Cross-validated Root Mean Square Research Articles

Exploring the Potential of Machine Learning Algorithms to Improve Diffusion Nuclear Magnetic Resonance Imaging Models Analysis

Purpose: This paper explores different machine learning (ML) algorithms for analyzing diffusion nuclear magnetic resonance imaging (dMRI) models when analytical fitting shows restrictions. It reviews various ML techniques for dMRI analysis and evaluates their performance on different b-values range datasets, comparing them with analytical methods. Materials and Methods: After standard fitting for reference, four sets of diffusion-weighted nuclear magnetic resonance images were used to train/test various ML algorithms for prediction of diffusion coefficient (D), pseudo-diffusion coefficient (D*), perfusion fraction (f), and kurtosis (K). ML classification algorithms, including extra-tree classifier (ETC), logistic regression, C-support vector, extra-gradient boost, and multilayer perceptron (MLP), were used to determine the existence of diffusion parameters (D, D*, f, and K) within single voxels. Regression algorithms, including linear regression, polynomial regression, ridge, lasso, random forest (RF), elastic-net, and support-vector machines, were used to estimate the value of the diffusion parameters. Performance was evaluated using accuracy (ACC), area under the curve (AUC) tests, and cross-validation root mean square error (RMSECV). Computational timing was also assessed. Results: ETC and MLP were the best classifiers, with 94.1% and 91.7%, respectively, for the ACC test and 98.7% and 96.3% for the AUC test. For parameter estimation, RF algorithm yielded the most accurate results The RMSECV percentages were: 8.39% for D, 3.57% for D*, 4.52% for f, and 3.53% for K. After the training phase, the ML methods demonstrated a substantial decrease in computational time, being approximately 232 times faster than the conventional methods. Conclusions: The findings suggest that ML algorithms can enhance the efficiency of dMRI model analysis and offer new perspectives on the microstructural and functional organization of biological tissues. This paper also discusses the limitations and future directions of ML-based dMRI analysis.

Field Testing of Gamma-Spectroscopy Method for Soil Water Content Estimation in an Agricultural Field.

Gamma-ray spectroscopy (GRS) enables continuous estimation of soil water content (SWC) at the subfield scale with a noninvasive sensor. Hydrological applications, including hyper-resolution land surface models and precision agricultural decision making, could benefit greatly from such SWC information, but a gap exists between established theory and accurate estimation of SWC from GRS in the field. In response, we conducted a robust three-year field validation study at a well-instrumented agricultural site in Nebraska, United States. The study involved 27 gravimetric water content sampling campaigns in maize and soybean and 40K specific activity (Bq kg-1) measurements from a stationary GRS sensor. Our analysis showed that the current method for biomass water content correction is appropriate for our maize and soybean field but that the ratio of soil mass attenuation to water mass attenuation used in the theoretical equation must be adjusted to satisfactorily describe the field data. We propose a calibration equation with two free parameters: the theoretical 40K intensity in dry soil and a, which creates an "effective" mass attenuation ratio. Based on statistical analyses of our data set, we recommend calibrating the GRS sensor for SWC estimation using 10 profiles within the footprint and 5 calibration sampling campaigns to achieve a cross-validation root mean square error below 0.035 g g-1.

Controlling capillary flow by using laser-fabricated acrylic substrates with rectangular grooves for the improvement of repeatability of laser induced breakdown spectroscopy liquid analysis

