Gaussian Support Vector Machine Research Articles

Efficient Machine Learning Techniques to Classifying Cardiovascular Disease and Improve Prediction Analysis

Cardiovascular Disease (CVD) account for a large portion of the global health burden and are one of the main causes of decease worldwide. In the classification and forecasting of CVDs, Machine Learning (ML) techniques have demonstrated encouraging outcomes. In this research report, a comparative analysis of classification and prediction models for CVD is presented, including both linear and ensemble ML approaches. The paper compares ensemble models like Catboost, Histogram Gradient Boosting Machine (HGBM), and Extra Trees against linear models like Gaussian Nave Bayes, SVM, and KNN. The objective is to identify the most effective CVD prediction model by assessing its performance through accuracy, precision, sensitivity and F1 score as key evaluation metrics. Moreover, results show that ensemble models outperform linear models using advanced techniques such as boosting and histogram-based algorithms. The results underscore the critical role ensemble models play in accurately diagnosing and predicting cardiovascular disease and provide important new information to researchers and healthcare providers. Using these models has the potential to significantly improve patient outcomes and health management by enabling early detection and intervention

Integrated database of granular soils under triaxial shear and its application in the prediction of stress–strain relationship

This study presents a novel data generation framework that generates a large database for machine learning (ML)-based soil model predictions. The dataset comprised 216 sets of triaxial tests on morphologically mutated and gene-decayed granular samples. This database was then estimated using five widely utilized ML algorithms to predict the stress-strain relationship of granular soils. They include the support vector machine (SVM), bagged trees, Gaussian process regression (GPR), and back propagation neural network (BPNN) algorithms. Following the hyperparameter settlement, model training, and testing, all the trained models captured the effects of the multiscale particle morphology, initial packing state, and confining stress. The excellent training and testing performances indicate the superior quality of the generated dataset. The fine tree, exponential GPR, and BPNN outperformed the Gaussian SVM and bagged trees in terms of the predictive performance. Among them, the exponential GPR exhibits the best model performance in reflecting the particle morphology effect, whereas the fine tree and BPNN generally exhibit good predictive performance for missing local information. Furthermore, all the models are tested by the micro-tomography (μCT) experimental data. The findings of this study were validated through a comparison between the DEM and model prediction results.

Development of a Novel HS-GC/MS Method Using the Total Ion Spectra Combined with Machine Learning for the Intelligent and Automatic Evaluation of Food-Grade Paraffin Wax Odor Level.

The intensity of the odor in food-grade paraffin waxes is a pivotal quality characteristic, with odor panel ratings currently serving as the primary criterion for its assessment. This study presents an innovative method for assessing odor intensity in food-grade paraffin waxes, employing headspace gas chromatography with mass spectrometry (HS/GC-MS) and integrating total ion spectra with advanced machine learning (ML) algorithms for enhanced detection and quantification. Optimization was conducted using Box-Behnken design and response surface methodology, ensuring precision with coefficients of variance below 9%. Analytical techniques, including hierarchical cluster analysis (HCA) and principal component analysis (PCA), efficiently categorized samples by odor intensity. The Gaussian support vector machine (SVM), random forest, partial least squares regression, and support vector regression (SVR) algorithms were evaluated for their efficacy in odor grade classification and quantification. Gaussian SVM emerged as superior in classification tasks, achieving 100% accuracy, while Gaussian SVR excelled in quantifying odor levels, with a coefficient of determination (R2) of 0.9667 and a root mean square error (RMSE) of 6.789. This approach offers a fast, reliable, robust, objective, and reproducible alternative to the current ASTM sensory panel assessments, leveraging the analytical capabilities of HS-GC/MS and the predictive power of ML for quality control in the petrochemical sector's food-grade paraffin waxes.

