Linear Support Vector Machine Model Research Articles

Impact of Velocity of Detonation and Charge per Bank Cubic Meters on Flyrock Throw Prediction Using Support Vector Machine

One of the ambient effects of production blasting is flyrock. To effectively manage flyrock throw distance in mining, there is the necessity to successfully envisage blasting output without sacrificing the hazardous impact of flyrock which may result in fatality and operational shutdown. For flyrock throw distance prediction, velocity of detonation (VOD) and charge per bank cubic meter (CPBCM) are not usually included. This paper focuses on the use of support vector machine (SVM) regression to ascertain the impact of VOD and CPBCM on flyrock throw predictions. The machine learning models were linear support vector machine (LSVM), quadratic Gaussian support vector machine (QGSVM), fine Gaussian support vector machine (FGSVM), medium Gaussian support vector machine (MGSVM), and cubic Gaussian support vector machine (CGSVM). The outcome indicates that FGSVM was the most sensitive with a 4% improvement when VOD and CPBCM were included. As a result, the LSVM model provides a suitable AI competitive alternative tool for flyrock throw prediction in mining operations by incorporating VOD and CPBCM.

Economic-threshold-based classification of soybean aphid, Aphis glycines, infestations in commercial soybean fields using Sentinel-2 satellite data

The soybean aphid (SBA), Aphis glycines Matsumura (Hemiptera: Aphididae), is a significant insect pest of soybean, Glycine max (L.) Merrill (Fabales: Fabaceae), and field treatment decisions for this pest are based on average field populations. Previous studies indicated that ground- and drone-based red-edge and near-infrared remote sensing can be used to detect plant stress caused by SBA infestations in soybean. However, it remains to be determined if remote sensing for SBA can be expanded to field or landscape scale using satellite-based platforms. Thus, this research was conducted in three steps to determine the potential of using Sentinel-2 satellite data for the classification of SBA infestations in soybean fields using simulated and actual Sentinel-2 satellite spectral reflectance. In the first step, as a proof of concept, hyperspectral data from cage studies were used to simulate Sentinel-2 bands and vegetation indices (VIs), conducted in nine trials at multiple locations between 2013 and 2021. The effects of SBA from caged plants on simulated data were evaluated with random intercept linear mixed models. The satellite simulation indicated a significant effect of SBA on the spectral reflectance of caged soybean plants (p < 0.05) for four satellite bands (5, 6, 7, and 8A) and five VIs (NDVI, GNDVI, SAVI, OSAVI, and NDRE). In the second step, actual Sentinel-2 spectral reflectance and corresponding aphid counts of commercial soybean fields, collected from 2017 to 2019, were obtained. The relationship between SBA counts and Sentinel-2 spectral reflectance from commercial soybean fields were evaluated with general linear models. A significant effect of SBA was observed for three satellite bands (6, 7, and 8A) and three VIs (NDVI, SAVI, and OSAVI). In the third step, linear support vector machine (LSVM) models for the classification of SBA infestations as above or below a previously determined economic threshold of 250 aphids per plant were developed using simulated Sentinel-2 bands and VIs from the caged plots, and were tested on actual Sentinel-2 data from commercial soybean fields. The best LSVM model for the classification of aphids in soybean reached 91% accuracy, 85.7% sensitivity, and 93.3% specificity. Thus, simulations with caged plots can be used as an indication of the potential of using satellite data for the detection of plant stresses on a larger scale. Furthermore, this study advances decision-making for SBA, and the developed LSVM model can be used to update regional and local monitoring for the management of SBA.

Machine learning approaches for cardiovascular hypertension stage estimation using photoplethysmography and clinical features.

