Support Vector Machine Research Articles

Enhancing sound source localization and music teaching through integrated computational resource allocation

In contemporary educational and computational settings, the incorporation of cutting-edge technologies like sound source localization and personalized music teaching helps in offering an effective resource allocation strategies. Previous systems for sound localization and music teaching frequently lacked real-time flexibility and effective resource use, reducing their efficiency in dynamic learning settings and tasks involving computation. To overcome these shortcomings, the SoundLocMusicTeachRA (SLMTRA) algorithm is presented, a single, integrated platform made to maximize sound localization accuracy, improve music teaching efficiency, and enhance computational resource oversight. However, the existing study did not highlight the importance of computation resource allocation but this proposed algorithm will address it. SLMTRA uses a new Bagging ensemble approach incorporating Random Forest (RF), Decision Trees (DT), Naive Bayes (NB), Support Vector Machine (SVM), and K-Nearest Neighbor (KNN), with hyperparameter tuning to enhance the effectiveness of the approach. These classifiers are trained utilizing sound localization datasets from recordings made with microphones, music teaching feedback datasets from data on student performance, and resource allocation datasets from metrics for computer utilization. Experimental findings indicate SLMTRA’s high accuracy in sound source localization, improved music teaching feedback capacities, as well as effective resource allocation tactics, guaranteeing the best performance of the system. The implementation of SLMTRA represents a noteworthy development in combining sound localization, music teaching, and resource allocation within a unified computational framework, offering a more flexible and effective system compared to previous methodologies.

Smartphone-Based Leaf Colorimetric Analysis of Grapevine (Vitis vinifera L.) Genotypes

Leaf chlorophyll content is a key indicator of plant physiological status in viticulture; therefore, regular evaluation to obtain data for nutrient supply and canopy management is of vital importance. The measurement of pigmentation is most frequently carried out with hand-held instruments, destructive off-site spectrophotometry, or remote sensing. Moreover, smartphone-based applications also ensure a promising way to collect colorimetric information that could correlate with pigmentation. In this study, four grapevine genotypes were investigated using smartphone-based RGB (Red, Green, Blue) and CIE-L*a*b* colorimetry and a portable chlorophyll meter. The objective of this study was to evaluate the correlation between leaf chlorophyll concentration and RGB- or CIE-L*a*b*-based color indices. A further aim was to find an appropriate model for discriminating between the genotypes by leaf coloration. For these purposes, fully developed leaves of ‘Chardonnay’, ‘Sauvignon blanc’, and ‘Pinot noir’ clones 666 and 777 were investigated with the Color Grab smartphone application to obtain RGB and CIE-L*a*b* values. Using these color values, chroma, hue, and a further 31 color indices were calculated. Chlorophyll concentrations were determined using an Apogee MC100 device, and the values were correlated with color values and color indices. The results showed that the chlorophyll concentration and color indices significantly differed between the genotypes. According to the results, certain color indices show a different direction in their relationship with leaf pigmentation for different grapevine genotypes. The same index showed a positive correlation for the leaf chlorophyll concentration for one variety and a negative correlation for another, which raises the possibility that the relationship is genotype-specific and not uniform within species. In light of this result, further study of the species specificity of the commonly used vegetation indices is warranted. Support Vector Machine (SVM) analysis of the samples based on color properties showed a 71.63% classification accuracy, proving that coloration is an important ampelographic feature for the identification and assessment of true-to-typeness.

Multi-View Fusion-Based Automated Full-Posture Cattle Body Size Measurement

Cattle farming is an important part of the global livestock industry, and cattle body size is the key indicator of livestock growth. However, traditional manual methods for measuring body sizes are not only time-consuming and labor-intensive but also incur significant costs. Meanwhile, automatic measurement techniques are prone to being affected by environmental conditions and the standing postures of livestock. To overcome these challenges, this study proposes a multi-view fusion-driven automatic measurement system for full-attitude cattle body measurements. Outdoors in natural light, three Zed2 cameras were installed covering different views of the channel. Multiple images, including RGB images, depth images, and point clouds, were automatically acquired from multiple views using the YOLOv8n algorithm. The point clouds from different views undergo multiple denoising to become local point clouds of the cattle body. The local point clouds are coarsely and finely aligned to become a complete point cloud of the cattle body. After detecting the 2D key points on the RGB image created by the YOLOv8x-pose algorithm, the 2D key points are mapped onto the 3D cattle body by combining the internal parameters of the camera and the depth values of the corresponding pixels of the depth map. Based on the mapped 3D key points, the body sizes of cows in different poses are automatically measured, including height, length, abdominal circumference, and chest circumference. In addition, support vector machines and Bézier curves are employed to rectify the missing and deformed circumference body sizes caused by environmental effects. The automatic body measurement system measured the height, length, abdominal circumference, and chest circumference of 47 Huaxi Beef Cattle, a breed native to China, and compared the results with manual measurements. The average relative errors were 2.32%, 2.27%, 3.67%, and 5.22%, respectively, when compared with manual measurements, demonstrating the feasibility and accuracy of the system.

