Support Vector Machine Analysis Research Articles

Disrupted Cerebellar-Default Mode Network Functional Connectivity in Major Depressive Disorder With Gastrointestinal Symptoms.

Gastrointestinal (GI) symptoms are one of the common somatic symptoms presented in patients with major depressive disorder (MDD). Higher frequency of GI symptoms and higher GI symptom burden were linked to greater depression severity and increased risk of suicide ideation. However, few studies have explored the underlying mechanisms of GI symptoms in MDD. Based on previous studies, the cerebellar-DMN circuits may play a potentially critical role in GI symptoms comorbid with depression. Fifty-two first-episode drug-naive patients with MDD (35 with GI symptoms and 17 without GI symptoms) and 28 matched healthy controls were recruited in the current study and underwent resting-state functional magnetic resonance imaging scan. Cerebellar seed-based functional connectivity maps were established. Relative to depressed patients without GI symptoms, significantly increased cerebellar-anterior default mode network (DMN) connectivities were found in those with GI symptoms. Both increased and decreased functional connectivities were found between cerebellum and posterior DMN in patients with GI symptoms compared with those without GI symptoms and healthy controls. Moreover, the right Crus I - right superior temporal gyrus connectivity value was related to severity of GI symptoms and depression in all patients with MDD. The support vector machine analysis demonstrated a satisfactory classification accuracy (89%) of the disrupted cerebellar-DMN connectivities for correctly identifying MDD patients with GI symptoms. These results revealed the possible neural mechanisms for the involvement of cerebellar-DMN circuits in GI symptoms co-occurred with MDD.

Analyzing Classification Performance of fNIRS-BCI for Gait Rehabilitation Using Deep Neural Networks.

This research presents a brain-computer interface (BCI) framework for brain signal classification using deep learning (DL) and machine learning (ML) approaches on functional near-infrared spectroscopy (fNIRS) signals. fNIRS signals of motor execution for walking and rest tasks are acquired from the primary motor cortex in the brain’s left hemisphere for nine subjects. DL algorithms, including convolutional neural networks (CNNs), long short-term memory (LSTM), and bidirectional LSTM (Bi-LSTM) are used to achieve average classification accuracies of 88.50%, 84.24%, and 85.13%, respectively. For comparison purposes, three conventional ML algorithms, support vector machine (SVM), k-nearest neighbor (k-NN), and linear discriminant analysis (LDA) are also used for classification, resulting in average classification accuracies of 73.91%, 74.24%, and 65.85%, respectively. This study successfully demonstrates that the enhanced performance of fNIRS-BCI can be achieved in terms of classification accuracy using DL approaches compared to conventional ML approaches. Furthermore, the control commands generated by these classifiers can be used to initiate and stop the gait cycle of the lower limb exoskeleton for gait rehabilitation.

Probing the proteome to explore potential correlates of increased Alzheimer's-related cerebrovascular disease in adults with Down syndrome.

Cerebrovascular disease is associated with symptoms and pathogenesis of Alzheimer's disease (AD) among adults with Down syndrome (DS). The cause of increased dementia-related cerebrovascular disease in DS is unknown. We explored whether protein markers of neuroinflammation are associated with markers of cerebrovascular disease among adults with DS. Participants from the Alzheimer's disease in Down syndrome (ADDS) study with magnetic resonance imaging (MRI) scans and blood biomarker data were included. Support vector machine (SVM) analyses examined the relationship of blood-based proteomic biomarkers with MRI-defined cerebrovascular disease among participants characterized as having cognitive decline (n=36, mean age ± SD=53±6.2) and as being cognitively stable (n=78, mean age=49±6.4). Inflammatory and AD markers were associated with cerebrovascular disease, particularly among symptomatic individuals.The pattern suggested relatively greater inflammatory involvement among cognitively stable individuals and greater AD involvement among those with cognitively decline. The findings help to generate hypotheses that both inflammatory and AD markers are implicated in cerebrovascular disease among those with DS and point to potential mechanistic pathways for further examination.

Machine Learning Model to Stratify the Risk of Lymph Node Metastasis for Early Gastric Cancer: A Single-Center Cohort Study.