Grading the Dominant Pathological Indices in Liver Diseases from Pathological Images Using Radiomics Methods

Purpose: This study aims to diagnose the severity of important pathological indices, i.e., fibrosis, steatosis, lobular inflammation, and ballooning from the pathological images of the liver tissue based on extracted features by radiomics methods.
 Materials and Methods: This research uses the pathological images obtained from liver tissue samples for 258 laboratory mice. After preprocessing the images and data augmentation, a collection of texture feature sets extracted by gray-level-based algorithms, including Global, Gray-level Co-Occurrence Matrix (GLCM), Gray-level Run length Matrix (GLRLM), Gray-level Size Zone Matrix (GLSZM), and Neighboring Gray Tone Difference Matrix (NGTDM) algorithms. Then, advanced methods of classification, namely Support Vector Machine (SVM), Random Forest (RF), Quadratic Discriminant Analysis (QDA), K-Nearest Neighbors (KNN), Logistic Regression (LR), Naïve Bayes (NB), and Multi-layer Perceptrons (MLP) are employed. This procedure is provided separately for each of the four indices of fibrosis level in 6 grading classes, steatosis in 5 grading classes, inflammation in 4 grading classes, and ballooning in 3 grading classes. For a comparison of the output of these algorithms, the accuracy value obtained from the evaluation data is presented for the performance of different methods.
 Results: The results showed that, compared to other methods, the Gaussian SVM algorithm provides a better response to the classification of the grading of liver disease among all the indices from the pathological images due to its structural features. This value of accuracy was calculated at 84.30% for fibrosis, 90.55% for steatosis, 81.11% for inflammation, and 95.98% for ballooning.
 Conclusion: This fully automatic framework based on advanced radiomics algorithms and machine learning from pathological images can be very useful in clinical procedures and be considered as an assistant or a substitute for pathologists’ diagnoses.

Assessment of food safety risk using machine learning-assisted hyperspectral imaging: Classification of fungal contamination levels in rice grain

A rapid and nondestructive assessment of food safety risk using machine learning-assisted hyperspectral imaging was developed for classification of fungal contamination in brown rice grain. Brown rice was inoculated with Penicillium. The fungal infected rice was then mixed with healthy rice to obtain 0 %, 5 %, 25 %, 50 % and 100 % (w/w) contamination of infected rice. Volatile compounds including pentamethyl-heptane, decane, dodecane, 3-octanone, and 1-octen-3-ol were found in fungal infected rice, as analyzed using gas chromatography-mass spectrometry. The HSI system was used to collect spectral reflectance and spatial data of the samples covering the wavelength range of 400–1000 nm. The hypercubed data were analyzed using machine learning algorithms, including principal component analysis (PCA), discriminant factor analysis (DFA) and support vector machine (SVM). Using PCA for data reduction, 3 principal components were extracted with a cumulative variance of 90.53 %. DFA (linear and quadratic algorithms) and SVM (linear, quadratic, cubic, and Gaussian algorithms) were then used to classify the samples. HSI integrated with Gaussian SVM gave 93.4% accuracy which was best for classifying rice with different percentages of contamination. The image analysis gave a pseudo-color distribution map which facilitated the visualization of the contaminated rice by presenting data in an uncomplicated image. The machine learning-assisted HSI can be used as a rapid, nondestructive and chemical-free tool for an assessment of food safety risk for rice grain.

Intelligent classification of major depressive disorder using rs-fMRI of the posterior cingulate cortex

Major Depressive Disorder (MDD) is a widespread psychiatric condition that affects a significant portion of the global population. The classification and diagnosis of MDD is crucial for effective treatment. Traditional methods, based on clinical assessment, are subjective and rely on healthcare professionals' expertise. Recently, there's growing interest in using Resting-State Functional Magnetic Resonance Imaging (rs-fMRI) to objectively understand MDD's neurobiology, complementing traditional diagnostics. The posterior cingulate cortex (PCC) is a pivotal brain region implicated in MDD which could be used to identify MDD from healthy controls. Thus, this study presents an intelligent approach based on rs-fMRI data to enhance the classification of MDD.Original rs-fMRI data were collected from a cohort of 430 participants, comprising 197 patients and 233 healthy controls. Subsequently, the data underwent preprocessing using DPARSF, and the amplitudes of low-frequency fluctuation values were computed to reduce data dimensionality and feature count. Then data associated with the PCC were extracted. After eliminating redundant features, various types of Support Vector Machines (SVMs) were employed as classifiers for intelligent categorization. Ultimately, we compared the performance of each algorithm, along with its respective optimal classifier, based on classification accuracy, true positive rate, and the area under the receiver operating characteristic curve (AUC-ROC). Upon analyzing the comparison results, we determined that the Random Forest (RF) algorithm, in conjunction with a sophisticated Gaussian SVM classifier, demonstrated the highest performance. Remarkably, this combination achieved a classification accuracy of 81.9 % and a true positive rate of 92.9 %.In conclusion, our study improves the classification of MDD by supplementing traditional methods with rs-fMRI and machine learning techniques, offering deeper neurobiological insights and aiding accuracy, while emphasizing its role as an adjunct to clinical assessment.