Machine learning-based design of double corrugated steel plate shear walls

PurposeIn this study, machine learning (ML) algorithms were employed to predict the shear capacity and behavior of DCSWs.Design/methodology/approachIn this study, ML algorithms were employed to predict the shear capacity and behavior of DCSWs. Various ML techniques, including linear regression (LR), support vector machine (SVM), decision tree (DT), random forest (RF), extreme gradient boosting (XGBoost) and artificial neural network (ANN), were utilized. The ML models were trained using a dataset of 462 numerical and experimental samples. Numerical models were generated and analyzed using the finite element (FE) software Abaqus. These models underwent push-over analysis, subjecting them to pure shear conditions by applying a target displacement solely to the top of the shear walls without interaction from a frame. The input data encompassed eight survey variables: geometric values and material types. The characterization of input FE data was randomly generated within a logical range for each variable. The training and testing phases employed 90 and 10% of the data, respectively. The trained models predicted two output targets: the shear capacity of DCSWs and the likelihood of buckling. Accurate predictions in these areas contribute to the efficient lateral enhancement of structures. An ensemble method was employed to enhance capacity prediction accuracy, incorporating select algorithms.FindingsThe proposed model achieved a remarkable 98% R-score for estimating shear strength and a corresponding 98% accuracy in predicting buckling occurrences. Among all the algorithms tested, XGBoost demonstrated the best performance.Originality/valueIn this study, for the first time, ML algorithms were employed to predict the shear capacity and behavior of DCSWs.

Multi-label text classification via secondary use of large clinical real-world data sets.

Procedural coding presents a taxing challenge for clinicians. However, recent advances in natural language processing offer a promising avenue for developing applications that assist clinicians, thereby alleviating their administrative burdens. This study seeks to create an application capable of predicting procedure codes by analysing clinicians' operative notes, aiming to streamline their workflow and enhance efficiency. We downstreamed an existing and a native German medical BERT model in a secondary use scenario, utilizing already coded surgery notes to model the coding procedure as a multi-label classification task. In comparison to the transformer-based architecture, we were levering the non-contextual model fastText, a convolutional neural network, a support vector machine and logistic regression for a comparative analysis of possible coding performance. About 350,000 notes were used for model adaption. By considering the top five suggested procedure codes from medBERT.de, surgeryBERT.at, fastText, a convolutional neural network, a support vector machine and a logistic regression, the mean average precision achieved was 0.880, 0.867, 0.870, 0.851, 0.870and0.805 respectively. Support vector machines performed better for surgery reports with a sequence length greater than 512, achieving a mean average precision of 0.872 in comparison to 0.840 for fastText, 0.837 for medBERT.de and 0.820 for surgeryBERT.at. A prototypical front-end application for coding support was additionally implemented. The problem of predicting procedure codes from a given operative report can be successfully modelled as a multi-label classification task, with a promising performance. Support vector machines as a classical machine learning method outperformed the non-contextual fastText approach. FastText with less demanding hardware resources has reached a similar performance to BERT-based models and has shown to be more suitable for explaining the predictions efficiently.