Simple SummaryEndoscopic resection (ER) is a treatment option for clinically T1a early gastric cancer (EGC) without suspicion of lymph node metastasis (LNM). In patients with non-curative resection after ER, additional surgery is recommended owing to the LNM risk. However, of those patients treated with additional surgery after ER, the actual rate of LNM was about 5–10%; that is, the other patients underwent unnecessary surgeries. Therefore, it is crucial to estimate LNM risk in EGC patients to determine additional management after ER. We derived a machine learning (ML) model to stratify the LNM risk in EGC patients and validate its performance. The constructed ML model, which showed good performance with an area under the receiver operating characteristic of 0.85 or higher, could stratify LNM risk into very low (<1%), low (<3%), intermediate (<7%), and high (≥7%) risk categories. These findings suggest that the ML model can stratify the LNM risk in EGC patients.Stratification of the risk of lymph node metastasis (LNM) in patients with non-curative resection after endoscopic resection (ER) for early gastric cancer (EGC) is crucial in determining additional treatment strategies and preventing unnecessary surgery. Hence, we developed a machine learning (ML) model and validated its performance for the stratification of LNM risk in patients with EGC. We enrolled patients who underwent primary surgery or additional surgery after ER for EGC between May 2005 and March 2021. Additionally, patients who underwent ER alone for EGC between May 2005 and March 2016 and were followed up for at least 5 years were included. The ML model was built based on a development set (70%) using logistic regression, random forest (RF), and support vector machine (SVM) analyses and assessed in a validation set (30%). In the validation set, LNM was found in 337 of 4428 patients (7.6%). Among the total patients, the area under the receiver operating characteristic (AUROC) for predicting LNM risk was 0.86 in the logistic regression, 0.85 in RF, and 0.86 in SVM analyses; in patients with initial ER, AUROC for predicting LNM risk was 0.90 in the logistic regression, 0.88 in RF, and 0.89 in SVM analyses. The ML model could stratify the LNM risk into very low (<1%), low (<3%), intermediate (<7%), and high (≥7%) risk categories, which was comparable with actual LNM rates. We demonstrate that the ML model can be used to identify LNM risk. However, this tool requires further validation in EGC patients with non-curative resection after ER for actual application.

Dimensionless Artificial Intelligence-Based Model for Multiphase Flow Pattern Recognition in Horizontal Pipe

Summary Multiphase flow analysis attracts a lot of attention from researchers from diverse disciplines. There are several studies including experimental, theoretical modeling, and numerical analysis that were carried out to investigate the multiphase flow. However, many facets of multiphase flow are still unresolved owing to the extremely complex nature of the multiphase flow. The complex interactions of the different phases are leading to different flow regimes that are difficult to predict but essential for developing the computational model. The identification of the flow pattern is still a challenging task. One of the growing fields is the machine learning approach, which can address such complex problems. This study aims to use machine learning to develop models that can identify the flow patterns in multiphase flow. To achieve the objective, a large set of experimental data was collected. The effect of fluid properties, such as density, viscosity, and surface tension, on the flow pattern was introduced by changing the fluid properties. The wide range of data was processed by applying a machine learning technique for predicting the flow regimes. The models were built using dimensionless parameters to extend their validity for various design and operational conditions. This approach enables to capture the main flow pattern as well as subcategories of flow patterns in the horizontal pipe. Comparison and analyses of the different machine learning tools were carried out to investigate classification of multiphase flow patterns. The results showed that different artificial intelligence (AI) methods can predict the flow pattern in horizontal pipes with high accuracy. The results of using Reynold’s number for liquid (ReL) and gas (ReG) as an input to predict the flow patterns are deficient in accuracy for the support vector machine (SVM) and discriminant analysis (DA). However, the prediction capability of the model was improved by introducing Weber’s number for liquid (WeL) and gas (WeG) along with the Reynolds numbers (ReL and ReG). The improvement in the flow pattern prediction owing to the introduction of Weber’s number is speculated because of the capturing hydrodynamic phenomenon (inertia and surface tension) owing to change in fluid properties. It infers that capturing hydrodynamic phenomena affecting the flow pattern and their transition is essential for the prediction of flow patterns in multiphase flow.