Cardiovascular diseases (CVDs) are a leading cause of death worldwide, with hypertension emerging as a significant risk factor. Early detection and treatment of hypertension can significantly reduce the risk of developing CVDs and related complications. This work proposes a novel approach employing features extracted from the acceleration photoplethysmography (APG) waveform, alongside clinical parameters, to estimate different stages of hypertension. The current study used a publicly available dataset and a novel feature extraction algorithm to extract APG waveform features. Three distinct supervised machine learning algorithms were employed in the classification task, namely: Decision Tree (DT), Linear Discriminant Analysis (LDA), and Linear Support Vector Machine (LSVM). Results indicate that the DT model achieved exceptional training accuracy of 100% during cross-validation and maintained a high accuracy of 96.87% on the test dataset. The LDA model demonstrated competitive performance, yielding 85.02% accuracy during cross-validation and 84.37% on the test dataset. Meanwhile, the LSVM model exhibited robust accuracy, achieving 88.77% during cross-validation and 93.75% on the test dataset. These findings underscore the potential of APG analysis as a valuable tool for clinicians in estimating hypertension stages, supporting the need for early detection and intervention. This investigation not only advances hypertension risk assessment but also advocates for enhanced cardiovascular healthcare outcomes.

Gene Expression Machine Learning Models Classify Pediatric AML Subtypes with High Performance

Pediatric acute myeloid leukemia (pAML) encompasses over 20 molecular subtypes driven by unique genetic alterations, including hallmark chromosomal rearrangements and less frequently, point mutations or tandem duplications. Collectively, many of these pAML molecular categories are enriched in pediatric populations and are not represented by current classification systems, including the recently updated WHO or ICC. While fusion detection from RNA-Seq-based approaches is robust, many fusion negative subtypes would need to be defined by expression-based approaches as mutation calling from RNA-Seq data is less developed. Nevertheless, this could be challenging for subtypes with similar transcriptional profiles, such as those with shared HOX expression patterns, including NPM1, NUP98r, UBTF, DEK::NUP214, and KMT2A-PTD. To aid in the appropriate molecular classification of pAML, which is crucial for prognosis, we developed and compared three gene expression-based classifiers. A total of 1707 pAML gene expression profiles were mapped and analyzed from three distinct sources (St. Jude = 659; TARGET = 168; AAML1031 = 880). Raw read count data was normalized and scaled to obtain a relative expression value in transcripts per million (TPM), which served as input for feature selection, model training, and testing. Ground truth labels for all 1707 samples were obtained through multi-omics analysis, including whole genome sequencing, to identify fusions and mutations. For validation purposes, the data was stratified by subtypes and split 70/30 into training (n=1187) and testing (n=520). Three machine learning models were selected: random forest, XGboost, and linear support vector machine (SVM). Each sample had gene expression TPM data for 60,754 transcripts, out of which 20,004 transcripts related to protein-coding genes were incorporated for feature selection. Feature selection was performed using a median absolute deviation (MAD) algorithm to select the 3000 transcripts with the highest variability. Top variable 3000 genes were selected to allow for adequate tuning of the number of predictors. Each model was independently trained using stratified cross-validation and Monte-Carlo search for hyperparameter tuning. The best model was selected based on the Matthews Correlation Coefficient (MCC). Each model was tested on the hold-out TPM set with z-score normalization. On the hold-out testing set (n=520), the linear SVM model outperformed the random forest and XGboost models on five performance metrics across all subtypes (sensitivity=0.9577; precision=0.9577; specificity=0.9978; F1=0.96; accuracy=0.9958). The random forest (sensitivity=0.9154; precision=0.9154; specificity=0.9955; F1=0.92; accuracy=0.9915) and XGboost models (sensitivity=0.9231; precision=0.9231; specificity=0.9960; F1=0.92; accuracy=0.9923) also performed well across all subtypes. Although feature selection was shared across all three models, performance within each subtype varied between models. The linear SVM model demonstrated strong performance overall, driven by high specificity in classifying the KMT2Ar subgroup (n=127) and equal sensitivity across the GLIS-rearranged (n=14), GATA1 (n=10), BCL11B (n=7), CBFB::MYH11 (n=60), CEBPA (n=33), and RUNX1::RUNX1T1 (n=78) subtypes. The primary difference in performance between models is the high false positive rate for KMT2Ar and NPM1 (n=55) in the random forest and XGboost models. A preliminary hypothesis for this might be due to the large representation of KMT2Ar and NPM1 in the training data (24.85% and 10.78%, respectively). Synthetic upsampling (SMOTE) for the training dataset (n=1369) counteracts bias towards the majority classes and increases performance for the random forest (sensitivity=0.9327; precision=0.9327; specificity=0.9965; F1=0.93; accuracy=0.9933), XGboost (sensitivity=0.9404; precision=0.9404; specificity=0.9969; F1=0.94; accuracy=0.9940) and linear SVM (sensitivity=0.9615; precision=0.9615; specificity=0.9980; F1=0.96; accuracy=0.9962) models. Conjointly, these models demonstrate the utility and effectiveness of a machine learning approach for classifying pAML samples from transcriptome sequencing data, which may have broad clinical and research utility, especially for fusion negative subtypes.