An Analysis of the WPT Function for Pattern Optimization to Detect Defects in Bearings

New trends in maintenance techniques are oriented to digitization and prognosis. The new electronic devices based on IoT (Internet of Things) technology among others that support the industry 4.0 paradigm let enhance the traditional condition monitoring techniques to better understand and predict the state of a machine in service. Related to maintenance applications, one of the important steps in condition monitoring tasks for fault diagnosis is the selection of the optimal pattern to provide accurate results (avoiding fault positives/negatives) with adequate computation time. When implementing this, the selection of optimal parameters and thresholds for setting alarms are important to detect problems in the machine before the failure occurs. Vibratory signals have been proved to be a good variable to determine their mechanical behavior. Nevertheless, parameters obtained from time domain measurements are not computationally efficient nor good patterns to compare different machine conditions. In this sense, tools that represent the frequency domain or time–frequency domain have been useful to detect defects in rotating elements such as bearings. In this work, defects in ball bearings are studied using wavelet packet transform. For this, a methodology will be developed for the optimal selection of the mother wavelet, incorporating intelligent classification systems, and using a medium Gaussian support vector machine model. In this way, it will be verified that the correct selection of this function influences both the results and the ease and reliability of detection. The results using the selected mother wavelet will be compared to those using Daubechies 6, since it is the mother wavelet that has been used in previous works and which was selected based on experience. For it, vibratory signals are obtained from a testbench with different bearing conditions: healthy bearings and defective bearings (inner and outer race).

Developing a novel image marker to predict the clinical outcome of neoadjuvant chemotherapy (NACT) for ovarian cancer patients

ObjectiveNeoadjuvant chemotherapy (NACT) is one kind of treatment for advanced stage ovarian cancer patients. However, due to the nature of tumor heterogeneity, the clinical outcomes to NACT vary significantly among different subgroups. Partial responses to NACT may lead to suboptimal debulking surgery, which will result in adverse prognosis. To address this clinical challenge, the purpose of this study is to develop a novel image marker to achieve high accuracy prognosis prediction of NACT at an early stage. MethodsFor this purpose, we first computed a total of 1373 radiomics features to quantify the tumor characteristics, which can be grouped into three categories: geometric, intensity, and texture features. Second, all these features were optimized by principal component analysis algorithm to generate a compact and informative feature cluster. This cluster was used as input for developing and optimizing support vector machine (SVM) based classifiers, which indicated the likelihood of receiving suboptimal cytoreduction after the NACT treatment. Two different kernels for SVM algorithm were explored and compared. A total of 42 ovarian cancer cases were retrospectively collected to validate the scheme. A nested leave-one-out cross-validation framework was adopted for model performance assessment. ResultsThe results demonstrated that the model with a Gaussian radial basis function kernel SVM yielded an AUC (area under the ROC [receiver characteristic operation] curve) of 0.806 ± 0.078. Meanwhile, this model achieved overall accuracy (ACC) of 83.3%, positive predictive value (PPV) of 81.8%, and negative predictive value (NPV) of 83.9%. ConclusionThis study provides meaningful information for the development of radiomics based image markers in NACT treatment outcome prediction.

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.