Abnormal Degree Centrality as a Potential Imaging Biomarker for Right Temporal Lobe Epilepsy: A Resting-state Functional Magnetic Resonance Imaging Study and Support Vector Machine Analysis

Previous studies have reported altered neuroimaging features in right temporal lobe epilepsy (rTLE). However, the alterations in degree centrality (DC) as a diagnostic method for rTLE have not been reported. Therefore, we aimed to explore abnormalities in the DC of the rTLE and whether such alterations could be applied to the diagnosis of rTLE. Resting-state functional magnetic resonance imaging (fMRI) was used to scan 82 patients with rTLE and 69 healthy controls. The DC and support vector machine (SVM) methods were used for an analysis of the imaging data. Compared to the control group, the rTLE patients exhibited lower DC values in the right hippocampus, right superior temporal gyrus, and right caudate. Compared to the control group, the rTLE patients showed higher DC values in the right medial superior frontal gyrus (SFGmed), left dorsolateral superior frontal gyrus (SFGdor), right inferior parietal lobule (IPL), and the left postcentral. The highest diagnostic accuracy of 99.34% (150/151), based on SVM analysis, was demonstrated for the combination of abnormal DC in the right IPL and the left SFGdor, along with a sensitivity of 100% (82/82), and a specificity of 98.55% (68/69) for the differentiation of rTLE patients from healthy controls. The study demonstrated abnormal functional connectivity in rTLE patients. Thus, a distinctive DC pattern may serve as an imaging marker for the diagnosis of rTLE patients.

Comprehensive analysis of peripheral blood non-coding RNAs identifies a diagnostic panel for fungal infection after transplantation

The occurrence of fungal infection seriously affects the survival and life quality of transplanted patients. The accurate diagnosis is of particular importance in the early stage of infection. To develop a novel diagnostic method for this kind of patient, we established a post-transplant immunosuppressed mice model with fungus inoculation and collected their peripheral blood at specific time points after infection. After screening by microarray, differentially expressed miRNAs and lncRNAs were selected and homologously analyzed with those of human beings from the gene database. These miRNAs and lncRNAs candidates were validated by qRT-PCR in peripheral blood samples from transplanted patients. We found that, compared with normal transplanted patients, the levels of miR-215 and miR-let-7 c were up-regulated in the plasma of patients with fungal infection (P < 0.01), while levels of miR-154, miR-193a, NR_027669.1, and NR_036506.1 were down-regulated in their peripheral blood mononuclear cells (P < 0.01). Principal component analysis shows that the expression pattern of the above RNAs was different between the two groups. A 6-noncoding-RNA detection panel was established by the support vector machine analysis, whose area under the ROC curve was 0.927. The accuracy, precision, sensitivity, and specificity of this model were 0.928, 0.919, 0.944, and 0.910, respectively. Though our detection panel has excellent diagnostic efficacy, its clinical application value still needs to be further confirmed by multi-center prospective clinical trials.

A Sustainable Model for Emergency Medical Services in Developing Countries: A Novel Approach Using Partial Outsourcing and Machine Learning.

IntroductionUnlike Western countries, many low- and middle-income countries (LMIC), like India, have a de-centralized emergency medical services (EMS) involving both semi-government and non-government organizations. It is alarming that due to the absence of a common ecosystem, the utilization of resources is inefficient, which leads to shortage of available vehicles and larger response time. Fragmentation of emergency supply chain resources motivates us to propose a new vehicle routing and scheduling model equipped with novel features to ensure minimal response time using existing resources.Materials and MethodsThe data set of medical and fire-related emergencies from January 2018 to May 2018 of Uttarakhand State in India was provided by GVK Emergency Management and Research Institute (GVK EMRI) also known as 108 EMSs was used in the study. The proposed model integrates all the available EMS vehicles including partial outsourcing to non-ambulatory vehicles like police vans, taxis, etc., using a novel two-echelon heuristic approach. In the first stage, an offline learning model is developed to yield the deployment strategy for EMS vehicles. Seven well researched machine learning (ML) algorithms were analyzed for parameter prediction namely random forest (RF), convolutional neural network (CNN), k-nearest neighbor (KNN), classification and regression tree (CART), support vector machine (SVM), logistic regression (LR), and linear discriminant analysis (LDA). In the second stage, a real-time routing model is proposed for EMS vehicle routing at the time of emergency, considering partial outsourcing.Results and DiscussionThe results indicate that the RF classifier outperforms the LR, LDA, SVM, CNN, CART and NB classifier in terms of both accuracy as well as F-1 score. The proposed vehicle routing and scheduling model for automated decision-making shows an improvement of 42.1%, 54%, 27.9% and 62% in vehicle assignment time, vehicle travel time from base to scene, travel time from scene to hospital, and total response time, respectively, in urban areas.

