Radial Basis Function Support Vector Machine Research Articles

Applying a nomogram based on preoperative CT to predict early recurrence of laryngeal squamous cell carcinoma after surgery.

To identify the value of a computed tomography (CT)-based radiomics model to predict probability of early recurrence (ER) in patients diagnosed with laryngeal squamous cell carcinoma (LSCC) after surgery. Pre-operative CT scans of 140 LSCC patients treated by surgery are reviewed and selected. These patients are randomly split into the training set (n = 97) and test set (n = 43). The regions of interest of each patient were delineated manually by two senior radiologists. Radiomics features are extracted from CT images acquired in non-enhanced, arterial, and venous phases. Variance threshold, one-way ANOVA, and least absolute shrinkage and selection operator algorithm are used for feature selection. Then, radiomics models are built with five algorithms namely, k-nearest neighbor (KNN), logistic regression (LR), linear support vector machine (LSVM), radial basis function SVM (RSVM), and polynomial SVM (PSVM). Clinical factors are selected using univariate and multivariate logistic regressions. Last, a radiomics nomogram incorporating the radiomics signature and clinical factors is built to predict ER and its efficiency is evaluated by receiver operating characteristic (ROC) curve and calibration curve. Decision curve analysis (DCA) is also used to evaluate clinical usefulness. Four features are remarkably associated with ER in patients with LSCC. Applying to test set, the area under the ROC curves (AUCs) of KNN, LR, LSVM, RSVM, and PSVM are 0.936, 0.855, 0.845, 0.829, and 0.794, respectively. The radiomics nomogram shows better discrimination (with AUC: 0.939, 95% CI: 0.867-0.989) than the best radiomics model and the clinical model. Predicted and actual ERs in the calibration curves are in good agreement. DCA shows that the radiomics nomogram is clinically useful. The radiomics nomogram, as a noninvasive prediction tool, exhibits favorable performance for ER prediction of LSCC patients after surgery.

IMPROVING DIGITAL SIGNATURE VERIFICATION ACCURACY THROUGH SUPPORT VECTOR MACHINE LEARNING: A COMPARATIVE STUDY

Digital signatures are widely used in electronic documents, and their verification is crucial to ensure document authenticity and security. However, digital signature verification can be challenging, especially when dealing with large amounts of data. In this paper, we present a comparative study of three Support Vector Machine (SVM) based methods for improving digital signature verification accuracy. We used a dataset of 10,000 digital signatures and compared the performance of linear SVM, polynomial SVM, and radial basis function (RBF) SVM. Our results showed that all three SVM-based methods improved the accuracy of digital signature verification compared to traditional methods. The RBF SVM method was found to be the most effective method for improving accuracy, with an accuracy of 98%.

Label-free surface-enhanced Raman spectroscopy of serum with machine-learning algorithms for gallbladder cancer diagnosis

Gallbladder cancer (GBC) is a rare but frequently fatal biliary tract malignancy that is typically discovered when it is already advanced. In this study, we investigated a novel technique for the quick and non-invasive diagnosis of GBC based on serum surface-enhanced Raman spectroscopy (SERS). SERS spectra of serum from 41 patients with GBC and 72 normal subjects were recorded. Principal component analysis-linear discriminant analysis (PCA-LDA), and PCA-support vector machine (PCA-SVM), Linear SVM and Gaussian radial basis function-SVM (RBF-SVM) algorithms were used to establish the classification models, respectively. When the Linear SVM was used, the overall diagnostic accuracy for classifying the two groups could achieve 97.1%, and when RBF-SVM was used, the diagnostic sensitivity of GBC was 100%. The results demonstrated that SERS combination with a machine learning algorithm is a promising candidate to be one of the diagnostic tools for GBC in the future.

