Search NASA⌕ Search

SEARCH · Search NASA

Results for “Linear Regression”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 415 records · Page 23

Radiation Exposure in Biliary Procedures Performed to Manage Anastomotic Strictures in Pediatric Liver Transplant Recipients: Comparison Between Radiation Exposure Levels Using an Image Intensifier and a Flat-Panel Detector-Based System

Purpose: The aim of this study was to estimate radiation exposure in pediatric liver transplants recipients who underwent biliary interventional procedures and to compare radiation exposure levels between biliary interventional procedures performed using an image intensifier-based angiographic system (IIDS) and a flat panel detector-based interventional system (FPDS). Materials and Methods: We enrolled 34 consecutive pediatric liver transplant recipients with biliary strictures between January 2008 and March 2013 with a total of 170 image-guided procedures. The dose-area product (DAP) and fluoroscopy time was recorded for each procedure. The mean age was 61 months (range 4-192), and mean weight was 17 kg (range 4-41). The procedures were classified into three categories: percutaneous transhepatic cholangiography and biliary catheter placement (n = 40); cholangiography and balloon dilatation (n = 55); and cholangiography and biliary catheter change or removal (n = 75). Ninety-two procedures were performed using an IIDS. Seventy-eight procedures performed after July 2010 were performed using an FPDS. The difference in DAP between the two angiographic systems was compared using Wilcoxon rank-sum test and a multiple linear regression model. Results: Mean DAP in the three categories was significantly greater in the group of procedures performed using the IIDS compared with those performed using the FPDS. Statistical analysis showed a p value = 0.001 for the PTBD group, p = 0.0002 for the cholangiogram and balloon dilatation group, and p = 0.00001 for the group with cholangiogram and biliary catheter change or removal. Conclusion: In our selected cohort of patients, the use of an FPDS decreases radiation exposure.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Cone-Beam Computed Tomography (CBCT) Versus CT in Lung Ablation Procedure: Which is Faster?

AimTo compare cone-beam CT (CBCT) versus computed tomography (CT) guidance in terms of time needed to target and place the radiofrequency ablation (RFA) electrode on lung tumours.Materials and MethodsPatients at our institution who received CBCT- or CT-guided RFA for primary or metastatic lung tumours were retrospectively included. Time required to target and place the RFA electrode within the lesion was registered and compared across the two groups. Lesions were stratified into three groups according to their size (<10, 10–20, >20 mm). Occurrences of electrode repositioning, repositioning time, RFA complications, and local recurrence after RFA were also reported.ResultsForty tumours (22 under CT, 18 under CBCT guidance) were treated in 27 patients (19 male, 8 female, median age 67.25 ± 9.13 years). Thirty RFA sessions (16 under CBCT and 14 under CT guidance) were performed. Multivariable linear regression analysis showed that CBCT was faster than CT to target and place the electrode within the tumour independently from its size (β = −9.45, t = −3.09, p = 0.004). Electrode repositioning was required in 10/22 (45.4 %) tumours under CT guidance and 5/18 (27.8 %) tumours under CBCT guidance. Pneumothoraces occurred in 6/14 (42.8 %) sessions under CT guidance and in 6/16 (37.5 %) sessions under CBCT guidance. Two recurrences were noted for tumours receiving CBCT-guided RFA (2/17, 11.7 %) and three after CT-guided RFA (3/19, 15.8 %).ConclusionCBCT with live 3D needle guidance is a useful technique for percutaneous lung ablation. Despite lesion size, CBCT allows faster lung RFA than CT.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Will X-ray Safety Glasses Become Mandatory for Radiological Vascular Interventions?