To enhance the consistency of Laser-Induced Breakdown Spectroscopy (LIBS) in detecting dried liquid droplets, a rectangular groove substrate prepared by laser etching of a polymethyl methacrylate (PMMA) acrylic board is proposed in this work to suppress the Coffee Ring Effect (CRE). Located at the corner of the substrate, the right-angled heat source structure induces a significant inward thermal capillary flow within the groove, thus inhibiting solute capillary flow. At the later stages of heating, the edge deposition that occurs during evaporation is suppressed by the sawtooth-like microscopic grooves, along with the capillary effects in associated with the liquid. The rationale of CRE suppression by the rectangular groove substrate is theoredically analyzed based on the coupled thermal-solute capillary flow theory. With LIBS used to scan the samples within rectangular and circular groove substrates, it is revealed that the relative standard deviation (RSD) of spectral intensities for Cd I 259.28 nm and Cu I 324.27 nm in the rectangular groove substrate with a depth of 0.1 mm (RSD = 4.2%, RSD = 2.1%) is less significant than in the circular groove substrate (RSD = 41.2%, RSD = 39.3%). On average, the RSD is 7.6% and 7.3%, respectively, which indicates the effect of the rectangular groove substrate on CRE suppression. On this basis, calibration curves are drawn for Cd and Cu to indicate the good fitting results for the rectangular groove substrate, with the determination coefficient (R2) reaching 0.997 and 0.996 for Cd and Cu, respectively. The limit of detection (LOD) is 0.48 μg/mL and 0.35 μg/mL, respectively. The cross-validation root mean square error (RMSECV) is 2.02 μg/mL and 1.36 μg/mL, respectively. These results demonstrate that the rectangular groove substrate is effective in suppressing the CRE, and in enhancing the sensitivity of detection and the accuracy of heavy metal prediction, which provides guidance for platform selection in the pre-processing stage of LIBS technology.

A portable dry film FTIR instrument for industrial food and bioprocess applications.

The main objective of this study was to design, build, and test a compact, multi-well, portable dry film FTIR system for industrial food and bioprocess applications. The system features dry film sampling on a circular rotating disc comprising 31 wells, a design that was chosen to simplify potential automation and robotic sample handling at a later stage. Calibration models for average molecular weight (AMW, 200 samples) and collagen content (68 samples) were developed from the measurements of industrially produced protein hydrolysate samples in a controlled laboratory environment. Similarly, calibration models for the prediction of lactate content in samples from cultivation media (59 samples) were also developed. The portable dry film FTIR system showed reliable model characteristics which were benchmarked with a benchtop FTIR system. Subsequently, the portable dry film FTIR system was deployed in a bioprocessing plant, and protein hydrolysate samples were measured at-line in an industrial environment. This industrial testing involved building a calibration model for predicting AMW using 60 protein hydrolysate samples measured at-line using the portable dry film FTIR system and subsequent model validation using a test set of 26 samples. The industrial calibration in terms of coefficient of determination (R2 = 0.94), root mean square of cross-validation (RMSECV = 194 g mol-1), and root mean square of prediction (RMSEP = 162 g mol-1) demonstrated low prediction errors as compared to benchtop FTIR measurements, with no statistical difference between the calibration models of the two FTIR systems. This is to the authors' knowledge the first study for developing and employing a portable dry film FTIR system in the enzymatic protein hydrolysis industry for successful at-line measurements of protein hydrolysate samples. The study therefore suggests that the portable dry film FTIR instrument has huge potential for in/at-line applications in the food and bioprocessing industries.

Calibration for iron redox state and oxygen fugacity in silicate glasses from x-ray absorption spectroscopy

Prior work to quantify glass Fe-redox ratios in situ using x-ray absorption spectroscopy (XAS) has relied heavily on the pre-edge region. The pre-edge-based methods do not take advantage of other information in the spectra above the pre-edge, up to and past the main edge. This study updates the previous calibration (Dyar et al., 2016a) with additional experimental glasses spanning a wide range of geologically relevant compositions (SiO2 weight % = 38.5–78.0), equilibrated over a range of oxygen fugacity (fO2) values from reducing (IW-2) to oxidizing (air), for which Fe XAS analysis has been collected for the whole spectral region (pre-edge, main edge, EXAFS). The resulting dataset was used to evaluate multiple multivariate statistical models for their ability to accurately predict log fO2 or %Fe3+. Accuracies are estimated using cross-validated root mean square error, for which purpose unseen test data are held out from the regressions. Datasets covering small compositional ranges were tested and compared with datasets covering broad compositional ranges. These ‘stratified’ sets improved predictive accuracy in some cases; however, the extra complications associated with data stratification did not justify the small improvement in accuracy. Overall, a simple partial least squares (PLS) model based on all data collected over the entire XAS region was found to result in a universal calibration for determination of log fO2 and %Fe3+ in glasses.