Application of machine learning in predicting non-alcoholic fatty liver disease using anthropometric and body composition indices

Non-alcoholic fatty liver disease (NAFLD) is the most common chronic liver disease, which can progress from simple steatosis to advanced cirrhosis and hepatocellular carcinoma. Clinical diagnosis of NAFLD is crucial in the early stages of the disease. The main aim of this study was to apply machine learning (ML) methods to identify significant classifiers of NAFLD using body composition and anthropometric variables. A cross-sectional study was carried out among 513 individuals aged 13 years old or above in Iran. Anthropometric and body composition measurements were performed manually using body composition analyzer InBody 270. Hepatic steatosis and fibrosis were determined using a Fibroscan. ML methods including k-Nearest Neighbor (kNN), Support Vector Machine (SVM), Radial Basis Function (RBF) SVM, Gaussian Process (GP), Random Forest (RF), Neural Network (NN), Adaboost and Naïve Bayes were examined for model performance and to identify anthropometric and body composition predictors of fatty liver disease. RF generated the most accurate model for fatty liver (presence of any stage), steatosis stages and fibrosis stages with 82%, 52% and 57% accuracy, respectively. Abdomen circumference, waist circumference, chest circumference, trunk fat and body mass index were among the most important variables contributing to fatty liver disease. ML-based prediction of NAFLD using anthropometric and body composition data can assist clinicians in decision making. ML-based systems provide opportunities for NAFLD screening and early diagnosis, especially in population-level and remote areas.

Differentiation of NaCl, NaOH, and β-Phenylethylamine Using Ultraviolet Spectroscopy and Improved Adaptive Artificial Bee Colony Combined with BP-ANN Algorithm.

The aim of this study is to enhance the classification performance of the back-propagation-artificial neural network (BP-ANN) algorithm for NaCl, NaOH, β-phenylethylamine (PEA), and their mixture, as well as to avoid the defects of the artificial bee colony (ABC) algorithm such as prematurity and local optimization. In this paper, a method that combined an improved adaptive artificial bee colony (IAABC) algorithm and BP-ANN algorithm was proposed. This method improved the ABC algorithm by adding an adaptive local search factor and mutation factor; meanwhile, it can enhance the abilities of the global optimization and local search of the ABC algorithm and avoid prematurity. The extracted score vectors of the principal component of the ultraviolet (UV) spectrum were used as the input variable of the BP-ANN algorithm. The IAABC algorithm was used to optimize the weight and threshold of the BP-ANN algorithm, and the iterative algorithm was repeated until the output accuracy was reached. The output variable was the classification results of NaCl, NaOH, PEA, and the mixture. Meanwhile, the proposed IAABC-BP-ANN algorithm was compared with discriminant analysis (DA), sigmaid-support vector machine (SVM), radial basis function-SVM (RBF-SVM), BP-ANN, and ABC-BP-ANN. Then, the above algorithms were used to classify NaCl, NaOH, PEA, and the mixture, respectively. In the experiment, four indicators, accuracy, recall, precision, and F-score, were used as the evaluation criteria. In addition, the regression equation parameters of the mixture for the testing set were obtained by BP-ANN, ABC-BP-ANN, and IAABC-BP-ANN models. All of the results showed that IAABC-BP-ANN exhibits better performance than other algorithms. Therefore, IAABC-BP-ANN combined with UV spectroscopy is a potential identification tool for the detection of NaCl, NaOH, PEA, and the mixture.

Improved Parkinsonian tremor quantification based on automatic label modification and SVM with RBF kernel

Objective. The quantitative assessment of Parkinsonian tremor, e.g. (0, 1, 2, 3, 4) according to the Movement Disorder Society-Unified Parkinson’s Disease Rating Scale, is crucial for treating Parkinson’s disease. However, the tremor amplitude constantly fluctuates due to environmental and psychological effects on the patient. In clinical practice, clinicians assess the tremor severity for a short duration, whereas manual tremor labeling relies on the clinician’s physician experience. Therefore, automatic tremor quantification based on wearable inertial sensors and machine learning algorithms is affected by the manual labels of clinicians. In this study, an automatic modification method for the labels judged by clinicians is presented to improve Parkinsonian tremor quantitation. Approach. For the severe overlapping of dynamic feature range between different severities, an outlier modification algorithm (PCA-IQR) based on the combination of principal component analysis and interquartile range statistic rule is proposed to learn the blurred borders between different severity scores, thereby optimizing the labels. Afterward, according to the modified feature vectors, a support vector machine (SVM) with a radial basis function (RBF) kernel is proposed to classify the tremor severity. The classifier models of SVM with RBF kernel, k-nearest neighbors, and SVM with the linear kernel are compared. Main results. Experimental results show that the proposed method has high classification performance and excellent model generalization ability for tremor quantitation (accuracy: 97.93%, precision: 97.96%, sensitivity: 97.93%, F1-score: 97.94%). Significance. The proposed method may not only provide valuable assistance for clinicians to assess the tremor severity accurately, but also provides self-monitoring for patients at home and improve the assessment skills of clinicians.