Machine Learning-Based Cry Diagnostic System for Identifying Septic Newborns

Processing the newborns' cry audio signal (CAS) provides valuable information about the newborns' condition. This information can be used to diagnose the disease. This article analyzes the CASs of newborns under two months old using machine learning approaches to develop an automatic diagnostic system for identifying septic infants from healthy ones. Septic infants have not been studied in this context. The proposed features include Mel frequency cepstral coefficients and the prosodic features of tilt, rhythm, and intensity. The performance of each feature set was evaluated using a collection of classifiers, including Support Vector Machine (SVM), decision tree, and discriminant analysis. We also examined the majority voting method for improving the classification results and feature manipulation and multiple classifier framework, which has not previously been reported in the literature on developing an automatic diagnostic system based on the infant's CAS. We tested our methodology on two datasets of expiration and inspiration episodes of newborns' CASs. The framework of the concatenation of all feature sets using quadratic SVM resulted in the best F-score with 86% for the expiration dataset. Furthermore, the framework of tilt feature set with quadratic discriminant with 83.90% resulted in the best F-score for inspiration. We found out that septic infants cry differently than healthy infants through these experiments. Thus, our proposed method can be used as a noninvasive tool for identifying septic infants from healthy ones only based on their CAS.

Serum Inflammatory Biomarkers Contribute to the Prognosis Prediction in High-Grade Glioma.

BackgroundTo evaluate the prognostic value of serum inflammatory biomarkers and develop a risk stratification model for high-grade glioma (HGG) patients based on clinical, laboratory, radiological, and pathological factors.Materials and MethodsA retrospective study of 199 patients with HGG was conducted. Patients were divided into a training cohort (n = 120) and a validation cohort (n = 79). The effects of potential associated factors on the overall survival (OS) time were investigated and the benefits of serum inflammatory biomarkers in improving predictive performance was assessed. Univariable and multivariable Cox regression analyses, the least absolute shrinkage and selection operator (LASSO) regression analysis, and support vector machines (SVM) were used to select variables for the final nomogram model.ResultsAfter multivariable Cox, LASSO, and SVM analysis, in addition to 3 other clinico-pathologic factors, platelet-to-lymphocyte ratio (PLR) >144.4 (hazard ratio [HR], 2.05; 95% confidence interval [CI], 1.25–3.38; P = 0.005) were left for constructing the predictive model. The model with PLR exhibited a better predictive performance than that without them in both cohorts. The nomogram based on the model showed an excellent ability of discrimination in the entire cohort (C-index, 0.747; 95%CI, 0.706–0.788). The calibration curves showed good consistency between the predicted and observed survival probability.ConclusionOur study confirmed the prognostic value of serum inflammatory biomarkers including PLR and established a comprehensive scoring system for the OS prediction in HGG patients.

Detection of breast cancer by ATR-FTIR spectroscopy using artificial neural networks.

In this study, three (3) neural networks (NN) were designed to discriminate between malignant (n = 78) and benign (n = 88) breast tumors using their respective attenuated total reflection Fourier transform infrared (ATR-FTIR) spectral data. A proposed NN-based sensitivity analysis was performed to determine the most significant IR regions that distinguished benign from malignant samples. The result of the NN-based sensitivity analysis was compared to the obtained results from FTIR visual peak identification. In training each NN models, a 10-fold cross validation was performed and the performance metrics–area under the curve (AUC), accuracy, positive predictive value (PPV), specificity rate (SR), negative predictive value (NPV), and recall rate (RR)–were averaged for comparison. The NN models were compared to six (6) machine learning models–logistic regression (LR), Naïve Bayes (NB), decision trees (DT), random forest (RF), support vector machine (SVM) and linear discriminant analysis (LDA)–for benchmarking. The NN models were able to outperform the LR, NB, DT, RF, and LDA for all metrics; while only surpassing the SVM in accuracy, NPV and SR. The best performance metric among the NN models was 90.48% ± 10.30% for AUC, 96.06% ± 7.07% for ACC, 92.18 ± 11.88% for PPV, 94.19 ± 10.57% for NPV, 89.04% ± 16.75% for SR, and 94.34% ± 10.54% for RR. Results from the proposed sensitivity analysis were consistent with the visual peak identification. However, unlike the FTIR visual peak identification method, the NN-based method identified the IR region associated with C–OH C–OH group carbohydrates as significant. IR regions associated with amino acids and amide proteins were also determined as possible sources of variability. In conclusion, results show that ATR-FTIR via NN is a potential diagnostic tool. This study also suggests a possible more specific method in determining relevant regions within a sample’s spectrum using NN.