PurposeThe annual permissible radiation ocular lens dose has been reduced to 20 millisieverts (mSv) in the current European directive 2013/59/Euratom. The aim of this study was to evaluate the personal radiation dose for vascular interventions with special focus on ocular lens dose.Materials and MethodsFrom May 2016 to October 2016, the personal radiation doses of two interventionists and four technicians were prospectively recorded during 206 vascular interventions. The position of personnel, intervention type and fluoroscopy time were recorded. Parameters evaluated were total body dose measured by film dosimeter, hand dose measured by ring thermoluminescent dosimeter (TLD) and ocular lens dose measured by TLD placed in front of the safety glasses. Linear regression analysis was used to estimate the dose at 2 and 5 years.ResultsThe ocular lens dose, hand and total body dose of the two interventionists were 11/5, 56/47 and 0.6 mSv each, respectively. The estimated 5-year ocular dose was 113.08 mSv (95% CI 38.2–187.97)/40.95 (95% CI 16.9–64.7). Similarly, hand dose was 608.4 mSv (95% CI 442.78–774.38)/514.47 (95% CI 329.83–699.10) and body dose 6.07 mSv (95% CI 4.70–8.22)/5.12 (95% CI 3.65–6.59), respectively. Amongst four technicians, only the first assistant showed recordings of 0.3 mSv body dose, 2 mSv ocular lens dose and 5 mSv hand dose.ConclusionThe yearly ocular lens dose, particularly for interventionists dealing with complex interventions, could cross the permitted yearly limit set by the new Euratom directive. Therefore, X-ray safety glasses would become mandatory for complex radiological vascular interventions.Level of EvidenceLevel III, non-randomized controlled cohort/follow-up study.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Impact of Patient and Lesion Characteristics on Drug-Coated Balloon Angioplasty in the Femoropopliteal Artery: A Pooled Analysis of Four Randomized Controlled Multicenter Trials

Objectives: The principal objective of this pooled analysis was to investigate various patient and lesion characteristics on late lumen loss (LLL) after drug-coated balloon (DCB) angioplasty. Background: Four randomized controlled trials (THUNDER, FEMPAC, PACIFIER, CONSEQUENT) were pooled to investigate the influence of various patient and lesion characteristics on DCB angioplasty and on plain old balloon angioplasty (POBA) in patients with femoropopliteal artery disease. Methods: Angiographic data from 355 patients were pooled to assess the impact of patient (demographics, cardiovascular risk factors, cardiovascular co-morbidities, Rutherford stages) and lesion-/procedure-related (location, occlusion, length, restenosis, calcification, subintimal crossing, post-dilatation, dissection, stenting) characteristics on LLL. Linear regression models were utilized with LLL as the dependent variable to determine the predictive value of cardiovascular and lesion-/procedure-related factors. Results: Observational statistics revealed that LLL was lower in the DCB group as compared to POBA independent of all tested patient variables. LLL after DCB was also independent of most lesion and procedural characteristics except for lesion length and bailout stenting. LLL increased with lesion length in both treatment groups. Bailout stenting did not improve LLL in the DCB group but did so in the POBA group (0.74 ± 1.07 mm vs. 1.22 ± 1.36 mm, p = 0.043). Conclusions: DCB was superior to POBA for all tested patient subgroups and lesion subgroups. Our results suggest that all patients and lesions benefit to a similar degree from the use of DCB. DCB-PTA should therefore be preferred to POBA in all patients with steno-occlusive femoropopliteal lesions.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Contrasting interannual and multidecadal NAO variability

Decadal and longer timescale variability in the winter North Atlantic Oscillation (NAO) has considerable impact on regional climate, yet it remains unclear what fraction of this variability is potentially predictable. This study takes a new approach to this question by demonstrating clear physical differences between NAO variability on interannual-decadal (<30 year) and multidecadal (>30 year) timescales. It is shown that on the shorter timescale the NAO is dominated by variations in the latitude of the North Atlantic jet and storm track, whereas on the longer timescale it represents changes in their strength instead. NAO variability on the two timescales is associated with different dynamical behaviour in terms of eddy-mean flow interaction, Rossby wave breaking and blocking. The two timescales also exhibit different regional impacts on temperature and precipitation and different relationships to sea surface temperatures. These results are derived from linear regression analysis of the Twentieth Century and NCEP-NCAR reanalyses and of a high-resolution HiGEM General Circulation Model control simulation, with additional analysis of a long sea level pressure reconstruction. Evidence is presented for an influence of the ocean circulation on the longer timescale variability of the NAO, which is particularly clear in the model data. Finally, as well as providing new evidence of potential predictability, these findings are shown to have implications for the reconstruction and interpretation of long climate records

54 ENVIRONMENTAL SCIENCES↗

Drivers and potential predictability of summer time North Atlantic polar front jet variability