Prediction of Early Mortality in Esophageal Cancer Patients with Liver Metastasis Using Machine Learning Approaches

Patients with esophageal cancer liver metastasis face a high risk of early mortality, making accurate prediction crucial for guiding clinical decisions. However, effective predictive tools are currently limited. In this study, we used clinicopathological data from 1897 patients diagnosed with esophageal cancer liver metastasis between 2010 and 2020, which were sourced from the SEER database. Prognostic factors were identified using univariate and multivariate logistic regression, and seven machine learning models, including extreme gradient boosting (XGBoost) and support vector machine (SVM), were developed to predict early mortality. The models were evaluated using Receiver Operating Characteristic (ROC) curves, calibration curves, decision curve analysis (DCA), and F1 scores. Results showed that 40% of patients experienced all-cause early mortality and 38% had cancer-specific early mortality. Key predictors of early mortality included age, location, chemotherapy, and lung metastasis. Among the models, XGBoost performed best in predicting all-cause early mortality, while SVM excelled in predicting cancer-specific early mortality. These findings demonstrate that machine learning models, particularly XGBoost and SVM, can serve as valuable tools for predicting early mortality in patients with esophageal cancer liver metastasis, aiding clinical decision making.

INNOVATIVE MACHINE LEARNING APPROACHES TO FOSTER FINANCIAL INCLUSION IN MICROFINANCE

This study examines the application of machine learning algorithms to enhance financial inclusion in microfinance, focusing on credit scoring, risk and fraud detection, and customer segmentation. We performed feature engineering and employed models such as Logistic Regression, Decision Trees, Random Forests, Gradient Boosting Machines (XGBoost and LightGBM), Support Vector Machines (SVM), Autoencoders, Isolation Forests, and K-means Clustering. LightGBM achieved the highest accuracy (89.6%) and AUC (0.92) in credit scoring, while Random Forests demonstrated strong performance in both loan approval (86.7% accuracy) and fraud detection (87.6% accuracy, AUC of 0.88). SVM also performed competitively, and unsupervised methods like Autoencoders and Isolation Forests showed potential for anomaly detection but required further refinement.K-means Clustering excelled in customer segmentation with a silhouette score of 0.72, enabling tailored services based on client demographics. Our findings highlight the significant impact of machine learning on improving credit scoring accuracy, reducing fraud risks, and enhancing customer service delivery in microfinance, thereby promoting financial inclusion for underserved populations. Ethical considerations and model interpretability are crucial, particularly for smaller institutions. This study advocates for the broader adoption of machine learning in the microfinance sector.

Advanced Image Preprocessing and Integrated Modeling for UAV Plant Image Classification

The automatic identification of plant species using unmanned aerial vehicles (UAVs) is a valuable tool for ecological research. However, challenges such as reduced spatial resolution due to high-altitude operations, image degradation from camera optics and sensor limitations, and information loss caused by terrain shadows hinder the accurate classification of plant species from UAV imagery. This study addresses these issues by proposing a novel image preprocessing pipeline and evaluating its impact on model performance. Our approach improves image quality through a multi-step pipeline that includes Enhanced Super-Resolution Generative Adversarial Networks (ESRGAN) for resolution enhancement, Contrast-Limited Adaptive Histogram Equalization (CLAHE) for contrast improvement, and white balance adjustments for accurate color representation. These preprocessing steps ensure high-quality input data, leading to better model performance. For feature extraction and classification, we employ a pre-trained VGG-16 deep convolutional neural network, followed by machine learning classifiers, including Support Vector Machine (SVM), random forest (RF), and Extreme Gradient Boosting (XGBoost). This hybrid approach, combining deep learning for feature extraction with machine learning for classification, not only enhances classification accuracy but also reduces computational resource requirements compared to relying solely on deep learning models. Notably, the VGG-16 + SVM model achieved an outstanding accuracy of 97.88% on a dataset preprocessed with ESRGAN and white balance adjustments, with a precision of 97.9%, a recall of 97.8%, and an F1 score of 0.978. Through a comprehensive comparative study, we demonstrate that the proposed framework, utilizing VGG-16 for feature extraction, SVM for classification, and preprocessed images with ESRGAN and white balance adjustments, achieves superior performance in plant species identification from UAV imagery.