Can an inertial measurement unit, combined with machine learning, accurately measure ground reaction forces in cricket fast bowling?

ABSTRACT This study examined whether an inertial measurement unit (IMU) could measure ground reaction force (GRF) during a cricket fast bowling delivery. Eighteen male fast bowlers had IMUs attached to their upper back and bowling wrist. Each participant bowled 36 deliveries, split into three different intensity zones: low = 70% of maximum perceived bowling effort, medium = 85%, and high = 100%. A force plate was embedded into the bowling crease to measure the ground truth GRF. Three machine learning models were used to estimate GRF from the IMU data. The best results from all models showed a mean absolute percentage error of 22.1% body weights (BW) for vertical and horizontal peak force, 24.1% for vertical impulse, 32.6% and 33.6% for vertical and horizontal loading rates, respectively. The linear support vector machine model had the most consistent results. Although results were similar to other papers that have estimated GRF, the error would likely prevent its use in individual monitoring. However, due to the large differences in raw GRFs between participants, researchers may be able to help identify links among GRF, injury, and performance by categorising values into levels (i.e., low and high).

Predicting the Risk of Developing Type 1 Diabetes Using a One-Week Continuous Glucose Monitoring Home Test With Classification Enhanced by Machine Learning: An Exploratory Study.

Detection of two or more autoantibodies (Ab) in the blood might describe those individuals at increased risk of developing type 1 diabetes (T1D) during the following years. The aim of this exploratory study is to propose a high versus low T1D risk classifier using machine learning technology based on continuous glucose monitoring (CGM) home data. Forty-two healthy relatives of people with T1D with mean ± SD age of 23.8 ± 10.5 years, HbA1c (glycated hemoglobin) of 5.3% ± 0.3%, and BMI (body mass index) of 23.2 ± 5.2 kg/m2 with zero (low risk; N = 21), and ≥2 (high risk; N = 21) Ab, were enrolled in an NIH (National Institutes of Health)-funded TrialNet ancillary study. Participants wore a CGM for a week and consumed three standardized liquid mixed meals (SLMM) instead of three breakfasts. Glycemic features were extracted from two-hour post-SLMM CGM traces, compared across groups, and used in four supervised machine learning Ab risk status classifiers. Recursive Feature Elimination (RFE) algorithm was used for feature selection; classifiers were evaluated through 10-fold cross-validation, using the receiver operating characteristic area under the curve (AUC-ROC) to select the best classification model. The percent time of glucose >180 mg/dL (T180), glucose range, and glucose CV (coefficient of variation) were the only significant differences between the glycemic features in the two groups with P values of .040, .035, and .028 respectively. The linear SVM (Support Vector Machine) model with RFE features achieved the best performance of classifying low-risk versus high-risk individuals with AUC-ROC = 0.88. A machine learning technology, combining a potentially self-administered one-week CGM home test, has the potential to reliably assess the T1D risk.

Recurrence prediction of lung adenocarcinoma using an immune gene expression and clinical data trained and validated support vector machine classifier.