P264 Machine learning can accurately predict development of Inflammatory Bowel Disease

Abstract Background Prediction of inflammatory bowel disease (IBD) development is the subject of intense study. We have previously shown that significant changes occur in routine laboratory tests in the 5 years preceding IBD diagnosis. Using routinely collected laboratory results, we developed machine learning models to predict development of IBD. Methods We extracted data from the electronic medical records of Maccabi Health Services and included patients with IBD ≥ 16 years of age with a minimum of 5 years follow up and an age and sex matched healthy cohort. The outcome measured was entry into the IBD registry. The study population was split into training and validation cohorts and using laboratory values in the 5 years prior to IBD diagnosis as features, we assessed the performance of supervised learning models and performed interpretability analysis with unsupervised dimensionality reduction. Results 5643 patients with IBD and 17199 healthy people were included in this study. 3039 (53.8%) had Crohn’s disease (CD), 2322 (41.1%) ulcerative colitis (UC) and 282 (5%) indeterminate colitis. The mean age of the IBD and healthy cohorts at study inclusion was 39.3 ± 16.5 and 38.5 ± 15.6 years, respectively. Weighted nearest neighbour imputation was used to correct for data missingness. Unsupervised Uniform Manifold Approximation Projection dimensionality reduction correctly separated the cohort into healthy and IBD clusters (Figure 1A). Furthermore, this algorithm identified two distinct groups correlating to CD or UC diagnosis (Figure 1B). Supervised learning models, namely neural net (AUC 0.95), linear support vector machine (SVM) (AUC 0.93), radial basis function SVM (AUC 0.85) and logistic regression (AUC 0.93) models, accurately predicted IBD diagnosis (Figure 2). These models were repeated without C-reactive protein and the predictive value was not significantly changed. Moreover, we excluded data in the 2 years preceding IBD diagnosis to determine whether model performance was reduced and the predictive value remained the same. Interestingly, using only data from 3-5 years preceding IBD diagnosis, the model could better distinguish between CD and UC (Figure 3). Conclusion In this study, machine learning models could accurately predict a diagnosis of IBD using laboratory data from the 5 years preceding IBD diagnosis. Prospective studies are needed to test these models in a real world setting.

Development and Validation of Multivariable Prediction Algorithms to Estimate Future Walking Behavior in Adults: Retrospective Cohort Study.