The variability of the North Atlantic polar front jet stream is crucial in determining summer weather around the North Atlantic basin. Recent extreme summers in western Europe and North America have highlighted the need for greater understanding of this variability, in order to aid seasonal forecasting and mitigate societal, environmental and economic impacts. Here we find that simple linear regression and composite models based on a few predictable factors are able to explain up to 35 % of summertime jet stream speed and latitude variability from 1955 onwards. Sea surface temperature forcings impact predominantly on jet speed, whereas solar and cryospheric forcings appear to influence jet latitude. The cryospheric associations come from the previous autumn, suggesting the survival of an ice-induced signal through the winter season, whereas solar influences lead jet variability by a few years. Regression models covering the earlier part of the twentieth century are much less effective, presumably due to decreased availability of data, and increased uncertainty in observational reanalyses. Wavelet coherence analysis identifies that associations fluctuate over the study period but it is not clear whether this is just internal variability or genuine non-stationarity. Finally we identify areas for future research.

Meteorology & Atmospheric Sciences↗

Time-temperature-superposition analysis of diverse datasets by the minimum-arclength method: long-term prediction with uncertainty margins

In a recent publication, we carried out an extensive analysis of an unsupervised method of determining optimum shift factors in time-temperature-superposition of accelerated-aging data that involves minimizing the vertical arclength to obtain the master curve. For synthetic Arrhenius data with a variety of noise distributions, the work showed that, in conjunction with bootstrap-resampling, the method can produce reliable estimates of the mean activation energy along with uncertainty quantification. Here, we apply the above method to six different datasets taken from the published literature and demonstrate accurate prediction of mean activation energy from the data as-is without the need for any pre-processing or fitting. We also compare uncertainty margins computed by second-order bootstrap with that by linear regression theory and show that the former appears to provide consistent margins in the presence of common noise types in real data, including intra-isotherm measurement-errors, sample-to-sample variations, and intrinsic deviation from perfect Arrhenius behavior.

36 MATERIALS SCIENCE↗

Association of small vessel disease with tau pathology

Emerging evidence suggests that small vessel disease (SVD) is a risk factor for clinical dementia and may contribute to AD neuropathological changes. Watershed brain regions are located at the most distal areas between arterial territories, making them vulnerable to SVD-related changes. We examined the association of pathologic markers of SVD, specifically arteriolosclerosis in watershed brain regions, with AD pathologic changes. Participants (N= 982; mean age-at-death = 90; 69% women) were enrolled as part of one of two cohort studies of aging and dementia. At autopsy, neuropathological evaluation included semi-quantitative grading of arteriolosclerosis pathology from 2 cortical watershed regions: the anterior watershed (AWS) and posterior watershed (PWS), densities for cortical β-amyloid and tau-tangle pathology, and other common age-related pathologies. Linear regression models examined the association of watershed arteriolosclerosis pathology with β-amyloid and tau-tangle burden. In follow-up analyses, available ex-vivo MRI and proteomics data in a subset of decedents were leveraged to examine the association of whole brain measure of WMH, as a presumed MRI marker of SVD, with β-amyloid and tau-tangle burden, as well as to examine the association of watershed arteriolosclerosis with proteomic tau. Watershed arteriolosclerosis was common, with 45% of older persons having moderate-to-severe arteriolosclerosis pathology in the AWS region, and 35% in the PWS. In fully adjusted models that controlled for demographics and common age-related pathologies, an increase in severity of PWS arteriolosclerosis was associated with a higher burden of tau-tangle burden, specifically neocortical tau burden, but not with β-amyloid. AWS arteriolosclerosis was not associated with β-amyloid or tau pathology. Ex-vivo WMH was associated with greater tau-tangle pathology burden but not β-amyloid. Furthermore, PWS arteriolosclerosis was associated with higher abundance of tau phosphopeptides, that promote formation of tau aggregates. These data provide compelling evidence that SVD, specifically posterior watershed arteriolosclerosis pathology, is linked with tau pathological changes in the aging brain.