Immune microenvironment plays a critical role in cancer from onset to relapse. Machine learning (ML) algorithm can facilitate the analysis of lab and clinical data to predict lung cancer recurrence. Prompt detection and intervention are crucial for long-term survival in lung cancer relapse. Our study aimed to evaluate the clinical and genomic prognosticators for lung cancer recurrence by comparing the predictive accuracy of four ML models. A total of 41 early-stage lung cancer patients who underwent surgery between June 2007 and October 2014 at New York University Langone Medical Center were included (with recurrence, n=16; without recurrence, n=25). All patients had tumor tissue and buffy coat collected at the time of resection. The CIBERSORT algorithm quantified tumor-infiltrating immune cells (TIICs). Protein-protein interaction (PPI) network and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis were conducted to unearth potential molecular drivers of tumor progression. The data was split into training (75%) and validation sets (25%). Ensemble linear kernel support vector machine (SVM) ML models were developed using optimized clinical and genomic features to predict tumor recurrence. Activated natural killer (NK) cells, M0 macrophages, and M1 macrophages showed a positive correlation with progression. Conversely, T CD4+ memory resting cells were negatively correlated. In the PPI network, TNF and IL6 emerged as prominent hub genes. Prediction models integrating clinicopathological prognostic factors, tumor gene expression (45 genes), and buffy coat gene expression (47 genes) yielded varying receiver operating characteristic (ROC)-area under the curves (AUCs): 62.7%, 65.4%, and 59.7% in the training set, 58.3%, 83.3%, and 75.0% in the validation set, respectively. Notably, merging gene expression with clinical data in a linear SVM model led to a significant accuracy boost, with an AUC of 92.0% in training and 91.7% in validation. Using ML algorithm, immune gene expression data from tumor tissue and buffy coat may enhance the precision of lung cancer recurrence prediction.

Investigation of Ensemble Learning Method in Predicting Early Xerostomia Using Radiomics Features from CT and MRI

There are different side effects in radiotherapy of head and neck cancer including xerostomia. Xerostomia is due to irradiation of salivary glands during treatment. In the present study, the main aim is evaluation and validation of individual and ensemble classifiers for prediction of early xerostomia in radiotherapy of head and neck cancer. Totally 62 patients diagnosed with head and neck cancer were evaluated prospectively. For this purpose, patients' demographic information, dosimetric data, and the extracted features from CT and T1 and T2 MR images before radiotherapy treatments were used as input for the predictive model. There were two main steps in this study: 1- Comparison of the features from T1 and T2 MR images for prediction of xerostomia, and 2- Evaluation of the performance of the radiomics models using individual classifiers and combination of the classifiers using ensemble learning. Feature extraction was performed using 3D Slicer software and for both parotids, 642 features were extracted from both CT, and MRI and Pearson statistical test was used for feature selection. For model development, four models were used consist of Random Tree (RT), Neural Network (NN), Linear Support Vector Machine (LSVM) and Bayesian Network (BS) classifiers. Different algorithms combined by using the ensemble learning method. For evaluation of the performance of the constructed models, sensitivity, specificity, area under curve (AUC), and receiver operating characteristic (ROC) curve were utilized. The results of the present study show that the extracted features from T1 weighted images have superior prediction ability compared to the T2 weighted ones. The RT and LSVM models which were based on T1 weighted images have better performance than those with T2 weighted images. The AUC in the test group for these models are 0.95 and 0.76, respectively, while for the T2 weighted images the corresponding values are 0.75 and 0.63. The AUC for the test group for the RT-BN, RT-LSVM-BN and RT-NN-LSVM-BN models amount to 0.96, 0.91, and 0.88, respectively. The results show that the radiomics features from images before radiotherapy can be used as personalized and unique biomarkers for prediction of xerostomia. Ensemble classifiers can be more efficient than individual classifiers in prediction of early xerostomia.

A-229 Immune Biomarkers Associated with COVID-19 Disease Severity in an Urban, Hospitalized Population