SUPPORT VECTOR MACHINES FOR FORECASTING NON-SCHEDULED PASSENGER AIR TRANSPORTATION

Forecasting non-scheduled passenger air transportation demand is essential for effective operational planning and decision-making. In this abstract, we explore the use of Gaussian Support Vector Machines (SVM) methods in forecasting nonscheduled passenger air transportation processes. SVM is a type of supervised machine learning algorithm that can be applied to various domains, including nonscheduled passenger air transportation. In classification and regression tasks, SVMs are considered especially useful. SVMs can be used to forecast passenger demand for specific routes or flights. By analysing historical data, including factors such as time of day, day of the week, etc., SVMs can help airlines estimate future passenger demand. This method is crucial for optimising ticket pricing and managing seat inventory. This research proposes the implementation of different Gaussian SVM methods for the forecasting of non-scheduled passenger air transportation.

The Lassa fever cases and mortality in Nigeria: Quantile Regression versus Machine Learning Models

Introduction. Lassa fever (LF) is caused by the Lassa fever virus (LFV). It is endemic in West Africa, of which 25% of the infections are ascribed to Nigeria. This disease affects mostly the productive age and hence a proper understanding of the dynamics of this disease will help in formulating policies that would help in curbing the spread of LF. Objectives. The objective of this study is to compare the performance of quantile regression models with that of Machine Learning models. Methods. Data between between 7th January 2018 2018 and 17th December, 2022 on suspected cases, confirmed cases and deaths resulting from LF were retrieved from the Nigeria Centre for Disease Control (NCDC). The data obtained were fitted to quantile regression models (QRM) at 25%, 50% and 75% as well as to Machine learning models. The response variable being confirmed cases and mortality due to Lassa fever in Nigeria while the independent variables were total confirmed cases, the week, month and year. Results. Result showed that the highest monthly mean confirmed cases (56) and mortality (9) from LF were reported in February. The first quarter of the year reported the highest cases of both confirmed cases and deaths in Nigeria. Result also revealed that for the confirmed cases, quantile regression at 50% outperformed the best of the MLM, Gaussian-matern5/2 GPR (RMSE= 10.3393 versus 11.615), while for mortality, the medium Gaussian SVM (RMSE =1.6441 versus 1.8352) outperformed QRM. Conclusion. Quantile regression model at 50% better captured the dynamics of the confirmed cases of LF in Nigeria while the medium Gaussian SVM better captured the mortality of LF in Nigeria. Among the features selected, confirmed cases was found to be the most important feature that drive its mortality with the implication that as the confirmed cases of Lassa fever increases, is a significant increase in its mortality. This therefore necessitates a need for a better intervention measures that will help curb Lassa fever mortality as a result of the increase in the confirmed cases. There is also a need for promotion of good community hygiene which could include; discouraging rodents from entering homes and putting food in rodent proof containers to avoid contamination to help hart the spread of Lassa fever in Nigeria.

Classification of Weaving Motifs Based on Their Area of Origin Using the Support Vector Machine Algorithm

Indonesia has many cultural riches in the form of traditional fabrics, one of which is woven fabrics. Woven fabrics from each region showcase distinctive motifs, manifesting the local community’s daily life, culture, natural conditions, and beliefs. The diverse weaving motifs pose a challenge in determining the origin of the woven fabrics. It highlights the necessity of a system to detect and identify woven fabrics. Texture analysis was performed using the Gray Level Co-occurrence Matrix (GLCM). A classification method based on a Support Vector Machine (SVM) consisting of four models: Linear SVM, Quadratic SVM, Cubic SVM, and Fine Gaussian SVM was developed in this research. The images of woven fabrics came from three regions in Indonesia: Sumatra, Kalimantan, and Nusa Tenggara. This research utilized 240 training images and 12 testing images. The testing results unveiled that the Cubic SVM model, which achieved a 100% accuracy rate in 1.0835s, was the optimum SVM model for the weaving classification.

G-Induced Loss of Consciousness Prediction Using a Support Vector Machine.