A new prediction model for total body water estimation by BIA in children with normal and excessive weight

Various methods, including bioelectrical impedance analysis (BIA), are used for total body water (TBW) estimation. The objective of our study by BIA was to develop a new predication model based on corrected TBW for normal adult BMI, a concept similar to the standardization of glomerular filtration rate by relating it to the average adult body surface area. We measured TBW by BIA in 335 children 3-21 years old with normal or excessive body weight. Based on our data, we derived a new prediction model for TBW (L) for females {[(72.784+0.4093×weight)∗Corrected TBW]/100} and males {[(57.944+0.6551×weight)∗Corrected TBW]/100}. For validation, we compared our prediction model with three other models on TBW by BIA and dilution methods. Our model's error size to predict TBW showed lower cross-validated root mean square error (CV-RMSE) as compared to three other models versus our dataset by BIA and two other datasets by dilution methods. Our model also showed a smaller error (2.059) in CV-RMSE as compared to other models by dilution methods (2.126, 2.873, and 4.384) for normal and excessive weight combined. This implies that our model is more robust when excessive weight individuals are included in the data.. Our prediction model for TBW estimation by BIA performs better as compared to some other models based on BIA and dilution method datasets. Furthermore, our prediction model is the only one that is devised to be applicable to children and young adults with both normal as well as excessive weight.

Estimating Regional PM2.5 Concentrations in China Using a Global-Local Regression Model Considering Global Spatial Autocorrelation and Local Spatial Heterogeneity

Linear regression models are commonly used for estimating ground PM2.5 concentrations, but the global spatial autocorrelation and local spatial heterogeneity of PM2.5 distribution are either ignored or only partially considered in commonly used models for estimating PM2.5 concentrations. Therefore, taking both global spatial autocorrelation and local spatial heterogeneity into consideration, a global-local regression (GLR) model is proposed for estimating ground PM2.5 concentrations in the Yangtze River Delta (YRD) and in the Beijing, Tianjin, Hebei (BTH) regions of China based on the aerosol optical depth data, meteorological data, remote sensing data, and pollution source data. Considering the global spatial autocorrelation, the GLR model extracts global factors by the eigenvector spatial filtering (ESF) method, and combines the fraction of them that passes further filtering with the geographically weighted regression (GWR) method to address the local spatial heterogeneity. Comprehensive results show that the GLR model outperforms the ordinary GWR and ESF models, and the GLR model has the best performance at the monthly, seasonal, and annual levels. The average adjusted R2 of the monthly GLR model in the YRD region (the BTH region) is 0.620 (0.853), which is 8.0% and 7.4% (6.8% and 7.0%) higher than that of the monthly ESF and GWR models, respectively. The average cross-validation root mean square error of the monthly GLR model is 7.024 μg/m3 in the YRD region, and 9.499 μg/m3 in the BTH region, which is lower than that of the ESF and GWR models. The GLR model can effectively address the spatial autocorrelation and spatial heterogeneity, and overcome the shortcoming of the ordinary GWR model that overfocuses on local features and the disadvantage of the poor local performance of the ordinary ESF model. Overall, the GLR model with good spatial and temporal applicability is a promising method for estimating PM2.5 concentrations.