Abstract Background We sought to identify immune biomarkers associated with severe Coronavirus disease 2019 (COVID-19) in patients admitted to a large public county hospital during the early phase of the SARS-CoV-2 pandemic. We hypothesized that we could identify clinically relevant immune markers at the time of initial hospital admission that could be used to predict the course of COVID-19 illness. Methods The study population consisted of SARS-CoV-2 positive patients admitted for COVID-19 (n = 58) or controls (n = 14) at the Los Angeles County University of Southern California Medical Center between April-December 2020. Immunologic markers including chemokine/cytokines (IL-6, IL-8, IL-10, IP-10, MCP-1, TNFα) and serologic markers against SARS-CoV-2 antigens (including spike subunits S1 and S2, receptor binding domain (RBD), and nucleocapsid (N)) were assessed in serum collected on the day of admission using custom MILLIPLEX® immunoassay panels. Result values were computed using mean fluorescent intensity in individual samples fit to a standard curve using a 5PL logistic formula with power law variance. Comparison of patient demographic, clinical, cytokines and immunoglobulins characteristics between mild vs moderate/severe COVID-19 groups were conducted using Wilcoxon tests for continuous variables and Chi-square tests for categorical variables. Linear support vector machine models were fitted to perform the binary classification task of predicting mild vs moderate/severe COVID-19 using the python library scikit-learn. Results SARS-CoV-2 antibody levels were significantly elevated in patients with the highest COVID-19 disease severity, with IgM S1, IgG N, IgG RBD, IgG S1, and IgG S2 showing statistical significance between mild vs moderate/severe disease group medians (P = 0.037, 0.032, 0.007, 0.003, and 0.015, respectively). Of the chemokines/cytokines tested, only IP-10 showed significance across the disease groups (medians 640.8 pg/mL in mild, 493.3 pg/mL in moderate/severe, and 259.9 pg/mL in control, overall P = 0.005). The linear support vector machine model achieved an accuracy of 64% and an AUROC of 0.81 in predicting COVID-19 severity status. The most important clinical variables for predicting disease severity were white blood cell count, diastolic blood pressure, and platelet count, while the most important serologic markers were IgG anti-SARS-CoV-2 N, S1, S2, TNF-α, IP-10, and IL-10. Conclusion Our results suggest that IP-10 and anti-SARS-CoV-2 antibody measurements could be useful to identify patients most likely to experience the most severe forms of the disease. Strengths of this study include a focus on a racially and ethnically diverse patient population and a combined analysis of both cytokine/chemokine and immune response (antibody) biomarkers. However, we emphasize that our subjects were enrolled at a time before widespread vaccination against SARS-CoV-2.

RETRACTED ARTICLE: Research on Data Analysis of Efficient Innovation and Entrepreneurship Practice Teaching Based on LightGBM Classification Algorithm

The transformation of innovation and entrepreneurship practice teaching and education methods has put forward higher requirements for the intelligence and personalization of online education platforms. The aim of this study is to predict learning outcomes based on students' learning outcomes and habits, identify weak areas of knowledge, and provide targeted guidance and recommend the most suitable teaching resources. According to the concept of LightGBM model and the method of Feature selection, the research puts forward an integrated classification model ELO–LightGBM based on Elo Rating System (ELO) scoring system and Light Gradient Boosting Machine (LightGBM), trying to further mine the potential information of the practical teaching management data set. The model obtained a score of 0.7928 when using the dataset training, and a large number of comparative experiments were carried out between the ELO–LightGBM model and other classification models in different public datasets. The experimental results proved that the ELO–LightGBM model is more accurate than other classification models. In the comparative experiment on the practical teaching data set, the accuracy of the ELO–LightGBM model also surpassed the LightGBM model and the linear support vector machine model that performed well in small data sets, and the model was in the accuracy rate. The accuracy rate of winners in the comparison of micro-average is as high as 82.6%. It can be seen that the ELO–LightGBM model is of great significance to the intelligence and personalization of the online education platform.

Performance comparison of representative methods for few-shot speech gender analysis

As one of the basic tasks of the famous and important speech emotion recognition system, gender recognition based on speech analysis has attracted a lot of research interests in recent year. This paper considers the model performance of different machine learning methods on gender recognition in few-shot learning. In this article, we use three types of methods, namely Support Vector Machine (SVM), Convolutional Neural Networks (CNN), and Random forest. Among them, in the experiments of the SVM method, we report the linear SVM and nonlinear SVM models separately. Different experiments were performed with respect to the number of features and number of iterations and the results and possible issues were evaluated.