INTRODUCTION: Gravity-induced loss of consciousness (G-LOC) is a major threat to fighter pilots and may result in fatal accidents. The brain has a period of 5-6 s from the onset of high +Gz exposure, called the functional buffer period, during which transient ischemia is tolerated without loss of consciousness. We tried to establish a method for predicting G-LOC within the functional buffer period by using machine learning. We used a support vector machine (SVM), which is a popular classification algorithm in machine learning.METHODS: The subjects were 124 flight course students. We used a linear soft-margin SVM, a nonlinear SVM Gaussian kernel function (GSVM), and a polynomial kernel function, for each of which 10 classifiers were built every 0.5 s from the onset of high +Gz exposure (Classifiers 0.5-5.0) to predict G-LOC. Explanatory variables used for each SVM were age, height, weight, with/without anti-G suit, +Gz level, cerebral oxyhemoglobin concentration, and deoxyhemoglobin concentration.RESULTS: The performance of GSVM was better than that of other SVMs. The accuracy of each classifier of GSVM was as follows: Classifier 0.5, 58.1%; 1.0, 54.8%; 1.5, 57.3%; 2.0, 58.1%; 2.5, 64.5%; 3.0, 63.7%; 3.5, 65.3%; 4.0, 64.5%; 4.5, 64.5%; and 5.0, 64.5%.CONCLUSION: We could predict G-LOC with an accuracy rate of approximately 65% from 2.5 s after the onset of high +Gz exposure by using GSVM. Analysis of a larger number of cases and factors to enhance accuracy may be needed to apply those classifiers in centrifuge training and actual flight.Ohrui N, Iino Y, Kuramoto K, Kikukawa A, Okano K, Takada K, Tsujimoto T. G-induced loss of consciousness prediction using a support vector machine. Aerosp Med Hum Perform. 2024; 95(1):29-36.

A real-time air-writing model to recognize Bengali characters

<abstract><p>Air-writing is a widely used technique for writing arbitrary characters or numbers in the air. In this study, a data collection technique was developed to collect hand motion data for Bengali air-writing, and a motion sensor-based data set was prepared. The feature set as then utilized to determine the most effective machine learning (ML) model among the existing well-known supervised machine learning models to classify Bengali characters from air-written data. Our results showed that medium Gaussian SVM had the highest accuracy (96.5%) in the classification of Bengali character from air writing data. In addition, the proposed system achieved over 81% accuracy in real-time classification. The comparison with other studies showed that the existing supervised ML models predicted the created data set more accurately than many other models that have been suggested for other languages.</p></abstract>

Lassa fever cases and mortality in Nigeria: Quantile Regression vs. Machine Learning Models.

Lassa fever (LF) is caused by the Lassa fever virus (LFV). It is endemic in West Africa, of which % of the infections are ascribed to Nigeria. This disease affects mostly the productive age and hence a proper understanding of the dynamics of this disease will help in formulating policies that would help in curbing the spread of LF. The objective of this study is to compare the performance of quantile regression models with that of Machine Learning models in. Data between between 7th January 2018 2018 and 17th December, 2022 on suspected cases, confirmed cases and deaths resulting from LF were retrieved from the Nigeria Centre for Disease Control (NCDC). The data obtained were fitted to quantile regression models (QRM) at 25, 50 and 75% as well as to Machine learning models. The response variable being confirmed cases and mortality due to Lassa fever in Nigeria while the independent variables were total confirmed cases, the week, month and year. Result showed that the highest monthly mean confirmed cases (56) and mortality (9) from LF were reported in February. The first quarter of the year reported the highest cases of both confirmed cases and deaths in Nigeria. Result also revealed that for the confirmed cases, quantile regression at 50% outperformed the best of the MLM, Gaussian-matern5/2 GPR (RMSE=10.3393 vs. 11.615), while for mortality, the medium Gaussian SVM (RMSE=1.6441 vs. 1.8352) outperformed QRM. Quantile regression model at 50% better captured the dynamics of the confirmed cases of LF in Nigeria while the medium Gaussian SVM better captured the mortality of LF in Nigeria. Among the features selected, confirmed cases was found to be the most important feature that drive its mortality with the implication that as the confirmed cases of Lassa fever increases, is a significant increase in its mortality. This therefore necessitates a need for a better intervention measures that will help curb Lassa fever mortality as a result of the increase in the confirmed cases. There is also a need for promotion of good community hygiene which could include; discouraging rodents from entering homes and putting food in rodent proof containers to avoid contamination to help hart the spread of Lassa fever in Nigeria.