Incidence and predictors of cardiogenic shock following surgical or transcatheter tricuspid valve intervention

AbstractObjectiveThe objective of this study was to evaluate the incidence of and risk factors associated with cardiogenic shock (CS) following surgery versus transcatheter tricuspid valve intervention (TTVI) for tricuspid regurgitation (TR).BackgroundSurgical therapy for TR is associated with high rates of CS. Postprocedural shock has not been studied following TTVI.MethodsWe performed a single‐center retrospective analysis of isolated tricuspid valve (TV) surgery or TTVI for TR. The primary outcome was postprocedural class D or E CS according to Society for Cardiovascular Angiography and Interventions (SCAI) CS classification scheme, and secondary outcome was in‐hospital mortality. Multivariable logistic regression modeling was performed for primary and secondary outcomes. Support vector machine analysis was performed for sensitivity analysis.ResultsFrom 2008 to 2020, a total of 122 patients underwent isolated TV surgery (n = 58, 14 TV repair, and 44 TV replacement) or TTVI (n = 64, 36 TV repair, and 28 TV replacement). Surgical patients were significantly younger than TTVI patients (67.5 vs. 80 years, p &lt; 0.0001). Multivariable modeling revealed an association between the primary outcome and surgery (odds ratio [OR]: 8.75, 95% confidence interval [CI]: 2.83, 27.03, p = 0.0002), as well as baseline central venous pressure (CVP, OR: 1.12, 95% CI: 1.02, 1.22, p = 0.016). Additionally, class DE CS was independently associated with in‐hospital mortality (OR: 5.21, 1.35, 20.09, p = 0.016). CVP and surgery were found to have highest importance indices in support vector machine analysis.ConclusionIn patients undergoing TV intervention for TR, surgery versus TTVI and elevated CVP are associated with advanced postprocedural CS. Patients developing advanced CS are at increased risk of in‐hospital mortality.

Comparative Analysis of Support Vector Machine and EGARCH Modelling of Zenith Bank Plc. Stock Price on Economic Growth in Nigeria

In this 21st century, stock exchange has become a crucial and determining factor for the global economy, any variability in this market affects personal, corporate financial activities and the economic growth a country. A lot of analysis using different models were explored by several authors. However, this study set out to investigates the statistical behaviour of Zenith Bank PLC stock price using Support Vector Machine (SVM) and EGARCH Model on economic growth in Nigeria by adopting quantitative techniques using the causal-comparative research method. The findings from this research shows a coefficient of the Zenith bank Low stock price at (0.546945) and high stock price at (0.453562) which indicates that low stock price is insignificant because is less than α = 0.05, while high stock price of greater than α = 0.05 is significant to predict the Zenith bank stock price. Findings from the study has further revealed that SVM mode approach is more appropriate than EGARCH model and recommends that stockholders, financial analyst and researchers who are investing in stock market should adapt the SVM model approach to determine the volatility rate or level of stochastic in the financial time series data.&#x0D; Keywords: Stock Price, Support Vector Machine (SVM), EGARCH Model, Stock Exchange, Economic Growth

Abnormal global-brain functional connectivity and its relationship with cognitive deficits in drug-naive first-episode adolescent-onset schizophrenia.

Abnormal functional connectivity (FC) has been reported in drug-naive first-episode adolescent-onset schizophrenia (AOS) with inconsistent results due to differently selected regions of interest. The voxel-wise global-brain functional connectivity (GFC) analysis can help explore abnormal FC in an unbiased way in AOS. A total of 48 drug-naive first-episode AOS as well as 31 sex-, age- and education-matched healthy controls were collected. Data were subjected to GFC, correlation analysis and support vector machine analyses. Compared with healthy controls, the AOS group exhibited increased GFC in the right middle frontal gyrus (MFG), and decreased GFC in the right inferior temporal gyrus, left superior temporal gyrus (STG)/precentral gyrus/postcentral gyrus, right posterior cingulate cortex /precuneus and bilateral cuneus. After the Benjamini-Hochberg correction, significantly negative correlations between GFC in the bilateral cuneus and Trail-Making Test: Part A (TMT-A) scores (r=-0.285, p=0.049), between GFC in the left STG/precentral gyrus/postcentral gyrus and TMT-A scores (r=-0.384, p=0.007), and between GFC in the right MFG and the fluency scores (r=-0.335, p=0.020) in the patients. GFC in the left STG/precentral gyrus/postcentral gyrus has a satisfactory accuracy (up to 86.08%) in classifying patients from controls. AOS shows abnormal GFC in the brain areas of multiple networks, which bears cognitive significance. These findings suggest potential abnormalities in processing self-monitoring and sensory prediction, which further elucidate the pathophysiology of AOS.