Modeling landslide susceptibility based on convolutional neural network coupling with metaheuristic optimization algorithms

ABSTRACT Landslides are one of the most common geological hazards worldwide, especially in Sichuan Province (Southwest China). The current study's main purposes are to explore the potential applications of convolutional neural networks (CNN) hybrid ensemble metaheuristic optimization algorithms, namely beluga whale optimization (BWO) and coati optimization algorithm (COA), for landslide susceptibility mapping in Sichuan Province (China). For this aim, fourteen landslide conditioning factors were compiled in a spatial database. The effectiveness of the conditioning factors in the development of the landslide predictive model was quantified using the linear support vector machine model. The receiver operating characteristic (ROC) curve (AUC), the root mean square error, and six statistical indices were used to test and compare the three resultant models. For the training dataset, the AUC values of the CNN-COA, CNN-BWO and CNN models were 0.946, 0.937 and 0.855, respectively. In terms of the validation dataset, the CNN-COA model exhibited a higher AUC value of 0.919, while the AUC values of the CNN-BWO and CNN models were 0.906 and 0.805, respectively. The results indicate that the CNN-COA model, followed by the CNN-BWO model, and the CNN model, offers the best overall performance for landslide susceptibility analysis.

Machine Learning Prediction of Iron Deficiency Anemia in Chinese Premenopausal Women 12 Months after Sleeve Gastrectomy.

Premenopausal women, who account for more than half of patients for bariatric surgery, are at higher risk of developing postoperative iron deficiency anemia (IDA) than postmenopausal women and men. We aimed at establishing a machine learning model to evaluate the risk of newly onset IDA in premenopausal women 12 months after sleeve gastrectomy (SG). Premenopausal women with complete clinical records and undergoing SG were enrolled in this retrospective study. Newly onset IDA after surgery, the main outcome, was defined according to the age- and gender-specific World Health Organization criteria. A linear support vector machine model was developed to predict the risk of IDA after SG with the top five important features identified during feature selection. Four hundred and seven subjects aged 31.0 (Interquartile range (IQR): 26.0-36.0) years with a median follow-up period of 12 (IQR 7-13) months were analyzed. They were divided into a training set and a validation set with 285 and 122 individuals, respectively. Preoperative ferritin, age, hemoglobin, creatinine, and fasting C-peptide were included. The model showed moderate discrimination in both sets (area under curve 0.858 and 0.799, respectively, p < 0.001). The calibration curve indicated acceptable consistency between observed and predicted results in both sets. Moreover, decision curve analysis showed substantial clinical benefits of the model in both sets. Our machine learning model could accurately predict newly onset IDA in Chinese premenopausal women with obesity 12 months after SG. External validation was required before the model was used in clinical practice.

Immune biomarkers associated with COVID-19 disease severity in an urban, hospitalized population

ObjectivesWe sought to identify immune biomarkers associated with severe Coronavirus disease 2019 (COVID-19) in patients admitted to a large urban hospital during the early phase of the SARS-CoV-2 pandemic. DesignThe study population consisted of SARS-CoV-2 positive subjects admitted for COVID-19 (n = 58) or controls (n = 14) at the Los Angeles County University of Southern California Medical Center between April 2020 through December 2020. Immunologic markers including chemokine/cytokines (IL-6, IL-8, IL-10, IP-10, MCP-1, TNF-α) and serologic markers against SARS-CoV-2 antigens (including spike subunits S1 and S2, receptor binding domain, and nucleocapsid) were assessed in serum collected on the day of admission using bead-based multiplex immunoassay panels. ResultsWe observed that body mass index (BMI) and SARS-CoV-2 antibodies were significantly elevated in patients with the highest COVID-19 disease severity. IP-10 was significantly elevated in COVID-19 patients and was associated with increased SARS-CoV-2 antibodies. Interactions among all available variables on COVID-19 disease severity were explored using a linear support vector machine model which supported the importance of BMI and SARS-CoV-2 antibodies. ConclusionsOur results confirm the known adverse association of BMI on COVID-19 severity and suggest that IP-10 and SARS-CoV-2 antibodies could be useful to identify patients most likely to experience the most severe forms of the disease.

Untargeted Metabolomics Approach for the Differentiation between Panax vietnamensis var. vietnamensis and Panax vietnamensis var. fuscidiscus.