Depression Identification Using Asymmetry Matrix and Machine Learning Algorithms

Purpose: Developing an efficient and reliable method for the identification of depression is highly important. This paper aims to propose an approach for depression diagnosis using an interhemispheric asymmetry matrix and machine learning algorithms.
 Materials and Methods: First, an EEG signal was acquired from 24 depressed patients and 24 healthy subjects. The EEG signal was acquired from participants for 5 minutes in Eyes-Closed (EC) and 5 minutes in Eyes-Open (EO) condition. After preprocessing data, interhemispheric asymmetry for absolute and relative powers of theta and beta frequency bands, theta-to-alpha power ratio, and Individual Alpha Frequency (IAF) features were computed. Then, the proposed asymmetry matrix is used as a feature for statistical and classification analysis. In this paper, the classification was performed using a Support Vector Machine (SVM), logistic regression, and Multi-Layer Perceptron (MLP).
 Results: The results demonstrated that central and temporal theta absolute power, central and temporal IAF asymmetries in the EC condition and occipital beta absolute power, temporal theta relative power, temporal theta-to-alpha power ratio, and temporal IAF asymmetries in the EO condition have significant differences between depressed and healthy groups. Findings show that beta absolute power asymmetry in the occipital region and EO condition is a good biomarker for depression identification with 77.1% accuracy using the Gaussian SVM classifier.
 Conclusion: The results of this study show performance of proposed asymmetry matrix features in depression detection. Findings show that beta absolute power asymmetry in the occipital region and EO condition is a good biomarker for depression identification.

Optimized Machine learning techniques Enable prediction of organic dyes photophysical Properties: Absorption Wavelengths, emission Wavelengths, and quantum yields

Applications of organic dyes, ranging from basic research to industry, are functions of their photophysical properties. Two important aspects— (1) knowledge of the photophysical properties of existing dyes long before real applications and (2) discovery of new organic dyes with desired photophysical properties for either upgradation of existing or development of new applications—are needed to be addressed. These two cases are coupled together with the common goal of estimating photophysical properties with high accuracy at the minimum cost of time and money long before the hard-core laboratory experiment. For this purpose, machine learning-based techniques are the most suitable approach. In this study, we used optimized machine-learning techniques to assess a dataset of 3066 organic dyes, which were evaluated using three evaluation parameters: Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and the coefficient of determination (R2). The Quadratic Support Vector Machine (QSVM) was the best predictive model for RMSE-16.614, MAE-10.837, and R2-0.961 for absorption wavelengths and RMSE-23.636, MAE-16.278, and R2-0.929 for emission wavelengths. These R2 values are 0.7% and 0.4% greater than the Gradient Boost Regression Tree (GBRT) model's recently reported values of 0.954 and 0.925 for absorption and emission wavelengths, respectively. Furthermore, we estimated the quantum yield and found that the Coarse Gaussian Support Vector Machine (CGSVM) outperformed all examined models. For more validation of these models, we compared the predicted results with the experimental results of selective dyes. The proposed automated approach can be used for predicting photophysical properties without much computer programming knowledge.

Development and Validation of a Computed Tomography-Based Radiomics Nomogram for the Preoperative Prediction of Central Lymph Node Metastasis in Papillary Thyroid Microcarcinoma

This study aims to develop and validate a computed tomography (CT)-based radiomics nomogram for pre-operatively predicting central lymph node metastasis (CLNM) in patients with papillary thyroid microcarcinoma (PTMC) and explore the underlying biological basis by using RNA sequencing data. This study trained 452 PTMC patients across two hospitals from January 2012 to December 2020. The sets were randomly divided into the training (n = 339), internal test (n = 86), external test (n = 27) sets. Radiomics features were extracted from primary lesion's pre-operative CT images for each patient. After screening for features, five algorithms such as K-nearest neighbor, logistics regression, linear-support vector machine (SVM), Gaussian SVM, and polynomial SVM were used to establish the radiomics models. The performance of these five algorithms was evaluated and compared directly to radiologist's interpretation (CT-reported lymph node status). The radiomics signature score (Rad-score) was generated using a linear combination of the selected features. By combining the clinical risk factors and Rad score, a radiomics nomogram was established and compared with Rad-score and clinical model. The performance of the nomogram was evaluated based on the receiver operating characteristic (ROC) curve, calibration curve, and the decision curve analysis (DCA). The potential biological basis of nomogram was revealed by performing genetic analysis based on the RNA sequencing data. A total of 25 radiomic features were ultimately selected to train the machine learning models, and the five machine learning models outperformed the radiologists' interpretation by achieving area under the ROC curves (AUCs) ranging from 0.606 to 0.730 in the internal test set. By incorporating the Rad score and clinical risk factors (sex, age, tumor-diameter, and CT-reported lymph node status), this nomogram achieved AUCs of 0.800 and 0.803 in the internal and external test set, which were higher than that of the Rad-score and clinical model, respectively. Calibration curves and DCA also showed that the nomogram had good performance. As for the biological basis exploration, in patients predicted by nomogram to be PTC patients with CLMN, 109 genes were dysregulated, and some of them were associated with pathways and biological processes such as tumor angiogenesis. This radiomics nomogram successfully identified CLNM on pretreatment imaging across multiple institutions, exceeding the diagnostic ability of radiologists and had the potential to be integrated into clinical decision making as a non-invasive pre-operative tool.

