OLS regression combined with NIR spectroscopy provides a fast, cost-effective, and reliable method for assessing oil quality under diverse processing conditions, including deep-frying, making it highly suitable for quality control in the edible oils industry. Full article

Near-infrared (NIR) spectroscopy, combined with multivariate calibration techniques such as stepwise decorrelation of variables (SELECT) and ordinary least squares (OLS) regression, was used to develop robust, reduced-spectrum regression models for quantifying key phenolic compound markers in various olive oils. These oils included nine extra virgin olive oil (EVOO) varieties, refined olive oil (ROO) blended with virgin olive oil (VOO) or EVOO, and pomace olive oil, both with and without hydroxytyrosol (HTyr) supplementation. Olive oils were analyzed before and after deep frying. The results show that HTyr ranged from 7.28 mg/kg in Manzanilla (lowest) to 21.43 mg/kg in Royuela (highest). Tyrosol (Tyr) varied from 5.87 mg/kg in Royuela (lowest) to 14.86 mg/kg in Hojiblanca (highest). Similar trends were observed in all phenolic fractions across olive oil cultivars before and after deep-frying. HTyr supplementation significantly increased both HTyr and Tyr levels in non-fried and fried supplemented oils, with HTyr rising from single digits in some controls (around 0 mg/kg) to over 300 mg/kg in most of the supplemented samples. SELECT efficiently reduced redundancy by selecting the most vital wavelengths and thus significantly improved the regression models for key phenolic compounds, including HTyr, Tyr, caffeic acid, decarboxymethyl ligstroside aglycone in dialdehyde form (oleocanthal), decarboxymethyl oleuropein aglycone in dialdehyde form (oleacein), homovanillic acid, pinoresinol, oleuropein aglycone in oxidized aldehyde and hydroxylic form (OAOAH), ligstroside aglycone in oxidized aldehyde and hydroxylic form (LAOAH), and total phenolic content (TPC), achieving correlation coefficients (R) of 0.91–0.98. The SELECT-OLS method generated highly predictive models with minimal complexity, using at most 30 wavelengths out of 700. The number of decorrelated predictors varied, at 12, 14, 15, 30, 30, 21, 30, 30, 30, and 18 for HTyr, Tyr, caffeic acid, oleocanthal, oleacein, homovanillic acid, pinoresinol, OAOAH, LAOAH, and TPC, respectively, demonstrating the adaptability of the SELECT-OLS approach to different spectral patterns. These reliable calibration models enabled online and routine quantification of phenolic compounds in EVOO, VOO, ROO, including both non-fried and fried as well as supplemented and non-supplemented samples. They performed well across eight deep-frying conditions (3–6 h at 170–210 °C). Implementing an NIR instrument with optimized variable selection would simplify spectral analysis and reduce costs. The developed models all demonstrated strong predictive performance, with low leave-one-out mean prediction errors (LOOMPEs) with values of 15.69, 8.47, 3.64, 9.18, 16.71, 3.26, 8.57, 13.56, 56.36, and 82.38 mg/kg for HTyr, Tyr, caffeic acid, oleocanthal, oleacein, homovanillic acid, pinoresinol, OAOAH, LAOAH, and TPC, respectively. These results confirm that NIR spectroscopy combined with SELECT-OLS is a feasible, rapid, non-destructive, and eco-friendly tool for the reliable evaluation and quantification of phenolic content in edible oils. Full article