Panax vietnamensis var. vietnamensis (PVV) and Panax vietnamensis var. fuscidiscus (PVF) both belong to Panax vietnamensis species and are chemically and morphologically similar, making it hard to distinguish for the consumer. Herein, 42 PVF and 12 PVV samples were collected in Quang Nam and Lai Chau Province, respectively, and subsequently characterized by ITSr-DNA sequence data to verify their origins. Next, untargeted metabolomics combined with multivariate statistical analysis was developed to differentiate PVV and PVF. The metabolic profiles of PVV and PVF were found to be distinct and classified well using Partial Least-Squares Discriminant Analysis (PLS-DA) in the training set. Among them, seven ginsenosides were of high abundance in PVV, while six were of high abundance in PVF. Next, the test set was used to validate 13 putative differential markers found in the training set, illustrating a complete match with the expression patterns of these ginsenosides in the training set. Finally, PLS-DA and linear Support Vector Machine models both indicated distinct ginsenoside profiles of PVV and PVF without misclassification in the test set. Conclusively, the developed untargeted metabolomics approach might serve as a powerful tool for the authentication of PVV and PVF at the metabolome level.

Predicting Immunological Risk for Stage 1 and Stage 2 Diabetes Using a 1-Week CGM Home Test, Nocturnal Glucose Increments, and Standardized Liquid Mixed Meal Breakfasts, with Classification Enhanced by Machine Learning.

Background: Predicting the risk for type 1 diabetes (T1D) is a significant challenge. We use a 1-week continuous glucose monitoring (CGM) home test to characterize differences in glycemia in at-risk healthy individuals based on autoantibody presence and develop a machine-learning technology for CGM-based islet autoantibody classification. Methods: Sixty healthy relatives of people with T1D with mean ± standard deviation age of 23.7 ± 10.7 years, HbA1c of 5.3% ± 0.3%, and body mass index of 23.8 ± 5.6 kg/m2 with zero (n = 21), one (n = 18), and ≥2 (n = 21) autoantibodies were enrolled in an National Institutes of Health TrialNet ancillary study. Participants wore a CGM for a week and consumed three standardized liquid mixed meals (SLMM) instead of three breakfasts. Glycemic outcomes were computed from weekly, overnight (12:00-06:00), and post-SLMM CGM traces, compared across groups, and used in four supervised machine-learning autoantibody status classifiers. Classifiers were evaluated through 10-fold cross-validation using the receiver operating characteristic area under the curve (AUC-ROC) to select the best classification model. Results: Among all computed glycemia metrics, only three were different across the autoantibodies groups: percent time >180 mg/dL (T180) weekly (P = 0.04), overnight CGM incremental AUC (P = 0.005), and T180 for 75 min post-SLMM CGM traces (P = 0.004). Once overnight and post-SLMM features are incorporated in machine-learning classifiers, a linear support vector machine model achieved the best performance of classifying autoantibody positive versus autoantibody negative participants with AUC-ROC ≥0.81. Conclusion: A new technology combining machine learning with a potentially self-administered 1-week CGM home test can help improve T1D risk detection without the need to visit a hospital or use a medical laboratory. Trial registration: ClinicalTrials.gov registration no. NCT02663661.

Shared and distinct patterns of dynamic functional connectivity variability of thalamo-cortical circuit in bipolar depression and major depressive disorder.

Evidence has indicated abnormalities of thalamo-cortical functional connectivity (FC) in bipolar disorder during a depressive episode (BDD) and major depressive disorder (MDD). However, the dynamic FC (dFC) within this system is poorly understood. We explored the thalamo-cortical dFC pattern by dividing thalamus into 16 subregions and combining with a sliding-window approach. Correlation analysis was performed between altered dFC variability and clinical data. Classification analysis with a linear support vector machine model was conducted. Compared with healthy controls (HCs), both patients revealed increased dFC variability between thalamus subregions with hippocampus (HIP), angular gyrus and caudate, and only BDD showed increased dFC variability of the thalamus with superior frontal gyrus (SFG), HIP, insula, middle cingulate gyrus, and postcentral gyrus. Compared with MDD and HCs, only BDD exhibited enhanced dFC variability of the thalamus with SFG and superior temporal gyrus. Furthermore, the number of depressive episodes in MDD was significantly positively associated with altered dFC variability. Finally, the disrupted dFC variability could distinguish BDD from MDD with 83.44% classification accuracy. BDD and MDD shared common disrupted dFC variability in the thalamo-limbic and striatal-thalamic circuitries, whereas BDD exhibited more extensive and broader aberrant dFC variability, which may facilitate distinguish between these 2 mood disorders.