Development of efficient techniques for ASR System for Speech Detection and Recognization system using Gaussian Mixture Model- Universal Background Model

Some practical uses of ASR have been implemented, including the transcription of meetings and the usage of smart speakers. It is the process by which speech waves are transformed into text that allows computers to interpret and act upon human speech. Scalable strategies for developing ASR systems in languages where no voice transcriptions or pronunciation dictionaries exist are the primary focus of this work. We first show that the necessity for voice transcription into the target language can be greatly reduced through cross-lingual acoustic model transfer when phonemic pronunciation lexicons exist in the new language. Afterwards, we investigate three approaches to dealing with languages that lack a pronunciation lexicon. Secondly, we have a look at the efficiency of graphemic acoustic model transfer, which makes it easy to build pronunciation dictionaries. Thesis problems can be solved, in part, by investigating optimization strategies for training on huge corpora (such as GA+HMM and DE+HMM). In the training phase of acoustic modelling, the suggested method is applied to traditional methods. Read speech and HMI voice experiments indicated that while each data augmentation strategy alone did not always increase recognition performance, using all three techniques together did. Power normalised cepstral coefficient (PNCC) features are tweaked somewhat in this work to enhance verification accuracy. To increase speaker verification accuracy, we suggest employing multiple “Gaussian Mixture Model-Universal Background Model (GMM-UBM) and SVM classifiers”. Importantly, pitch shift data augmentation and multi-task training reduced bias by more than 18% absolute compared to the baseline system for read speech, and applying all three data augmentation techniques during fine tuning reduced bias by more than 7% for HMI speech, while increasing recognition accuracy of both native and non-native Dutch speech.

Gastrointestinal tract disorders classification using ensemble of InceptionNet and proposed GITNet based deep feature with ant colony optimization.

Computer-aided classification of diseases of the gastrointestinal tract (GIT) has become a crucial area of research. Medical science and artificial intelligence have helped medical experts find GIT diseases through endoscopic procedures. Wired endoscopy is a controlled procedure that helps the medical expert in disease diagnosis. Manual screening of the endoscopic frames is a challenging and time taking task for medical experts that also increases the missed rate of the GIT disease. An early diagnosis of GIT disease can save human beings from fatal diseases. An automatic deep feature learning-based system is proposed for GIT disease classification. The adaptive gamma correction and weighting distribution (AGCWD) preprocessing procedure is the first stage of the proposed work that is used for enhancing the intensity of the frames. The deep features are extracted from the frames by deep learning models including InceptionNetV3 and GITNet. Ant Colony Optimization (ACO) procedure is employed for feature optimization. Optimized features are fused serially. The classification operation is performed by variants of support vector machine (SVM) classifiers, including the Cubic SVM (CSVM), Coarse Gaussian SVM (CGSVM), Quadratic SVM (QSVM), and Linear SVM (LSVM) classifiers. The intended model is assessed on two challenging datasets including KVASIR and NERTHUS that consist of eight and four classes respectively. The intended model outperforms as compared with existing methods by achieving an accuracy of 99.32% over the KVASIR dataset and 99.89% accuracy using the NERTHUS dataset.