Artificial Intelligence in Audiology: A Scoping Review of Current Applications and Future Directions

The integration of artificial intelligence (AI) into medical disciplines is rapidly transforming healthcare delivery, with audiology being no exception. By synthesizing the existing literature, this review seeks to inform clinicians, researchers, and policymakers about the potential and challenges of integrating AI into audiological practice. The PubMed, Cochrane, and Google Scholar databases were searched for articles published in English from 1990 to 2024 with the following query: “(audiology) AND (“artificial intelligence” OR “machine learning” OR “deep learning”)”. The PRISMA extension for scoping reviews (PRISMA-ScR) was followed. The database research yielded 1359 results, and the selection process led to the inclusion of 104 manuscripts. The integration of AI in audiology has evolved significantly over the succeeding decades, with 87.5% of manuscripts published in the last 4 years. Most types of AI were consistently used for specific purposes, such as logistic regression and other statistical machine learning tools (e.g., support vector machine, multilayer perceptron, random forest, deep belief network, decision tree, k-nearest neighbor, or LASSO) for automated audiometry and clinical predictions; convolutional neural networks for radiological image analysis; and large language models for automatic generation of diagnostic reports. Despite the advances in AI technologies, different ethical and professional challenges are still present, underscoring the need for larger, more diverse data collection and bioethics studies in the field of audiology.

Research on gut microbiota characteristics of PBC patients at different ALBI grades based on machine learning

BackgroundThe Albumin-Bilirubin (ALBI) score and grade are widely used to stratify patients with primary biliary cholangitis (PBC) into different disease statuses and risk levels. Recent studies have increasingly highlighted the role of gut microbiota in autoimmune liver diseases. This study aimed to investigate the differences in gut microbiota among PBC patients with varying ALBI grades.MethodsClinical data and stool samples were collected from outpatient and inpatient PBC patients between 2019 and 2022. Gut microbiota profiles were obtained using 16S rDNA sequencing of stool samples. We analyzed alpha diversity, beta diversity, LEfSe analysis and pathway function prediction. Additionally, various machine learning methods—including random forest (RF), lasso, gradient boosting machine (GBM) and support vector machine (SVM)—were employed to identify key features and to build and validate predictive models using bootstrap techniques.ResultsClinical characteristics of ALBI grade 1 patients were comparatively better than those of ALBI grade 2 and 3 patients, including multiple laboratory indices. Gut microbiota analysis revealed that species richness and balance were higher in ALBI grade 1 patients. Both the comparison of the most abundant genera and the linear discriminant analysis (LDA) in LEfSe demonstrated that Lachnospira had a higher abundance and better discriminative ability in ALBI grade 1. Pathway function prediction indicated that sulfur metabolism was upregulated in higher ALBI grades. Furthermore, RF identified 10 specific genera, which were then used to build and validate models for discriminating PBC patients according to their ALBI grades. All three models, developed using different machine learning methods, demonstrated good discrimination ability (mean AUC 0.75–0.80).ConclusionThis study highlights significant differences in gut microbiota profiles among PBC patients with different ALBI grades. The increased abundance of Lachnospira and upregulation of sulfur metabolism pathways are notable in patients with lower ALBI grades. The machine learning models developed based on gut microbiota features offer promising tools for discriminating between PBC patients with varying disease severities, which could enhance the precision of treatment strategies.

Machine Learning for Prediction of Resistance Scores in Wheat (Triticum aestivum L.)

ABSTRACTMachine learning methods were shown to improve the prediction accuracies of genomic prediction of resistance scores compared to methods like RR‐BLUP, which were originally designed for metric rather than ordinal response values. We conducted a cross‐validation study with 361 wheat genotypes evaluated for five fungal diseases. Our objective was to compare the prediction accuracy and the ability to identify the most resistant genotypes of 19 genomic prediction approaches. Each approach consisted of a different combination of prediction method (RR‐BLUP, an alternative method with heterogeneous marker variances, Bayesian generalized linear regression with an ordinal response, support vector machine, gradient boosting machine and random forest), predictor (single SNP markers, LD‐based haplotype blocks, 250 variables generated with an autoencoder and SNPs identified with incremental feature selection) and response value (untransformed and logit‐transformed resistance scores). In our dataset, RR‐BLUP was consistently among the methods with the largest prediction accuracies and the best abilities to identify resistant genotypes in four of five investigated traits. However, in P. triticina, using gradient boosting machine and random forest instead of RR‐BLUP increased the prediction accuracy from 0.64 to 0.71, indicating that machine learning methods may have an advantage over linear models in genomic prediction. We also found that even though there was a positive correlation between the prediction accuracy and Cohen's , a measure to judge how well the most resistant genotypes can be identified, the correlation is not perfect and a large value for the prediction accuracy does not necessarily translate into an equally large value.