60 APPLIED LIFE SCIENCES↗

The solubility of titanite in silicate melt determined from growth and dissolution experiments

The solubility of titanite (CaTiSiO 5 ) in Si-rich melts was measured experimentally through growth experiments at 800–1000 °C, 0.5–1.0 GPa, log fO 2 ~ CCO–0.8, t = 72–168 h, and H 2 O = 0 to 4 wt.%, and in dissolution experiments at 925–1300 °C, 0.8 GPa, t = 18–118 h, and H 2 O = 1–10 wt.% in a piston-cylinder apparatus. Run product glasses in growth experiments were homogeneous, and iron loss suppressed ilmenite crystallization. Saturation concentrations in dissolution experiments were estimated by fitting measured diffusion profiles. Titanite solubility increases with increasing temperature and melt composition parameter \(M = {\text{ molar }}\left( {{\text{Na }} + {\text{ K }} + {\text{ 2Ca}}} \right)/\left( {{\text{Al }} \times {\text{ Si}}} \right)\) . Multiple linear regression of glass composition data from growth and dissolution experiments ( n = 29) plus 39 experiments from the LEPR database (Hirschmann et al. 2008 ) yielded the titanite solubility equation (adj. r 2 = 0.95): \(\left( {TiO_{2} } \right)^{melt} \left( {wt.\% } \right) = 0.978 \times M + 0.0048 \times T\left( K \right){-}5.90\) . This model correctly predicted undersaturation in 95% of 2344 experiments from the LEPR database that did not have titanite or rutile. Application to natural rocks yields saturation temperatures that are similar to independent temperature estimates. Finally, this equation should be useful for constraining the temperatures of titanite-saturated melts, for determining whether titanite saturation in magmatic source regions is likely, and for determining when titanite can crystallize and begin to exert an influence on melt geochemistry.

58 GEOSCIENCES↗

Temporal variation in the importance of a dominant consumer to stream nutrient cycling

Animal excretion can be a significant nutrient flux within ecosystems, where it supports primary production and facilitates microbial decomposition of organic matter. The effects of excretory products on nutrient cycling have been documented for various species and ecosystems, but temporal variation in these processes is poorly understood. We examined variation in excretion rates of a dominant grazing snail, Elimia clavaeformis, and its contribution to nutrient cycling, over the course of 14 months in a well-studied, low-nutrient stream (Walker Branch, east Tennessee, USA). Biomass-specific excretion rates of ammonium varied over twofold during the study, coinciding with seasonal changes in food availability (measured as gross primary production) and water temperature (multiple linear regression, R 2 = 0.57, P = 0.053). The contribution of ammonium excretion to nutrient cycling varied with seasonal changes in both biological (that is, nutrient uptake rate) and physical (that is, stream flow) variables. On average, ammonium excretion accounted for 58% of stream water ammonium concentrations, 26% of whole-stream nitrogen demand, and 66% of autotrophic nitrogen uptake. Phosphorus excretion by Elimia was contrastingly low throughout the year, supplying only 1% of total dissolved phosphorus concentrations. The high average N:P ratio (89:1) of snail excretion likely exacerbated phosphorus limitation in Walker Branch. To fully characterize animal excretion rates and effects on ecosystem processes, multiple measurements through time are necessary, especially in ecosystems that experience strong seasonality.

54 ENVIRONMENTAL SCIENCES↗

Hydrochemical and isotopic characteristics of groundwater in the northeastern Tennger Desert, northern China

Groundwater is typically the only water source in arid regions, and its circulation processes should be better understood for rational resource exploitation. Stable isotopes and major ions were investigated in the northeastern Tengger Desert, northern China, to gain insights into groundwater recharge and evolution. In the northern mountains, Quaternary unconsolidated sediments, exposed only in valleys between hills, form the main aquifer, which is mainly made of aeolian sand and gravel. Most of the mountain groundwater samples plot along the local meteoric water line (LMWL), with a more depleted signature compared to summer precipitation, suggesting that mountain groundwater was recharged by local precipitation during winter. Most of the groundwater was fresh, with total dissolved solids less than 1 g/L; dominant ions are Na{sup +}, SO{sub 4}{sup 2−} and Cl{sup −}, and all mineral saturation indices are less than zero. Evaporation, dissolution and cation exchange are the major hydrogeochemical processes. In the southern plains, however, the main aquifers are sandstone. The linear regression line of δD and δ{sup 18}O of groundwater parallels the LMWL but the intercept is lower, indicating that groundwater in the plains has been recharged by ancient precipitation rather than modern. Both calcite and dolomite phases in the plains groundwater are close to saturation, while gypsum and halite can still be dissolved into the groundwater. Different recharge mechanisms occur in the northern mountains and the southern plains, and the hydraulic connection between them is weak. Because of the limited recharge, groundwater exploitation should be limited as much as possible.