Physical inactivity is associated with numerous health risks, including cancer, cardiovascular disease, type 2 diabetes, increased health care expenditure, and preventable, premature deaths. The majority of Americans fall short of clinical guideline goals (ie, 8000-10,000 steps per day). Behavior prediction algorithms could enable efficacious interventions to promote physical activity by facilitating delivery of nudges at appropriate times. The aim of this paper is to develop and validate algorithms that predict walking (ie, >5 min) within the next 3 hours, predicted from the participants' previous 5 weeks' steps-per-minute data. We conducted a retrospective, closed cohort, secondary analysis of a 6-week microrandomized trial of the HeartSteps mobile health physical-activity intervention conducted in 2015. The prediction performance of 6 algorithms was evaluated, as follows: logistic regression, radial-basis function support vector machine, eXtreme Gradient Boosting (XGBoost), multilayered perceptron (MLP), decision tree, and random forest. For the MLP, 90 random layer architectures were tested for optimization. Prior 5-week hourly walking data, including missingness, were used for predictors. Whether the participant walked during the next 3 hours was used as the outcome. K-fold cross-validation (K=10) was used for the internal validation. The primary outcome measures are classification accuracy, the Mathew correlation coefficient, sensitivity, and specificity. The total sample size included 6 weeks of data among 44 participants. Of the 44 participants, 31 (71%) were female, 26 (59%) were White, 36 (82%) had a college degree or more, and 15 (34%) were married. The mean age was 35.9 (SD 14.7) years. Participants (n=3, 7%) who did not have enough data (number of days <10) were excluded, resulting in 41 (93%) participants. MLP with optimized layer architecture showed the best performance in accuracy (82.0%, SD 1.1), whereas XGBoost (76.3%, SD 1.5), random forest (69.5%, SD 1.0), support vector machine (69.3%, SD 1.0), and decision tree (63.6%, SD 1.5) algorithms showed lower performance than logistic regression (77.2%, SD 1.2). MLP also showed superior overall performance to all other tried algorithms in Mathew correlation coefficient (0.643, SD 0.021), sensitivity (86.1%, SD 3.0), and specificity (77.8%, SD 3.3). Walking behavior prediction models were developed and validated. MLP showed the highest overall performance of all attempted algorithms. A random search for optimal layer structure is a promising approach for prediction engine development. Future studies can test the real-world application of this algorithm in a "smart" intervention for promoting physical activity.

The Quality Detection and Recognition for Food Seasoning Based on an Artificial Olfactory System

The quality of food seasoning affects the flavor and the safety of the food. Traditional detection methods, such as test-paper, cyclotron or chromatography, have some limits in accuracy, detection range, time consumption, portability and cost. The artificial olfactory system (AOS) can detect and identify the food quality by simulating animal olfactory sensors, which has the advantages of short detection cycle, high sensitivity, and no complicated pre-processing. A sample data source for AOS to detect pepper powder was built by collecting the odor data of the condiment powders under different origins and adulteration conditions using the sensor array of the PEN3 electronic nose system (AIRSENSE, Germany). Then, machine learning algorithms including support vector machine, decision tree and random forest were introduced to form the intelligent models to evaluate the quality of the pepper powder from different origins and with different adulterated substances. The results indicated that the recognition accuracy reached to over 98% for all types of pepper samples with the radial basis function-support vector machine (RBF-SVM) model. The pure pepper could be distinguished from the adulterated samples successfully with the accuracy of near 100% using machine leaning models. Finally, other more machine learning algorithms were compared to recognize the states of different pepper powders. The results showed that the SVM method has the highest accuracy in classifying the adulteration of pepper powder among different machine learning models discussed in the research. It is feasible to detect and recognize the quality of the food seasoning with AOS based on machine learning algorithms and sensor array.

Machine Learning models for the Cognitive Stress Detection Using Heart Rate Variability Signals

Cognitive domains play a critical role in daily functioning. The prediction of cognitive stress state is important to better monitor work performance. This study aims to explore machine learning models to detect cognitive load or state using heart rate variability (HRV) signals. HRV data were recorded from thirty subjects during rest, two cognitive tasks (d2 Attention and Featuring Switcher task), and recovery. Seven HRV indexes from both time and frequency domains, extracted from raw R-R intervals, were used to identify whether subjects performed cognitive tasks or not. Five classifier models: linear support vector machine (LSVM), kernel SVM radial basis function, k-nearest neighbor (KNN), and random forest (RF), were trained and evaluated using a leave-one-out cross-validation approach. The accuracies and F1-score range from 0.54 to 0.62, with LSVM, showing the best. These acceptable performances indicate the machine learning approach could be used to further distinguish between rest and cognitive state. With the ubiquity of non-invasive and low-cost wearable devices, this finding offers insight to be incorporated into personal work performance monitoring in the digital age.