Given the highly variable distribution of seasonal snowpacks in complex mountainous environments, the accurate snow modeling of basin-wide snow water equivalent (SWE) requires a spatially distributed approach at a sufficiently fine grid resolution (<500 m) to account for the important processes in the seasonal evolution of a snowpack (e.g., wind redistribution of snow to resolve patchy snow cover in an alpine zone). However, even well-validated snow evolution models, such as SnowModel, are prone to errors when key model inputs, such as the precipitation and wind speed and direction, are inaccurate or only available at coarse spatial resolutions. Incorporating fine-spatial-resolution remotely sensed snow-covered area (SCA) information into spatially distributed snow modeling has the potential to refine and improve fine-resolution snow water equivalent (SWE) estimates. This study developed 30 m resolution SnowModel simulations across the Big Thompson River, Fraser River, Three Lakes, and Willow Creek Basins, a total area of 4212 km² in Colorado, for the water years 2000–2023, and evaluated the incorporation of a Moderate Resolution Imaging Spectroradiometer (MODIS) and Landsat SCA datasets into the model’s development and calibration. The SnowModel was calibrated spatially to the Landsat mean annual snow persistence (SP) and temporally to the MODIS mean basin SCA using a multi-objective calibration procedure executed using Latin hypercube sampling and a stepwise calibration process. The Landsat mean annual SP was also used to further optimize the SnowModel simulations through the development of a spatially variable precipitation correction field. The evaluations of the SnowModel simulations using the Airborne Snow Observatories’ (ASO’s) light detection and ranging (lidar)-derived SWE estimates show that the versions of the SnowModel calibrated to the remotely sensed SCA had an improved performance (mean error ranging from −28 mm to −6 mm) compared with the baseline simulations (mean error ranging from 69 mm to 86 mm), and comparable spatial patterns to those of the ASO, especially at the highest elevations. Furthermore, this study’s results highlight how a regularly updated 30 m resolution SCA could be used to further improve the calibrated SnowModel simulations to near real time (latency of 5 days or less). Full article

Accurately determining the hydraulic properties of soilless growing media is essential for optimizing water management in container-based horticulture and agriculture. The very rapid estimation of hydraulic properties using a Mini Disk Infiltrometer has great potential for practical use compared to the very time-consuming standard methods. The objectives of this study were (1) to calibrate simulated cumulative stepwise infiltration under different suctions with the measured data from Mini Disk Infiltrometer, (2) to evaluate the efficiency of the Hydrus-2D inverse model to predict water dynamics through substrates, (3) to compare the substrate hydraulic parameters obtained through the numerical inversion model to those obtained via laboratory methods, and (4) to provide recommendations on how to effectively use the MDI-based method for practical applications. This study employs numerical inversion of Mini Disk Infiltrometer (MDI) data to estimate the hydraulic parameters of three different growing media, namely white peat, thermally treated wood fibre (WF4), and Seedling substrate. Infiltration experiments were conducted under suction-controlled conditions using varying initial moisture contents, followed by numerical simulations using the Hydrus-2D model and the Van Genuchten equation to describe the hydraulic parameters. The results demonstrated strong agreement between observed and simulated infiltration data, particularly under moistened conditions, with high R² > 0.9 values indicating the model’s effectiveness. However, discrepancies were observed for substrates in their initial dry state, suggesting limitations in capturing early-stage infiltration dynamics. The findings highlighted the potential of numerical inversion methods for estimating substrate hydraulic properties but also revealed the need for methodological refinements. Modifying the Van Genuchten model or exploring alternative approaches such as the Brooks and Corey model may enhance accuracy. Extending the suction range of measurement techniques is also recommended to improve parameter estimation. This study provides important evidence that the inverse method based on MDI is an effective tool for rapidly determining the hydraulic functions of substrates, which are important in promoting sustainable horticultural practices. Future research should focus on refining parameter estimation methods and addressing model limitations to enhance the reliability of hydraulic property assessments in soilless growing media. Full article

Objective: Accurate risk stratification at an early stage may reduce the incidence of infection and improve the survival rate of recipients by adopting targeted interventions. This study aimed to develop a nomogram to predict the risk of multidrug-resistant organism (MDRO) infections in liver transplant (LT) recipients. Methods: We retrospectively collected clinical data from 301 LT recipients and randomly divided them into a training set (210 cases) and validation set (91 cases) using a 7:3 split ratio. Factors related to the risk of MDRO infection after LT were determined using univariate and multivariate bidirectional stepwise logistic regression. The model’s predictive performance and discrimination ability were evaluated using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA). Results: 56 (18.60%) patients developed a MDRO infection, including 37 (17.62%) in the training cohort and 19 (20.88%) in the validation cohort. Ultimately, five factors related to MDRO infection after LT surgery were established: ascites (OR = 3.48, 95% CI [1.33–9.14], p = 0.011), total bilirubin (OR = 1.01, 95% CI [1.01–1.01], p < 0.001), albumin (OR = 0.85, 95% CI [0.75–0.96], p = 0.010), history of preoperative ICU stay (OR = 1.09, 95% CI [1.01–1.17], p = 0.009), and length of ICU stay (OR = 3.70, 95% CI [1.39–9.84], p = 0.019). The model demonstrated strong discrimination, and the area under the curve (AUC), sensitivity, and specificity of the training set were 0.88 (95% CI [0.81–0.94]), 0.82 (95% CI [0.76–0.87]), and 0.86 (95% CI [0.75–0.98]), respectively, while for the validation set, they were 0.77 (95% CI [0.65–0.90]), 0.76 (95% CI [0.67–0.86]), and 0.68 (95% CI [0.48–0.89]). The mean absolute error (MAE) in the validation cohort was 0.029, indicating a high accuracy. DCA showed a clinical benefit within a threshold probability range of 0.1 to 0.7. Conclusions: This study developed a clinically accessible nomogram to predict the risk of MDRO infection in LT recipients, enabling early risk stratification and the real-time assessment of infection risk based on the length of postoperative ICU stay. The model incorporates five easily obtainable clinical parameters (ascites, total bilirubin, albumin, preoperative ICU stay history, and length of ICU stay) and demonstrates strong predictive performance, facilitating the early identification of high-risk patients. Future research should focus on refining the model by incorporating additional clinical factors (e.g., immunosuppressive therapy adherence) and validating its generalizability in multicenter, large-sample cohorts to enhance its clinical utility. Full article