58 GEOSCIENCES↗

MABAL: a Novel Deep-Learning Architecture for Machine-Assisted Bone Age Labeling

Bone age assessment (BAA) is a commonly performed diagnostic study in pediatric radiology to assess skeletal maturity. The most commonly utilized method for assessment of BAA is the Greulich and Pyle method (Pediatr Radiol 46.9:1269–1274, 2016; Arch Dis Child 81.2:172–173, 1999) atlas. The evaluation of BAA can be a tedious and time-consuming process for the radiologist. As such, several computer-assisted detection/diagnosis (CAD) methods have been proposed for automation of BAA. Classical CAD tools have traditionally relied on hard-coded algorithmic features for BAA which suffer from a variety of drawbacks. Recently, the advent and proliferation of convolutional neural networks (CNNs) has shown promise in a variety of medical imaging applications. There have been at least two published applications of using deep learning for evaluation of bone age (Med Image Anal 36:41–51, 2017; JDI 1–5, 2017). However, current implementations are limited by a combination of both architecture design and relatively small datasets. The purpose of this study is to demonstrate the benefits of a customized neural network algorithm carefully calibrated to the evaluation of bone age utilizing a relatively large institutional dataset. In doing so, this study will aim to show that advanced architectures can be successfully trained from scratch in the medical imaging domain and can generate results that outperform any existing proposed algorithm. The training data consisted of 10,289 images of different skeletal age examinations, 8909 from the hospital Picture Archiving and Communication System at our institution and 1383 from the public Digital Hand Atlas Database. The data was separated into four cohorts, one each for male and female children above the age of 8, and one each for male and female children below the age of 10. The testing set consisted of 20 radiographs of each 1-year-age cohort from 0 to 1 years to 14–15+ years, half male and half female. The testing set included left-hand radiographs done for bone age assessment, trauma evaluation without significant findings, and skeletal surveys. A 14 hidden layer-customized neural network was designed for this study. The network included several state of the art techniques including residual-style connections, inception layers, and spatial transformer layers. Data augmentation was applied to the network inputs to prevent overfitting. A linear regression output was utilized. Mean square error was used as the network loss function and mean absolute error (MAE) was utilized as the primary performance metric. MAE accuracies on the validation and test sets for young females were 0.654 and 0.561 respectively. For older females, validation and test accuracies were 0.662 and 0.497 respectively. For young males, validation and test accuracies were 0.649 and 0.585 respectively. Finally, for older males, validation and test set accuracies were 0.581 and 0.501 respectively. The female cohorts were trained for 900 epochs each and the male cohorts were trained for 600 epochs. An eightfold cross-validation set was employed for hyperparameter tuning. Test error was obtained after training on a full data set with the selected hyperparameters. Using our proposed customized neural network architecture on our large available data, we achieved an aggregate validation and test set mean absolute errors of 0.637 and 0.536 respectively. To date, this is the best published performance on utilizing deep learning for bone age assessment. Our results support our initial hypothesis that customized, purpose-built neural networks provide improved performance over networks derived from pre-trained imaging data sets. We build on that initial work by showing that the addition of state-of-the-art techniques such as residual connections and inception architecture further improves prediction accuracy. This is important because the current assumption for use of residual and/or inception architectures is that a large pre-trained network is required for successful implementation given the relatively small datasets in medical imaging. Instead we show that a small, customized architecture incorporating advanced CNN strategies can indeed be trained from scratch, yielding significant improvements in algorithm accuracy. It should be noted that for all four cohorts, testing error outperformed validation error. One reason for this is that our ground truth for our test set was obtained by averaging two pediatric radiologist reads compared to our training data for which only a single read was used. This suggests that despite relatively noisy training data, the algorithm could successfully model the variation between observers and generate estimates that are close to the expected ground truth.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Predicting Chemical Reaction Barriers with a Machine Learning Model