Can machine learning models predict maternal and newborn healthcare providers’ perception of safety during the COVID-19 pandemic? A cross-sectional study of a global online survey

BackgroundMaternal and newborn healthcare providers are essential professional groups vulnerable to physical and psychological risks associated with the COVID-19 pandemic. This study uses machine learning algorithms to create a predictive tool for maternal and newborn healthcare providers’ perception of being safe in the workplace globally during the pandemic.MethodsWe used data collected between 24 March and 5 July 2020 through a global online survey of maternal and newborn healthcare providers. The questionnaire was available in 12 languages. To predict healthcare providers’ perception of safety in the workplace, we used features collected in the questionnaire, in addition to publicly available national economic and COVID-19-related factors. We built, trained and tested five machine learning models: Support Vector Machine (SVM), Random Forest (RF), XGBoost, CatBoost and Artificial Neural Network (ANN) for classification and regression. We extracted from RF models the relative contribution of features in output prediction.ResultsModels included data from 941 maternal and newborn healthcare providers from 89 countries. ML models performed well in classification and regression tasks, whereby RF had 82% cross-validated accuracy for classification, and CatBoost with 0.46 cross-validated root mean square error for regression. In both classification and regression, the most important features contributing to output prediction were classified as three themes: (1) information accessibility, clarity and quality; (2) availability of support and means of protection; and (3) COVID-19 epidemiology.ConclusionThis study identified salient features contributing to maternal and newborn healthcare providers perception of safety in the workplace. The developed tool can be used by health systems globally to allow real-time learning from data collected during a health system shock. By responding in real-time to the needs of healthcare providers, health systems could prevent potential negative consequences on the quality of care offered to women and newborns.

Dielectric behavior of soil as a function of frequency, temperature, moisture content and soil texture: a deep neural networks based regression model

Dielectric behavior of soil has utmost applications in microwave remote sensing and soil treatment. In the present study, the soil's dielectric properties (Ɛ' and Ɛ") were measured using the vector network analyzer and an open-ended coaxial probe (85070E, Agilent Technologies) in the region of 0.2 to 14 GHz. The observed results showed that Ɛ' and Ɛ" strongly depend on frequency, texture, moisture content and temperature. A deep neural network (DNN) based multivariable regression model has been developed to model their behavior, using experimentally observed data to learn its parameters automatically. It shows a five-fold cross-validation root mean square errors (RMSE) of 0.0258 and 0.0336, and R2-scores of 1.0000 and 0.9998, between actual recorded and predicted values of Ɛ' and Ɛ", respectively. The results of the proposed DNN-based model have been compared with the response surface method (RSM) based model; among these, the DNN-based model shows significantly better results. Further, the DNN-based estimates of Ɛ' and Ɛ" for loam texture at a moisture content of 18% (i.e. in between observed experiments of 15% and 20%) are made and plotted with actual observed values at 15% and 20% to verify the predictive ability of the proposed DNN-based model. It shows an acceptable estimate of dielectric properties and the effectiveness of the fast and innovative DNN-based approach for predicting soil's dielectric properties depending upon multiple factors.

MEASUREMENT, PREDICTION AND VALIDATION OF HUMAN GAIT TORQUE FOR LOWER LIMB ASSISTIVE DEVICES

The lower limb rehabilitation process requires external control inputs based on joint kinematics and kinetics to its actuator for generating the motion of a robotic assistive device. This paper describes the procedures for measuring, predicting and validating the joint kinematics and kinetics of human gait for the lower limb exoskeleton. The high-speed multi-camera-based video system associated with passive optical markers and a force plate sensor is used to measure the gait data. The lower limb joint torques have also been predicted using three different machine learning regression models, viz. Decision Tree (DT), Gaussian Process Regression (GPR) and Support Vector Machine (SVM). The analysis is further validated with the geometry-based analytical method. Three lower limb joint torques have been predicted by utilizing gait period, joint angle, joint velocity and joint acceleration as the predictors. The lowest average 10-fold cross-validation Root Mean Square Error (RMSE) using the DT model between measured and predicted torques have been recorded for the ankle, knee and hip joint as 1.9521, 1.5527 and 1.5684[Formula: see text]Nm, respectively. This study enhances the field of medical rehabilitation by providing the required knowledge of human gait kinematics/kinetics and predicting the joint torque profile without any force sensor.