Klasifikcija i analiza skupa podataka MNIST pomoću algoritama PCA i SVM

Introduction/purpose: The utilization of machine learning methods has become indispensable in analyzing large-scale, complex data in contemporary data-driven environments, with a diverse range of applications from optimizing business operations to advancing scientific research. Despite the potential for insight and innovation presented by these voluminous datasets, they pose significant challenges in areas such as data quality and structure, necessitating the implementation of effective management strategies. Machine learning techniques have emerged as essential tools in identifying and mitigating these challenges and developing viable solutions to address them. The MNIST dataset represents a prominent example of a widely-used dataset in this field, renowned for its expansive collection of handwritten numerical digits, and frequently employed in tasks such as classification and analysis, as demonstrated in the present study. Methods: This study employed the MNIST dataset to investigate various statistical techniques, including the Principal Components Analysis (PCA) algorithm implemented using the Python programming language. Additionally, Support Vector Machine (SVM) models were applied to both linear and non-linear classification problems to assess the accuracy of the model. Results: The results of the present study indicate that while the PCA technique is effective for dimensionality reduction, it may not be as effective for visualization purposes. Moreover, the findings demonstrate that both linear and non-linear SVM models were capable of effectively classifying the dataset. Conclusion: The findings of the study demonstrate that SVM can serve as an efficacious technique for addressing classification problems.

Integration of Clinical and CT-Based Radiomic Features for Pretreatment Prediction of Pathologic Complete Response to Neoadjuvant Systemic Therapy in Breast Cancer.

The purpose of the present study was to examine the potential of a machine learning model with integrated clinical and CT-based radiomics features in predicting pathologic complete response (pCR) to neoadjuvant systemic therapy (NST) in breast cancer. Contrast-enhanced CT was performed in 329 patients with breast tumors (n = 331) before NST. Pyradiomics was used for feature extraction, and 107 features of seven classes were extracted. Feature selection was performed on the basis of the intraclass correlation coefficient (ICC), and six ICC thresholds (0.7-0.95) were examined to identify the feature set resulting in optimal model performance. Clinical factors, such as age, clinical stage, cancer cell type, and cell surface receptors, were used for prediction. We tried six machine learning algorithms, and clinical, radiomics, and clinical-radiomics models were trained for each algorithm. Radiomics and clinical-radiomics models with gray level co-occurrence matrix (GLCM) features only were also built for comparison. The linear support vector machine (SVM) regression model trained with radiomics features of ICC ≥0.85 in combination with clinical factors performed the best (AUC = 0.87). The performance of the clinical and radiomics linear SVM models showed statistically significant difference after correction for multiple comparisons (AUC = 0.69 vs. 0.78; p &lt; 0.001). The AUC of the radiomics model trained with GLCM features was significantly lower than that of the radiomics model trained with all seven classes of radiomics features (AUC = 0.85 vs. 0.87; p = 0.011). Integration of clinical and CT-based radiomics features was helpful in the pretreatment prediction of pCR to NST in breast cancer.

Computer vision model for estimating the mass and volume of freshly harvested Thai apple ber (Ziziphus mauritiana L.) and its variation with storage days

The physical properties of fruits are proportional to their mass and volume; this connection is used to determine the fruit qualities and in designing the novel postharvest machinery. The present study aimed to forecast the mass and volume of Thai apple ber (Ziziphus mauritiana L.) as a function of its physical properties measured using image processing techniques at different stages of ripening (1st day, 4th day, 7th day, and 10th day). The mass and volume models developed and analyzed the single variable regression, multilinear regressions, and mass regression based on volume. Among these models, linear support vector machine (SVM) was found appropriate. The experimental data analysis showed that the R2 of the linear SVM model for mass and volume of the projected area were 0.955 and 0.965, respectively. In contrast, for the multilinear regression model, R2 values were 0.967 and 0.972, respectively. For the mass prediction model, the R2 was 0.970 based on calculated volume showing a linear relationship. Thus, it was concluded that real-time measurement of physical properties of Thai apple ber using an image-processing technique to estimate the mass and volume is a precise and accurate approach.