We describe a modification of a previously described measurement–analysis protocol to determine the intrinsic properties of triaxial accelerometers by using a measurement protocol based on angular stepwise rotation in the Earth’s gravitational field. This study was conducted with MEMS triaxial accelerometers that were co-integrated in four consumer-grade wireless microsensors. The measurements were carried out on low-cost rotation tables in different laboratories in different countries to simulate the reproducibility environment encountered in inter-comparisons of calibration capabilities. We used a previously described calibration–uncertainty metric to independently characterize the overall uncertainty of the calibration and analysis process. The intrinsic property analysis suggested, and the uncertainty metric confirmed, an unacceptably large error in one combination of microsystem and low-cost rotation table. A simple modification of the analysis protocol provided a substantial improvement in the reproducibility of the protocol with all combinations of microsystem and rotation table. Later, measurements with a high-performance triaxial accelerometer using a significantly more expensive rotation table carried out at one location further validated the usefulness of this modification. The results reported here also demonstrate the existence of unidentified defects in one microsystem and one low-cost rotation table that interact with each other in ways not currently understood to produce anomalously large errors with the old protocol but not with the new protocol. Full article

In this work, kinetic models are assessed to describe bacterial cellulose (BC) production, substrate consumption, and biomass growth by K. xylinus in a batch-stirred tank bioreactor, under 700 rpm and 500 rpm agitation rates. The kinetic models commonly used for Acetobacter or Gluconacetobacter were fitted to published data and compared using the Akaike Information Criterion (AIC). A stepwise fitting procedure was proposed for model selection to reduce computation effort, including a first calibration in which only the biomass and substrate were simulated, a selection of the three most effective models in terms of AIC, and a calibration of the three selected models with the simulation of biomass, substrate, and product. Also, an uncoupled product equation involving a modified Monod substrate function is proposed for a 500 rpm agitation rate, leading to an improved prediction of BC productivity. The M2c and M1c models were the most efficient for biomass growth and substrate consumption for the combined AIC, under 700 rpm and 500 rpm agitation rates, respectively. The average coefficients of determination for biomass, substrate, and product predictions were 0.981, 0.994, and 0.946 for the 700 rpm agitation rate, and 0.984, 0.991, and 0.847 for the 500 rpm agitation rate. It is shown that the prediction of BC productivity is improved through the proposed substrate function, whereas the computation effort is reduced through the proposed model fitting procedure. Full article