A comparative analysis of artificial neural network and support vector machine for online transient stability prediction considering uncertainties

ABSTRACT Power system transient stability is an integral part of power system planning and operation. Conventional approaches to assess transient stability are time consuming and hence, are not suitable for online application. Thus, this paper presents a comparative analysis of two different machine learning (ML) algorithms, i.e. artificial neural network (ANN) and support vector machine (SVM), for online transient stability prediction, considering various uncertainties (load, network topology, fault type, fault location, and fault clearing time). Time-domain simulations were conducted using DIgSILENT PowerFactory software for obtaining the training data for ML algorithms. MATLAB was used to apply the ML algorithms (ANN and SVM), and to draw a comparison between them. The classification accuracy of ANN and SVM was found to be 98.4% and 94.4% (fixed topology), respectively. The classification accuracy of ANN and SVM was found to be 97.7% and 93.2% (topology change), respectively. The training time of ANN was less than that of SVM (for both topology cases). These results for the IEEE 14-bus system demonstrated that both ANN and SVM can rapidly estimate the transient stability, considering uncertainties, with a reasonable accuracy; however, ANN outperformed SVM as its classification performance and computational performance was determined to be superior.

Are dolphins modulating whistles in interspecific group contexts?

ABSTRACT Cetaceans depend highly on acoustic signals, such as whistles, for social communication. Mixed-species associations (MSAs), temporary or long-term encounters between different species, may play an important role on the acoustic dynamics of the species involved. Because of whistles’ communication function, we used these sounds to investigate the acoustic relationships of Tursiops truncatus when involved in different group contexts: in MSAs with two other delphinid species and in single species group. Acoustic recordings of T. truncatus single species groups and in associations with Globicephala melas and Grampus griseus were collected in the Western South Atlantic Ocean through a towed array of hydrophones. A total of 1267 whistles were analysed and acoustical parameters were extracted. To verify the differences among whistles produced in such contexts a support vector machine and random forest analysis were implemented. Both analyses revealed a clear separation of whistles from the single species versus the MSA as well as between both MSAs. The results indicate that interspecific associations may influence the whistle structure and suggests that T. truncatus whistles can be modified during interspecific interactions. These findings are useful to elucidate the possible factors underlying behavioural plasticity and interspecific associations and to understand dolphins’ acoustic communication.

Machine Learning Applied to Omics Datasets Predicts Mortality in Patients with Alcoholic Hepatitis.

Alcoholic hepatitis is a major health care burden in the United States due to significant morbidity and mortality. Early identification of patients with alcoholic hepatitis at greatest risk of death is extremely important for proper treatments and interventions to be instituted. In this study, we used gradient boosting, random forest, support vector machine and logistic regression analysis of laboratory parameters, fecal bacterial microbiota, fecal mycobiota, fecal virome, serum metabolome and serum lipidome to predict mortality in patients with alcoholic hepatitis. Gradient boosting achieved the highest AUC of 0.87 for both 30-day mortality prediction using the bacteria and metabolic pathways dataset and 90-day mortality prediction using the fungi dataset, which showed better performance than the currently used model for end-stage liver disease (MELD) score.