Accuracy and self-validation of automated bone age determination

The BoneXpert method for automated determination of bone age from hand X-rays was introduced in 2009 and is currently running in over 200 hospitals. The aim of this work is to present version 3 of the method and validate its accuracy and self-validation mechanism that automatically rejects an image if it is at risk of being analysed incorrectly. The training set included 14,036 images from the 2017 Radiological Society of North America (RSNA) Bone Age Challenge, 1642 images of normal Dutch and Californian children, and 8250 images from Tübingen from patients with Short Stature, Congenital Adrenal Hyperplasia and Precocious Puberty. The study resulted in a cross-validated root mean square (RMS) error in the Tübingen images of 0.62 y, compared to 0.72 y in the previous version. The RMS error on the RSNA test set of 200 images was 0.45 y relative to the average of six manual ratings. The self-validation mechanism rejected 0.4% of the RSNA images. 121 outliers among the self-validated images of the Tübingen study were rerated, resulting in 6 cases where BoneXpert deviated more than 1.5 years from the average of the three re-ratings, compared to 72 such cases for the original manual ratings. The accuracy of BoneXpert is clearly better than the accuracy of a single manual rating. The self-validation mechanism rejected very few images, typically with abnormal anatomy, and among the accepted images, there were 12 times fewer severe bone age errors than in manual ratings, suggesting that BoneXpert could be safer than manual rating.

Seasonal climate signals preserved in biochemical varves: insights from novel high-resolution sediment scanning techniques

Abstract. Varved lake sediments are exceptional archives of paleoclimatic information due to their precise chronological control and annual resolution. However, quantitative paleoclimate reconstructions based on the biogeochemical composition of biochemical varves are extremely rare, mainly because the climate–proxy relationships are complex and obtaining biogeochemical proxy data at very high (annual) resolution is difficult. Recent developments in high-resolution hyperspectral imaging (HSI) of sedimentary pigment biomarkers combined with micro X-ray fluorescence (µXRF) elemental mapping make it possible to measure the structure and composition of varves at unprecedented resolution. This provides opportunities to explore seasonal climate signals preserved in biochemical varves and, thus, assess the potential for annual-resolution climate reconstruction from biochemical varves. Here, we present a geochemical dataset including HSI-inferred sedimentary pigments and µXRF-inferred elements at very high spatial resolution (60 µm, i.e. > 100 data points per varve year) in varved sediments of Lake Żabińskie, Poland, over the period 1966–2019 CE. We compare these data with local meteorological observations to explore and quantify how changing seasonal meteorological conditions influenced sediment composition and varve formation processes. Based on the dissimilarity of within-varve multivariate geochemical time series, we classified varves into four types. Multivariate analysis of variance shows that these four varve types were formed in years with significantly different seasonal meteorological conditions. Generalized additive models (GAMs) were used to infer seasonal climate conditions based on sedimentary variables. Spring and summer (MAMJJA) temperatures were predicted using Ti and total C (Radj2=0.55; cross-validated root mean square error (CV-RMSE) = 0.7 ∘C, 14.4 %). Windy days from March to December (mean daily wind speed > 7 m s−1) were predicted using mass accumulation rate (MAR) and Si (Radj2=0.48; CV-RMSE = 19.0 %). This study demonstrates that high-resolution scanning techniques are promising tools to improve our understanding of varve formation processes and climate–proxy relationships in biochemical varves. This knowledge is the basis for quantitative high-resolution paleoclimate reconstructions, and here we provide examples of calibration and validation of annual-resolution seasonal weather inference from varve biogeochemical data.