In the past few decades, tremendous advances have been made in the understanding of catalysis at solid surfaces. Despite this, most discoveries of materials for improved catalytic performance are made by a slow trial and error process in an experimental laboratory. Computational simulations have begun to provide a way to rationally design materials for optimizing catalytic performance, but due to the high computational expense of calculating transition state energies, simulations cannot adequately screen the phase space of materials. In this work, we attempt to mitigate this expense by using a machine learning approach to predict the most expensive and most important parameter in a catalyst’s affinity for a reaction: the reaction barrier. Previous methods which used the step reaction energy as the only parameter in a linear regression had a mean absolute error (MAE) on the order of 0.4 eV, too high to be used predictively. Here in our work, we achieve a MAE of about 0.22 eV, a marked improvement towards the goal of computational prediction of catalytic activity.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Using diverse potentials and scoring functions for the development of improved machine-learned models for protein–ligand affinity and docking pose prediction

The advent of computational drug discovery holds the promise of significantly reducing the effort of experimentalists, along with monetary cost. More generally, predicting the binding of small organic molecules to biological macromolecules has far-reaching implications for a range of problems, including metabolomics. However, problems such as predicting the bound structure of a protein–ligand complex along with its affinity have proven to be an enormous challenge. In recent years, machine learning-based methods have proven to be more accurate than older methods, many based on simple linear regression. Nonetheless, there remains room for improvement, as these methods are often trained on a small set of features, with a single functional form for any given physical effect, and often with little mention of the rationale behind choosing one functional form over another. Moreover, it is not entirely clear why one machine learning method is favored over another. Here, we endeavor to undertake a comprehensive effort towards developing high-accuracy, machine-learned scoring functions, systematically investigating the effects of machine learning method and choice of features, and, when possible, providing insights into the relevant physics using methods that assess feature importance. Here, we show synergism among disparate features, yielding adjusted R 2 with experimental binding affinities of up to 0.871 on an independent test set and enrichment for native bound structures of up to 0.913. When purely physical terms that model enthalpic and entropic effects are used in the training, we use feature importance assessments to probe the relevant physics and hopefully guide future investigators working on this and other computational chemistry problems.

59 BASIC BIOLOGICAL SCIENCES↗

Development of Novel Biodegradable Enrofloxacin–Silica Composite for In Vitro Drug Release Kinetic Studies

A sustained drug delivery system is developed by using nonionic polymer to formulate drug release rate from silica capsules. To serve this purpose, silica capsules filled with poly(ethylene glycol) (PEG) were incorporated with a veterinary antibiotic drug enrofloxacin (ENF); as a model hydrophobic drug by using a general and facile sol–gel route. The physicochemical properties of the prepared drug-loaded composites were investigated by scanning electron microscope (SEM), nitrogen adsorption, Fourier transform infrared spectroscopy and thermal analysis (TGA). The impact of the media’s ionic strength on the drug release was evaluated over a range of 0–0.4 M to simulate the gastrointestinal feed in two physiological pH conditions. Sodium chloride was applied for ionic concentration adjustment due to its ability to salt out polymers in the midrange of the lyotropic series. Simultaneously, the drug release kinetics was evaluated by fitting experimental data to common empirical (zero-order, first order and Higuchi) and semi-empirical (Ritger–Peppas and Sahlin–Peppas) models. The drug release kinetics from capsules revealed a non-Fickian diffusion and pure relaxation-controlled release. Of these models, Sahlin–Peppas equation best fit the release data of ENF. To determine the best model, non-linear regressions were carried out.