A new estimate of global desert area and quantity ofdust emission

<p indent=0mm>Deserts and their margins have long presented a challenge for human habitation, as an unstable and easily degraded environment. Deserts are the most important source of global dust emission, and the dust flux can be used as an index of land degradation and climate change, as well as a way to assess the dust’s own effects on global climate. In this study, we used the Moderate Resolution Imaging Spectroradiometer (MODIS) data, combing with Support Vector Machine (SVM) analysis, trial-and-error method and visual interpretation analyses, to extract the global and regional desert area, and analyze the accuracy of classification of the extracted desert area. We use data on topography, precipitation, evapotranspiration, and surface wind speed to characterize the environment of the world’s deserts, in order to explore the natural conditions that produce and accompany desert and produce a clear objective definition for desert environment. Our results show that global desert area is <sc>~17.5×10⁶ km²,</sc> and the overall classificatory accuracy is 92%, according to the cross-checked by the multi sources data calibrations. Globally, desert is mainly distributed in eastern hemisphere and northern hemisphere; North Africa and East Asia have the largest desert area. The classificatory accuracy of all individual regions is greater than 75%. Global desert, as characterized by our multi-faceted definition, is mainly distributed in low regions below<sc>2000 m</sc> elevation, mean annual precipitation less than <sc>200 mm,</sc> and annual wind speed is about <sc>4–6 m/s.</sc> Based on our defined distribution of global deserts, we estimated dust emissions from those areas. Because dust emission is a small-scale, random process, it is necessary to collect field data of dust emission from typical landscape types, in order to describe dust emission and quantitatively assess their impact. We used data from the Portable <italic>In-Situ</italic> Wind Erosion Laboratory (PI-SWERL), which is a newly developed portable wind erosion instrument. Its observation data are in good agreement with the wind tunnel experiment results. Because of its small size and portability, it has great advantages in measuring dust emission at specific sites. We integrated measured data about 1400 PI-SWERL observation points of dust emission (PM₁₀), combing with land cover type, soil moisture and friction wind speed data, and obtained about 14000 points quantifying dust emission through interpolation. These 14000 dust emission amounts points are distributed across different areas of the desert. Using these 14000 points to make Tyson polygons, we estimated the annual dust emission amount of each polygon area, and then added each annual dust emission amounts polygon area to obtain the annual dust emission amount of global desert. Dust emission amount shows considerable geographical differences between different desert areas. The region around the Caspian Sea and Aral Sea in central Asia, Mesopotamia in western Asia, sandy areas in northern West Asia and eastern Australia have the highest annual dust emission rates. This result is in good agreement with previous published results on global dust source regions. The annual dust emission amounts in desert area are ~1792.65 Tg/a, and North Africa has the largest annual dust emission amounts, ~786.03 Tg/a, accounting for 43.85% of the total dust emission amounts of global desert. According to the relationship between dust deposition in land and sea, we estimate that ~1344.49 Tg/a dust is deposited in land and ~448.16 Tg/a in ocean. The global desert area interpreted from remote sensing images using our new standardized protocol, and our dust emission estimates, will facilitate further understanding of the environmental changes within deserts and their impacts.

Identification of peanut oil origins based on Raman spectroscopy combined with multivariate data analysis methods

This study aimed to use Raman spectroscopy to identify the producing areas of peanut oil and build a robust discriminant model to further screen out the characteristic spectra closely related to the origin. Raman spectra of 159 peanut oil samples from different provinces and different cities of the same province were collected. The obtained data were analyzed by stepwise linear discriminant analysis (SLDA), k-nearest neighbor analysis (k-NN), support vector machine (SVM) and multi-way analysis of variance. The results showed that the overall recognition rate of samples based on full spectra was higher than 90%. The producing origin, variety and their interaction influenced Raman spectra of peanut oil significantly, and 1 400–1 500 cm−1 and 1 600–1 700 cm−1 were selected as the characteristic spectra of origin and less affected by variety. The best classification model established by SLDA combined with characteristic spectra could rapidly and accurately identify peanut oil's origin.

A Feature Selection Approach Hybrid Grey Wolf and Heap-Based Optimizer Applied in Bearing Fault Diagnosis

An effective bearing fault diagnosis model based on machine learning is proposed in this study. The model can separate into three stages: feature extraction, feature selection, and classification. In the stage of feature extraction, multiresolution analysis (MRA) and fast Fourier transform (FFT) are applied to extract the features from the raw signal measured from the rotating machine. In the second stage, a powerful feature selection method is proposed and utilized in the stage of feature selection. The new feature selection method is based on grey wolf optimization (GWO) and heap-based optimizer (HBO) with new strategies combined. Finally, support vector machine (SVM) and linear discriminant analysis (LDA) are used as the classifier independently. To verify the capability of the proposed model, four different datasets are applied to test the model in this study, respectively University of California Irvine (UCI) benchmark dataset, bearing dataset, Case Western Reserve University (CWRU) benchmark dataset, and Machinery Failure Prevention Technology (MFPT) benchmark dataset. The proposed method is compared with the existing methods and can certify the robustness with the experiment results.