Using least angular regression to model the antibacterial potential of metronidazole complexes

Imidazole has anti-inflammatory, antituberculotic, antimicrobial, antimycotic, antiviral, and antitumor properties in the human body, to name a few. Metronidazole [1-(2-Hydroxyethyl)-2-methyl-5-nitroimidazole] is a widely used antiprotozoan and antibacterial medication. Using fourier transform infrared spectroscopy, the current study models the antibacterial activity of already synthesised Metronidazole (MTZ) complexes (MTZ-Benz, MTZ-Benz-Cu, MTZ-Benz-Cu-Cl_2CHCOOH, MTZ, MTZ-Cu, MTZ-Cu-Cl_2CHCOOH, MTZ-Benz-Ag, MTZ-Benz-Ag-Cl_2CHCOOH, MTZ-Ag and MTZ-Ag-Cl_2CHCOOH) against E. coli, B. bronceptica, S. epidermidis, B. pumilus and S. aureus. To characterise the Metronidazole complexes for antibacterial activity against 05 microbes, the least angular regression and least absolute shrinkage selection operators were used. Asymmetric Least Squares was used to correct the spectrum baseline. Least angular regression outperforms cross-validated root mean square error in the fitted models. Using Least angular regression, influential wavelengths that explain the variation in antibacterial activity of Metronidazole complexes were identified and mapped against functional groups.

Spaceborne height models reveal above ground biomass changes in tropical landscapes

The area-wide estimation of aboveground biomass (AGB) and its changes as a proxy for the sequestration and emission of carbon are currently associated with high uncertainties. Here we combined interferometric synthetic aperture radar (InSAR) height models derived from TanDEM-X with repeated ground-based inventories from the years 2012 and 2019 to estimate InSAR height and AGB changes in a structurally diverse and dynamic landscape in Sumatra, Indonesia. The results suggested that the InSAR height models were highly accurate and the relationship between InSAR height and AGB change resulted in a coefficient of determination R2 of 0.65 and a cross-validated root mean square error (RMSE) of 2.38 Mg ha−1 year−1, equivalent to 13.32% of the actual AGB difference range. The estimated AGB changes with TanDEM-X were further related to the initial canopy height and fire activities in the study area. Initial canopy heights and the occurrences of fires had a significant effect on the AGB change. In general, low canopy heights tend to be associated with increasing AGB over time, whereas high canopy heights tend to be associated with stable or decreasing AGB. As expected, fires had a negative impact on the AGB changes being more pronounced in forest areas compared to oil palm concessions. The results of this study are relevant for the utilization of spaceborne InSAR height models and its potential to estimate canopy height and AGB change on large spatial scales. It was demonstrated that these changes can be related to their sources and ecosystem processes. This AGB change estimation technique can be used to model the impacts of fires on AGB change and carbon emissions, which are important for sustainable forest management.

Research on the measurement method of printing ink content based on spectrum

In the printing quality inspection, measuring the primary color ink content of printed products is an important link. In order to obtain accurate measurement results in the ink content measurement system, this study preprocesses the interference information such as stray light and noise in the collected ink content spectrum information of cyan, magenta, and yellow samples firstly. Considering the limited denoising ability of common filtering methods in the frequency domain, a composite filtering method combining median filtering and wavelet transform is introduced to preprocess the collected spectral signal. The system not only suppressed the random interference, but restored original signal well. Then, considering that there are many redundant data in full-wavelength spectral reflectance data, the Successive Projections Algorithm (SPA) and Competitive Adaptive Reweighted Sampling (CARS) algorithm were used to extract characteristic wavelengths from the preprocessed information. After extracting the characteristic wavelengths, the primary color ink content prediction models were built based on the Partial Least Squares Regression (PLSR) and Support Vector Regression (SVR) algorithms. The test results show that SVR models based on SPA and CRAS feature extraction algorithms have better prediction effect than PLSR models. Among them, CARS-SVR algorithm models have the best prediction effect, and the modeling wavelengths of three primary color samples decreases by 82.75%, 69.25% and 90.50%, compared with the whole wavelengths. The Cross-Validation Root Mean Square Error (RMSECV) and coefficient of determination (R2) of CARS-SVR content prediction models are 0.0256, 0.0182, 0.0249 and 0.9931, 0.9966, 0.9936, respectively. The system showed better measurement accuracy.