High‐resolution reference evapotranspiration for arid Egypt: Comparative analysis and evaluation of empirical and artificial intelligence models

AbstractAccurate estimation of evapotranspiration has crucial importance in arid regions like Egypt, which suffers from the scarcity of precipitation and water shortages. This study provides an investigation of the performance of 31 widely used empirical equations and 20 models developed using five artificial intelligence (AI) algorithms to estimate reference evapotranspiration (ET0) to generate gridded high‐resolution daily ET0 estimates over Egypt. The AI algorithms include support vector machine‐radial basis function (SVM‐RBF), random forest (RF), group method of data handling neural network (GMDH‐NN), multivariate adaptive regression splines (MARS), and dynamic evolving neural fuzzy interference system (DENFIS). Daily observations records of 41 stations distributed over Egypt were used to calculate ET0 using FAO56 Penman–Monteith equation as a reference estimate. The multiparameter Kling‐Gupta efficiency (KGE) metric was used as an evaluation metric for its robustness in representing different statistical error/agreement characteristics in a single value. By category, the empirical equations based on radiation performed better in replicating FAO56‐PM followed by temperature‐ and mass‐transfer‐based ones. Ritchie equation was found to be the best overall in Egypt (median KGE 0.76) followed by Caprio (median KGE 0.64), and Penman (median KGE 0.52) equations based on station‐wise ranking. On the other hand, the RF model, having maximum and minimum temperatures, wind speed, and relative humidity as predictors, outperformed other AI algorithms. Overall, the RF model performed the best among all the AI models and empirical equations. The generated 0.10° × 0.10° daily estimates of ET0 enabled the detection of a significant increase of 0.12–0.16 mm·decade−1 in the agricultural‐dependent Nile Delta using the modified Mann–Kendall test and Sen's slope estimator.

A Framework on Performance Analysis of Mathematical Model-Based Classifiers in Detection of Epileptic Seizure from EEG Signals with Efficient Feature Selection.

Epilepsy is one of the neurological conditions that are diagnosed in the vast majority of patients. Electroencephalography (EEG) readings are the primary tool that is used in the process of diagnosing and analyzing epilepsy. The epileptic EEG data display the electrical activity of the neurons and provide a significant amount of knowledge on pathology and physiology. As a result of the significant amount of time that this method requires, several automated classification methods have been developed. In this paper, three wavelets such as Haar, dB4, and Sym 8 are employed to extract the features from A–E sets of the Bonn epilepsy dataset. To select the best features of epileptic seizures, a Particle Swarm Optimization (PSO) technique is applied. The extracted features are further classified using seven classifiers like linear regression, nonlinear regression, Gaussian Mixture Modeling (GMM), K-Nearest Neighbor (KNN), Support Vector Machine (SVM-linear), SVM (polynomial), and SVM Radial Basis Function (RBF). Classifier performances are analyzed through the benchmark parameters, such as sensitivity, specificity, accuracy, F1 Score, error rate, and g-means. The SVM classifier with RBF kernel in sym 8 wavelet features with PSO feature selection method attains a higher accuracy rate of 98% with an error rate of 2%. This classifier outperforms all other classifiers.

Performance improvement of machine learning models via wavelet theory in estimating monthly river streamflow

River streamflow is an essential hydrological parameters for optimal water resource management. This study investigates models used to estimate monthly time-series river streamflow data at two hydrological stations in the USA (Heise and Irwin on Snake River, Idaho). Five diverse types of machine learning (ML) model were tested, support vector machine-radial basis function (SVM-RBF), SVM-Polynomial (SVM-Poly), decision tree (DT), gradient boosting (GB), random forest (RF), and long short-term memory (LSTM). These were trained and tested alongside a conventional multiple linear regression (MLR). To improve the estimation and model performance, hybrid models were designed by coupling the models with wavelet theory (W). The models performance was assessed using root mean square error (RMSE), mean absolute error (MAE), coefficient of determination (R2), Nash-Sutcliffe efficiency (NSE), and Willmott’s index (WI). A side-by-side performance assessment of the stand-alone and hybrid models revealed that the coupled models exhibit better estimates of monthly river streamflow relative to the stand-alone ones. The statistical parameter values for the best model (W-LSTM4) during the test phase was RMSE = 36.533 m3/s, MAE = 26.912 m3/s, R2 = 0.947, NSE = 0.946, WI = 0.986 (Heise station), and RMSE = 33.378 m3/s, MAE = 24.562 m3/s, R2 = 0.952, NSE = 0.951, WI = 0.987 (Irwin station).