36 MATERIALS SCIENCE↗

Multi-region Modeling of Daily Global Solar Radiation with Artificial Intelligence Ensemble

Solar radiation data are crucial for the design and evaluation of solar energy systems, climatic studies, water resources management, estimating crop productivity, etc. As so, for locations where direct measurements are not available, reliable models may be developed to estimate solar radiation from more readily available data. In this study, two artificial intelligence (AI)-based models including artificial neural network and adaptive neuro-fuzzy inference systems, three temperature-based empirical models including Meza–Varas, Hargreaves–Samani, and Chen, and a conventional multi-linear regression (MLR) model were employed for multi-region daily global solar radiation estimation for Iraq. To ensure appropriate selection of input variables, sensitivity analysis was conducted to determine the dominant parameters. Finally, two ensemble approaches, neural average ensemble and simple average ensemble, were applied to improve the performance of the single models. For this purpose, daily meteorological data of maximum temperature (T{sub max}), minimum temperature (T{sub min}), mean temperature (T{sub mean}), relative humidity (R{sub H}), and wind speed (U{sub 2}) were obtained from January 2006 to December 2016 from four major cities in Iraq representing, north, west, south, and east regions. The results revealed that temperatures (T{sub max}, T{sub mean}, T{sub min}) and relative humidity are the dominant parameters. While temperature-based empirical models and MLR model could be employed to achieve reliable results, AI-based models are superior in performance to other models. Also promising improvement in daily global solar radiation modeling could be achieved by model ensemble. The results of this study affirmed that the provided ensemble approaches can increase the performance of single models up to 19.19%, 7.59%, and 16.81% in training, validation, and testing steps, respectively.

58 GEOSCIENCES↗

A Case Study on a Combination NDVI Forecasting Model Based on the Entropy Weight Method

It is critically meaningful to accurately predict NDVI (Normalized Difference Vegetation Index), which helps guide regional ecological remediation and environmental managements. In this study, a combination forecasting model (CFM) was proposed to improve the performance of NDVI predictions in the Yellow River Basin (YRB) based on three individual forecasting models, i.e., the Multiple Linear Regression (MLR), Artificial Neural Network (ANN), and Support Vector Machine (SVM) models. The entropy weight method was employed to determine the weight coefficient for each individual model depending on its predictive performance. Results showed that: (1) ANN exhibits the highest fitting capability among the four orecasting models in the calibration period, whilst its generalization ability becomes weak in the validation period; MLR has a poor performance in both calibration and validation periods; the predicted results of CFM in the calibration period have the highest stability; (2) CFM generally outperforms all individual models in the validation period, and can improve the reliability and stability of predicted results through combining the strengths while reducing the weaknesses of individual models; (3) the performances of all forecasting models are better in dense vegetation areas than in sparse vegetation areas.

54 ENVIRONMENTAL SCIENCES↗

Targeted metabolomic analysis identifies increased serum levels of GABA and branched chain amino acids in canine diabetes

Introduction Dogs with naturally occurring diabetes mellitus represent a potential model for human type 1 diabetes, yet signifcant knowledge voids exist in terms of the pathogenic mechanisms underlying the canine disorder. Untargeted metabolomic studies from a limited number of diabetic dogs identifed similarities to humans with the disease. Objective To expand and validate earlier metabolomic studies, identify metabolites that difer consistently between diabetic and healthy dogs, and address whether certain metabolites might serve as disease biomarkers. Methods Untargeted metabolomic analysis via liquid chromatography-mass spectrometry was performed on serum from diabetic (n=15) and control (n=15) dogs. Results were combined with those of our previously published studies using identical methods (12 diabetic and 12 control dogs) to identify metabolites consistently diferent between the groups in all 54 dogs. Thirty-two candidate biomarkers were quantifed using targeted metabolomics. Biomarker concentrations were compared between the groups using multiple linear regression (corrected P<0.0051 considered signifcant). Results Untargeted metabolomics identifed multiple persistent diferences in serum metabolites in diabetic dogs compared with previous studies. Therefore, targeted metabolomics showed increases in gamma amino butyric acid, valine, leucine, isoleucine, citramalate, and 2-hydroxyisobutyric acid in diabetic versus control dogs while indoxyl sulfate, N-acetyl-L-aspartic acid, kynurenine, anthranilic acid, tyrosine, glutamine, and tauroursodeoxycholic acid were decreased. Conclusion Several of these fndings parallel metabolomic studies in both human diabetes and other animal models of this disease. Given recent studies on the role of GABA and branched chain amino acids in human diabetes, the increase in serum concentrations in canine diabetes warrants further study of these metabolites as potential biomarkers, and to identify similarity in mechanisms underlying this disease in humans and dogs.

59 BASIC BIOLOGICAL SCIENCES↗