Rapid determination of benzalkonium chloride in aqueous samples by FTIR spectroscopy in tandem with chemometrics

In this research, a powerful regression model coupled with FTIR spectroscopy (Mid, 800–1700 cm−1) has been proposed as an efficient method for precise determination of Benzalkonium chloride (BAK) in aqueous samples. For this purpose, both partial least squares regression (PLS-R) and support vector regression (SVR) as methods of multivariate calibration were used for evaluation, and their results were compared. Accordingly, root mean square error of prediction, and leave-one-out cross-validation root mean square error, and correlation coefficients between the calculated (Rcal2) and the predicted (Rpred2) values were used. In comparison to PLS (Rpred2 = 0.975; RMSEP = 0.321), SVR had a higherRpred2 (0.991) and a lower value of root mean square error of prediction (RMSEP = 0.218). The lower detection limit was 0.00068% w/w for PLS and 0.0011% w/w for SVR model in a concentration range from 0.013 to 1% w/w. Hence, FTIR spectroscopy combined with SVR can be considered an efficient approach for real-time determination of BAK in aqueous samples.

Use of Multi-Temporal LiDAR to Quantify Fertilization Effects on Stand Volume and Biomass in Late-Rotation Coastal Douglas-Fir Forests

Forest fertilization is common in coastal British Columbia as a means to increase wood production and potentially enhance carbon sequestration. Generally, the effects of fertilization are determined by measuring sample plots pre- and post-treatment, resulting in fertilization effects being determined for a limited portion of the treatment area. Applications of remote sensing-based enhanced forest inventories have allowed for estimations to expand to the wider forested area. However, these applications have not focused on monitoring the effects of silvicultural treatments. The objective of this research was to examine if a multi-temporal application of the LiDAR area-based method can be used to detect the fertilization effects on volume, biomass, and height in a second-growth Douglas-fir (Pseudotsuga menziesii) stand. The study area on Vancouver Island was fertilized in January 2007, and sample plots were established in 2011. LiDAR acquisitions were made in 2004, prior to fertilization, and in 2008, 2011, and 2016, covering both treated and untreated areas. A total of 29 paired LiDAR blocks, comprised of four 20 m resolution raster cells, were selected on either side of the fertilization boundary for analysis of the effects across several different stand types differing in the percentage of Douglas-fir, site index, and age. Random forest (RF) plot-level models were developed to estimate total stem volume and total stem biomass for each year of LiDAR acquisition using an area-based approach. Plot level results showed an increase in stem volume by 13% fertilized over control from 2005 to 2011, which was similar to a 14% increase in above-ground carbon stocks estimated using a tree-ring stand reconstruction approach. Plot-level RF models showed R2 values of 0.86 (volume) and 0.92 (biomass) with relative cross-validated root mean square errors of 12.5% (volume) and 11.9% (biomass). For both the sample plots and LiDAR blocks, statistical results indicated no significant differences in volume or biomass between treatments. However, significant differences in height increments were detected between treatments in LiDAR blocks. The results from this research highlight the promising potential for the use of enhanced forest inventory methods to rapidly expand the assessment of treatment effects beyond sample plots to the stand, block, or landscape level.

Histamine Control in Raw and Processed Tuna: A Rapid Tool Based on NIR Spectroscopy.

The present study was designed to investigate whether near infrared (NIR) spectroscopy with minimal sample processing could be a suitable technique to rapidly measure histamine levels in raw and processed tuna fish. Calibration models based on orthogonal partial least square regression (OPLSR) were built to predict histamine in the range 10–1000 mg kg−1 using the 1000–2500 nm NIR spectra of artificially-contaminated fish. The two models were then validated using a new set of naturally contaminated samples in which histamine content was determined by conventional high-performance liquid chromatography (HPLC) analysis. As for calibration results, coefficient of determination (r2) > 0.98, root mean square of estimation (RMSEE) ≤ 5 mg kg−1 and root mean square of cross-validation (RMSECV) ≤ 6 mg kg−1 were achieved. Both models were optimal also in the validation stage, showing r2 values > 0.97, root mean square errors of prediction (RMSEP) ≤ 10 mg kg−1 and relative range error (RER) ≥ 25, with better results showed by the model for processed fish. The promising results achieved suggest NIR spectroscopy as an implemental analytical solution in fish industries and markets to effectively determine histamine amounts.