Modern hydrological modeling frequently incorporates global remote sensing or reanalysis products for multivariate calibration. Although these datasets significantly contribute to model accuracy, the inherent uncertainties in the datasets and multivariate calibration present challenges in the modeling process. To address this issue, this study introduces an adaptive, process-wise fitting framework for the iterative multivariate calibration of hydrological models using global remote sensing and reanalysis products. A distinctive feature is the “kinship” concept, which defines the relationship between model parameters and hydrological processes, highlighting their impacts and connectivity within a directed graph. The framework subsequently develops an enhanced particle swarm optimization (PSO) algorithm for stepwise calibration of hydrological processes. This algorithm introduces a learning rate that reflects the parameter’s kinship to the calibrated hydrological process, facilitating efficient exploration in search of suitable parameter values. This approach maximizes the performance of the calibrated process while ensuring a balance with other processes. To ease the impact of inherent uncertainties in the datasets, the Extended Triple Collocation (ETC) method, operating independently of ground truth data, is integrated into the framework to assess the simulation of the calibrated process using remote sensing products with inherent data uncertainty. This proposed approach was implemented with the SWAT model in both arid and humid basins. Five calibration schemes were designed and evaluated through a comprehensive comparison of their performance in three repeated experiments. The results highlight that this approach not only improved the accuracy of ET simulation across sub-basins but also enhanced the precision of streamflow at gauge stations, concurrently reducing parameter uncertainty. This approach significantly advances our understanding of hydrological processes, demonstrating the potential for both theoretical and practical applications in hydrology. Full article

Accumulating evidence has underscored the prognostic value of tumor-infiltrating immune cells in the tumor microenvironment of colon cancer (CC). In this retrospective study, based on publicly available transcriptome profiles and clinical data from the Gene Expression Omnibus and The Cancer Genome Atlas databases, we derived and verified an activated dendritic cell (aDC)-related gene signature (aDCRS) for predicting the survival outcomes and chemotherapy and immunotherapy response of CC patients. We quantified the infiltration abundance of 22 immune cell subtypes via the “CIBERSORT” R script. Univariate Cox proportional hazards (PHs) regression was used to identify aDC as the most robust protective cell type for CC prognosis. After selecting differentially expressed genes (DEGs) significantly correlated with aDC infiltration, we performed univariate Cox-PH regression, LASSO regression, and stepwise multivariate Cox-PH regression successively to screen out prognosis-related genes from selected DEGs for constructing the aDCRS. Receiver operating characteristic (ROC) curves and Kaplan–Meier (KM) analysis were employed to assess the discriminatory ability and risk-stratification capacity. The “oncoPredict” package, Cancer Treatment Response gene signature DataBase, and Tumor Immune Dysfunction and Exclusion algorithm were utilized to estimate the practicability of the aDCRS in predicting response to chemotherapy and immune checkpoint blockade. Gene set enrichment analysis and single-cell RNA sequencing analysis were also implemented. Furthermore, an aDCRS-based nomogram was constructed and validated via ROC curves, calibration plots and decision curve analysis. In conclusion, aDCRS and an aDCRS-based nomogram will facilitate precise prognosis prediction and individualized therapeutic interventions, thus improving the survival outcomes of CC patients in the future. Full article

Robot measurement systems with a binocular planar structured light camera (3D camera) installed on a robot end-effector are often used to measure workpieces’ shapes and positions. However, the measurement accuracy is jointly influenced by the robot kinematics, camera-to-robot installation, and 3D camera measurement errors. Incomplete calibration of these errors can result in inaccurate measurements. This paper proposes a joint calibration method considering these three error types to achieve overall calibration. In this method, error models of the robot kinematics and camera-to-robot installation are formulated using Lie algebra. Then, a pillow error model is proposed for the 3D camera based on its error distribution and measurement principle. These error models are combined to construct a joint model based on homogeneous transformation. Finally, the calibration problem is transformed into a stepwise optimization problem that minimizes the sum of the relative position error between the calibrator and robot, and analytical solutions for the calibration parameters are derived. Simulation and experiment results demonstrate that the joint calibration method effectively improves the measurement accuracy, reducing the mean positioning error from over 2.5228 mm to 0.2629 mm and the mean distance error from over 0.1488 mm to 0.1232 mm. Full article

Alzheimer’s disease is an incurable disorder that accounts for up to 70% of all dementia cases. While the prevalence of Alzheimer’s disease and other types of dementia has increased by more than 160% in the last 30 years, the rates of undetected cases remain critically high. The present work aims to address the underdetection of Alzheimer’s disease by proposing four logistic regression models that can be used as a foundation for community-based screening tools that do not require the participation of medical professionals. Our models make use of individual clock drawing errors as well as complementary patient data that is highly available and easily collectible. All models were controlled for age, education, and gender. The discriminative ability of the models was evaluated by area under the receiver operating characteristic curve (AUC), the Hosmer-Lemeshow test, and calibration plots were used to assess calibration. Finally, decision curve analysis was used to quantify clinical utility. We found that among 10 possible CDT errors, only 3 were informative for the detection of Alzheimer’s disease. Our base regression model, containing only control variables and clock drawing errors, produced an AUC of 0.825. The other three models were built as extensions of the base model with the step-wise addition of three groups of complementary data, namely cognitive features (semantic fluency score), genetic predisposition (family history of dementia), and cardio-vascular features (BMI, blood pressure). The addition of verbal fluency scores significantly improved the AUC compared to the base model (0.91 AUC). However, further additions did not make a notable difference in discriminatory power. All models showed good calibration. In terms of clinical utility, the derived models scored similarly and greatly outperformed the base model. Our results suggest that the combination of clock symmetry and clock time errors plus verbal fluency scores may be a suitable candidate for developing accessible screening tools for Alzheimer’s disease. However, future work should validate our findings in larger and more diverse datasets. Full article