Applying machine learning methods to predict geology using soil sample geochemistry

In this study we compared various machine learning techniques that used soil geochemistry to aid in geologic mapping. We tested six different sampling methods (undersample, oversample, Synthetic Minority Oversampling Technique (SMOTE), Adaptive Synthetic Sampling (ADASYN), SMOTE and Edited Nearest Neighbor (SMOTEENN), and SMOTE and Tomek links (SMOTETomek)). SMOTE performed best with ADASYN and SMOTETomek having slightly lower effectiveness. Nine machine learning algorithms (naïve Bayes, logistic regression, quadratic discriminant analysis, nearest neighbors, radial basis function support-vector machine, artificial neural network, random forest, AdaBoost classifier, and gradient boosting classifier) were compared and AdaBoost classifiers and gradient boosting classifiers were found to be most effective. Finally, we experimented with multiple classifier systems (MCS) testing different combinations of algorithms and various combinatorial functions. It was found that MCS can outperform individual models, and the best MCS combined nearest neighbors, radial basis function support-vector machine, artificial neural network, random forest, AdaBoost classifiers, and gradient boosting classifier, then applied a logistic regression to the probabilities output by the models. Ultimately, we created a tool that is able to adequately predict underlying geology in the study area using soil geochemistry.

Integrated analysis of plasma and urine reveals unique metabolomic profiles in idiopathic inflammatory myopathies subtypes.

ObjectivesIdiopathic inflammatory myopathies (IIM) are a class of autoimmune diseases with high heterogeneity that can be divided into different subtypes based on clinical manifestations and myositis‐specific autoantibodies (MSAs). However, even in each IIM subtype, the clinical symptoms and prognoses of patients are very different. Thus, the identification of more potential biomarkers associated with IIM classification, clinical symptoms, and prognosis is urgently needed.MethodsPlasma and urine samples from 79 newly diagnosed IIM patients (mean disease duration 4 months) and 52 normal control (NC) samples were analysed by high‐performance liquid chromatography of quadrupole time‐of‐flight mass spectrometry (HPLC‐Q‐TOF‐MS)/MS‐based untargeted metabolomics. Orthogonal partial least‐squares discriminate analysis (OPLS‐DA) were performed to measure the significance of metabolites. Pathway enrichment analysis was conducted based on the KEGG human metabolic pathways. Ten machine learning (ML) algorithms [linear support vector machine (SVM), radial basis function SVM, random forest, nearest neighbour, Gaussian processes, decision trees, neural networks, adaptive boosting (AdaBoost), Gaussian naive Bayes and quadratic discriminant analysis] were used to classify each IIM subtype and select the most important metabolites as potential biomarkers.ResultsOPLS‐DA showed a clear separation between NC and IIM subtypes in plasma and urine metabolic profiles. KEGG pathway enrichment analysis revealed multiple unique and shared disturbed metabolic pathways in IIM main [dermatomyositis (DM), anti‐synthetase syndrome (ASS), and immune‐mediated necrotizing myopathy (IMNM)] and MSA‐defined subtypes (anti‐Mi2+, anti‐MDA5+, anti‐TIF1γ+, anti‐Jo1+, anti‐PL7+, anti‐PL12+, anti‐EJ+, and anti‐SRP+), such that fatty acid biosynthesis was significantly altered in both plasma and urine in all main IIM subtypes (enrichment ratio > 1). Random forest and AdaBoost performed best in classifying each IIM subtype among the 10 ML models. Using the feature selection methods in ML models, we identified 9 plasma and 10 urine metabolites that contributed most to separate IIM main subtypes and MSA‐defined subtypes, such as plasma creatine (fold change = 3.344, P = 0.024) in IMNM subtype and urine tiglylcarnitine (fold change = 0.351, P = 0.037) in anti‐EJ+ ASS subtype. Sixteen common metabolites were found in both the plasma and urine samples of IIM subtypes. Among them, some were correlated with clinical features, such as plasma hypogeic acid (r = −0.416, P = 0.005) and urine malonyl carnitine (r = −0.374, P = 0.042), which were negatively correlated with the prevalence of interstitial lung disease.ConclusionsIn both plasma and urine samples, IIM main and MSA‐defined subtypes have specific metabolic signatures and pathways. This study provides useful clues for understanding the molecular mechanisms, searching potential diagnosis biomarkers and therapeutic targets for IIM.