Estimation of leaf water content from hyperspectral data of different plant species by using three new spectral absorption indices.

The leaf equivalent water thickness (EWT, g cm−2) and fuel moisture content (FMC, %) are key variables in ecological and environmental monitoring. Although a variety of hyperspectral vegetation indices have been developed to estimate the leaf EWT and FMC, most of these indices are defined considered two or three specific bands for a specific plant species, which limits their applicability. In this study, we proposed three new spectral absorption indices (SAI970, SAI1200, and SAI1660) for various plant types by considering the symmetry of the spectral absorption at 970 nm, 1200 nm and 1660 nm and spectral heterogeneity of different leaves. The indices were calculated considering the absorption peak and shoulder bands of each leaf instead of the same specific bands for all leaves. A pooled dataset of three tree species (camphor (VX), capricorn (VJ), and red-leaf plum (VL)) was used to test the performance of the SAIs in terms of the leaf EWT and FMC estimation. The results indicated that, first, SAI1200 was more suitable for estimating the EWT than FMC, whereas SAI970 and SAI1660 were more suitable for estimating the FMC. Second, SAI1200 achieved the most accurate estimation of the EWT with a cross-validation coefficient of determination (Rcv2) of 0.845 and relative cross-validation root mean square error (rRMSEcv) of 8.90%. Third, SAI1660 outperformed the other indices in estimating the FMC at the leaf level, with an Rcv2 of 0.637 and rRMSEcv of 8.56%. Fourth, SAI970 achieved a moderate accuracy in estimating the EWT (Rcv2 of 0.25 and rRMSEcv of 19.68%) and FMC (Rcv2 of 0.275 and rRMSEcv of 12.10%) at the leaf level. These results can enrich the application of the SAIs and demonstrate the potential of using SAI1200 to determine the leaf EWT and SAI1660 to obtain the leaf FMC among various plant types.

Mining data from milk mid-infrared spectroscopy and animal characteristics to improve the prediction of dairy cow's liveweight using feature selection algorithms based on partial least squares and Elastic Net regressions

Body weight (BW) of dairy cows is relevant for breeding programs and farm management to assess the maintenance requirements, reproduction performance, or health status of cow. Currently, it is still difficult to follow BW changes of individual cows routinely in large herds. Combined with animal characteristics, milk mid-infrared (MIR) spectrum was proposed as an additional source of information to predict BW under the framework of dairy herd improvement (DHI) programs. However, the presence of less informative variables in the prediction equation could impact negatively its robustness. This research aims to improve the robustness of BW regression models by applying a feature selection before modeling. A total of 5,920 BW records composed of animal characteristics and milk MIR spectrum were collected from Holstein cows. Three feature selection algorithms were applied to select the most informative variables: partial least squares regression (PLS) combined with sum of ranking difference (PLS-SRD), PLS combined with uninformative variables elimination (PLS-UVE), and the output of Elastic Net regression (EN). Four herd independent validation sets and the corresponding remained calibration datasets having on average 163 and 1,708 records, respectively, were used to develop models using PLS or EN approaches. Ten-fold cross-validation was conducted to parametrize each model. Parity, days in milk (DIM), milk yield (MY), and two MIR spectral points were selected as relevant variables to predict BW. PLS (root mean square error of validation, RMSEp = 60 kg) and EN (RMSEp = 60 kg) regressions employing these 5 predictors were more robust than the models developed without MIR or using MIR without feature selection. The EN models had a cross-validation root mean square error of around 53 kg. The 2 MIR points explained up to 4.20% variation in predicting BW. The RMSE of validation sets using another brand of spectrometer were around 64 kg. This study confirms the possibility to predict an indicator of BW from animal characteristics and MIR variables. The variable selection procedures improved the model's robustness and transferability. The accuracy of BW prediction seems to be sufficient to provide useful information for breeding program and farm management decisions under a DHI framework.