There is increasing interest in connected and automated vehicles (CAVs), since their implementation will transform the nature of transportation and promote social and economic change. Transition toward cooperative driving still requires the understanding of some key questions to assess the performances of CAVs and human-driven vehicles on roundabouts and to properly balance road safety and traffic efficiency requirements. In this view, this paper proposes a simulation-based methodological framework aiming to assess the presence of increasing proportions of CAVs on roundabouts operating at a high-capacity utilization level. A roundabout was identified in Palermo City, Italy, and built in Aimsun (version 20) to describe the stepwise methodology. The CAV-based curves of capacity by entry mechanism were developed and then used as target capacities. To calibrate the model parameters, the capacity curves were compared with the capacity data simulated by Aimsun. The impact on the safety and performance efficiency of a lane dedicated to CAVs was also examined using surrogate measures of safety. The paper ends with highlighting a general improvement with CAVs on roundabouts, and with providing some insights to assess the advantages of the automated and connected driving technologies in transitioning to smarter mobility. Full article

The aboveground dry biomass (AGDB) of winter wheat can reflect the growth and development of winter wheat. The rapid monitoring of AGDB by using hyperspectral technology is of great significance for obtaining the growth and development status of winter wheat in real time and promoting yield increase. This study analyzed the changes of AGDB based on a winter wheat irrigation experiment. At the same time, the AGDB and canopy hyperspectral reflectance of winter wheat were obtained. The effect of spectral preprocessing algorithms such as reciprocal logarithm (Lg), multiple scattering correction (MSC), standardized normal variate (SNV), first derivative (FD), and second derivative (SD); sample division methods such as the concentration gradient method (CG), the Kennard–Stone method (KS), and the sample subset partition based on the joint X–Y distances method (SPXY); sample division ratios such as 1:1 (Ratio1), 3:2 (Ratio2), 2:1 (Ratio3), 5:2 (Ratio4), and 3:1 (Ratio5); dimension reduction algorithms such as uninformative variable elimination (UVE); and modeling algorithms such as partial least-squares regression (PLSR), stepwise multiple linear regression (SMLR), artificial neural network (ANN), and support vector machine (SVM) on the hyperspectral monitoring model of winter wheat AGDB was studied. The results showed that irrigation can improve the AGDB and canopy spectral reflectance of winter wheat. The spectral preprocessing algorithm can change the original spectral curve and improve the correlation between the original spectrum and the AGDB of winter wheat and screen out the bands of 1400 nm, 1479 nm, 1083 nm, 741 nm, 797 nm, and 486 nm, which have a high correlation with AGDB. The calibration sets and validation sets divided by different sample division methods and sample division ratios have different data-distribution characteristics. The UVE method can obviously eliminate some bands in the full-spectrum band. SVM is the best modeling algorithm. According to the universality of data, the better sample division method, sample division ratio, and modeling algorithm are SPXY, Ratio4, and SVM, respectively. Combined with the original spectrum and by using UVE to screen bands, a model with stable performance and high accuracy can be obtained. According to the particularity of data, the best model in this study is FD-CG-Ratio4-Full-SVM, for which the R²_c, RMSE_c, R²_v, RMSE_v, and RPD are 0.9487, 0.1663 kg·m⁻², 0.7335, 0.3600 kg·m⁻², and 1.9226, respectively, which can realize hyperspectral monitoring of winter wheat AGDB. This study can provide a reference for the rational irrigation of winter wheat in the field and provide a theoretical basis for monitoring the AGDB of winter wheat by using hyperspectral remote sensing technology. Full article