A novel cross-species differential tumor classification method based on exosome-derived microRNA biomarkers established by human-dog lymphoid and mammary tumor cell lines' transcription profiles.

Background and Aim:Exosome-derived microRNA (miRNA) has been widely studied as a non-invasive candidate biomarker for tumor diagnosis in humans and dogs. Its application, however, was primarily focused on intraspecies usage for individual tumor type diagnosis. This study aimed to gain insight into its application as a cross-species differential tumor diagnostic tool; we demonstrated the process of identifying and using exosome-derived miRNA as biomarkers for the classification of lymphoid and mammary tumor cell lines in humans and dogs.Materials and Methods:Exosome-derived miRNA sequencing data from B-cell lymphoid tumor cell lines (n=13), mammary tumor cell lines (n=8), and normal mammary epithelium cultures (n=4) were pre-processed in humans and dogs. F-test and rank product (RP) analyses were used to select candidate miRNA orthologs for tumor cell line classification. The classification was carried out using an optimized support vector machine (SVM) with various kernel classifiers, including linear SVM, polynomial SVM, and radial basis function SVM. The receiver operating characteristic and precision-recall curves were used to assess the performance of all models.Results:MIR10B, MIR21, and MIR30E were chosen as the candidate orthologs from a total of 236 human-dog miRNA orthologs (p≤0.01, F-test score ≥10, and RP score ≤10). Their use of polynomial SVM provided the best performance in classifying samples from various tumor cell lines and normal epithelial culture.Conclusion:The study successfully demonstrated a method for identifying and utilizing candidate human-dog exosome-derived miRNA orthologs for differential tumor cell line classification. Such findings shed light on a novel non-invasive tumor diagnostic tool that could be used in both human and veterinary medicine in the future.

Towards automated in vivo parcellation of the human cerebral cortex using supervised classification of magnetic resonance fingerprinting residuals.

Accurate parcellation of the cerebral cortex in an individual is a guide to its underlying organization. The most promising in vivo quantitative magnetic resonance (MR)-based microstructural cortical mapping methods are yet to achieve a level of parcellation accuracy comparable to quantitative histology. We scanned 6 participants using a 3D echo-planar imaging MR fingerprinting (EPI-MRF) sequence on a 7T Siemens scanner. After projecting MRF signals to the individual-specific inflated model of the cortical surface, normalized autocorrelations of MRF residuals of vertices of 8 microstructurally distinct areas (BA1, BA2, BA4a, BA6, BA44, BA45, BA17, and BA18) from 3 cortical regions were used as feature vector inputs into linear support vector machine (SVM), radial basis function SVM (RBF-SVM), random forest, and k-nearest neighbors supervised classification algorithms. The algorithms' prediction performance was compared using: (i) features from each vertex or (ii) features from neighboring vertices. The neighborhood-based RBF-SVM classifier achieved the highest prediction score of 0.85 for classification of MRF residuals in the central region from a held-out participant. We developed an automated method of cortical parcellation using a combination of MR fingerprinting residual analysis and machine learning classification. Our findings provide the basis for employing unsupervised learning algorithms for whole-cortex structural parcellation in individuals.

Pressure wrist pulse signal analysis by sparse decomposition using improved Gabor function

Background and objective: In traditional Chinese medicine and Ayurvedic medicine, wrist pulse wave fluctuations are an important indicator for distinguishing different health states. Owing to the development of modern sensing technology, computational methods have been used in the analysis of pulse wave signals. The description and quantification of the peaks in the pulse wave is significant for the identification of health status.Methods: In this study, we decomposed the pressure pulse waveform of the radial artery into several components by sparse decomposition with an improved Gabor function. To better represent the position, shape, and relationship of the peaks, we designed an improved Gabor function structure based on the characteristics of the pulse waveform to generate a time-frequency dictionary. Compared with conventional representation methods, the shape of the Gabor function is more variable. In addition, owing to the limitation of windowing, the Gabor function can reduce the influence on other positions when it represents a specific position. Feature vectors consisting of decomposed components can be used for computerized pulse signal analysis and disease diagnosis.Results: In the binary classification of healthy and diseased pulse signals, the proposed method achieved the best results for health/diabetes, health/cardiac disease, health/hypertension, and health/nephropathy with accuracies of 93.54%, 73.42%, 88.42%, and 82.28%, respectively. The multi-classification performance of the different types of features was evaluated by six classifiers, and the proposed method obtained the highest classification performance with support vector machine-radial basis function for both balanced and imbalanced data.Conclusions: The results indicated that the proposed method enabled to obtain a smaller representation error and exhibited superior performance in distinguishing between the signals collected from patients and healthy individuals. Moreover, for the multi-classification of the pulse signals, the proposed method performed better than the state-of-the-art methods.

Radiomic Applications on Digital Breast Tomosynthesis of BI-RADS Category 4 Calcifications Sent for Vacuum-Assisted Breast Biopsy.

Background: A fair amount of microcalcifications sent for biopsy are false positives. The study investigates whether quantitative radiomic features extracted from digital breast tomosynthesis (DBT) can be an additional and useful tool to discriminate between benign and malignant BI-RADS category 4 microcalcification. Methods: This retrospective study included 252 female patients with BI-RADS category 4 microcalcifications. The patients were divided into two groups according to micro-histopathology: 126 patients with benign lesions and 126 patients with certain or possible malignancies. A total of 91 radiomic features were extracted for each patient, and the 12 most representative features were selected by using the agglomerative hierarchical clustering method. The binary classification task of the two groups was carried out by using four different machine-learning algorithms (i.e., linear support vector machine (SVM), radial basis function (RBF) SVM, logistic regression (LR), and random forest (RF)). Accuracy, sensitivity, sensibility, and the area under the curve (AUC) were calculated for each of them. Results: The best performance was achieved using the RF classifier (AUC = 0.59, 95% confidence interval 0.57–0.60; sensitivity = 0.56, 95% CI 0.54–0.58; specificity = 0.61, 95% CI 0.59–0.63; accuracy = 0.58, 95% CI 0.57–0.59). Conclusions: DBT-based radiomic analysis seems to have only limited potential in discriminating benign from malignant microcalcifications.

Intelligent detection of hard seeds of snap bean based on hyperspectral imaging

As a common problem in snap beans, hard seed has seriously affected the large-scale industrial planting and yield of snap bean. To realize accurate, quick and non-destructive identifying the hard seeds of snap bean is of great significance to avoiding the effects of hard seeds on germination and growth. This research was based on hyperspectral imaging (HSI) to achieve accurate detection of hard seeds of snap bean. This study obtained the characteristic spectra from the hyperspectral image of a single seed, and then combined the synthetic minority over-sampling technique (SMOTE) and Tomek links to balance the numbers of hard and non-hard seed samples. The characteristic wavelengths were extracted from the average spectrum. Then the average spectrum was processed by first derivative (1D). After that, the characteristic wavelengths could be extracted using successive projections algorithm (SPA). Finally, a radial basis function-support vector machine (RBF-SVM) model was established to realize the intelligent detection of hard seeds, and the detection accuracy rate reached 89.32%. The research results showed that HSI technology could achieved accurate, fast and non-destructive testing of the hard seeds of snap bean, which is of great significance to the large-scale and standardized planting of snap bean and increase the yield per unit area.