Research on Climactic Chapter Recognition of a Chinese Long Novel Based on Plot Description

Many readers continue to pursue Chinese long novels in the past several decades because of diverse characters and fascinating plots. The climactic chapter is an important part of a Chinese long novel, where the key conflict develops to the extreme point. However, how to quickly and accurately recognize the climactic chapter remains a common problem for many readers in their reading choices. This paper conducts research on recognizing the climactic chapter of a Chinese long novel by accurately describing its plot. The proposed method consists of two parts; one is the extraction of key elements, such as viewpoint paragraphs, non-viewpoint paragraphs, chapter keywords, major characters etc. The other part is the climactic chapter recognition, which applies the Bidirectional Gate Recurrent Unit (BiGRU) model and the multi-head attention to recognize the climactic chapter, on the basis of the chapter plot description matrix. Comparative experiments on the corpus named The Condor Trilogy show that the proposed method in this paper has a better recognition performance compared with the existing models, such as Naive Bayesian (NB), Support Vector Machine (SVM), Roberta-large, and the Bidirectional Long-Short Term Memory (BiLSTM) network. Ablation experiments validated the effectiveness of primary components in the proposed method.

Intra- and inter-network connectivity abnormalities associated with surgical outcomes in degenerative cervical myelopathy patients: a resting-state fMRI study

Resting-state functional MRI (fMRI) has revealed functional changes at the cortical level in degenerative cervical myelopathy (DCM) patients. The aim of this study was to systematically integrate static and dynamic functional connectivity (FC) to unveil abnormalities of functional networks of DCM patients and to analyze the prognostic value of these abnormalities for patients using resting-state fMRI. In this study, we collected clinical data and fMRI data from 44 DCM patients and 39 healthy controls (HC). Independent component analysis (ICA) was performed to investigate the group differences of intra-network FC. Subsequently, both static and dynamic FC were calculated to investigate the inter-network FC alterations in DCM patients. k-means clustering was conducted to assess temporal properties for comparison between groups. Finally, the support vector machine (SVM) approach was performed to predict the prognosis of DCM patients based on static FC, dynamic FC, and fusion of these two metrics. Relative to HC, DCM patients exhibited lower intra-network FC and higher inter-network FC. DCM patients spent more time than HC in the state in which both patients and HC were characterized by strong inter-network FC. Both static and dynamic FC could successfully classify DCM patients with different surgical outcomes. The classification accuracy further improved after fusing the dynamic and static FC for model training. In conclusion, our findings provide valuable insights into the brain mechanisms underlying DCM neuropathology on the network level.

Improved Bacterial Foraging Optimization Algorithm with Machine Learning-Driven Short-Term Electricity Load Forecasting: A Case Study in Peninsular Malaysia

Accurate electricity demand forecasting is crucial for ensuring the sustainability and reliability of power systems. Least square support vector machines (LSSVM) are well suited to handle complex non-linear power load series. However, the less optimal regularization parameter and the Gaussian kernel function in the LSSVM model have contributed to flawed forecasting accuracy and random generalization ability. Thus, these parameters of LSSVM need to be chosen appropriately using intelligent optimization algorithms. This study proposes a new hybrid model based on the LSSVM optimized by the improved bacterial foraging optimization algorithm (IBFOA) for forecasting the short-term daily electricity load in Peninsular Malaysia. The IBFOA based on the sine cosine equation addresses the limitations of fixed chemotaxis constants in the original bacterial foraging optimization algorithm (BFOA), enhancing its exploration and exploitation capabilities. Finally, the load forecasting model based on LSSVM-IBFOA is constructed using mean absolute percentage error (MAPE) as the objective function. The comparative analysis demonstrates the model, achieving the highest determination coefficient (R2) of 0.9880 and significantly reducing the average MAPE value by 28.36%, 27.72%, and 5.47% compared to the deep neural network (DNN), LSSVM, and LSSVM-BFOA, respectively. Additionally, IBFOA exhibits faster convergence times compared to BFOA, highlighting the practicality of LSSVM-IBFOA for short-term load forecasting.

Intelligent Inversion Analysis of Surrounding Rock Parameters and Deformation Characteristics of a Water Diversion Surge Shaft

The water diversion surge shaft is vital for a hydropower station. However, the complex geological properties of the surrounding rock make it challenging to obtain its mechanical parameters. A method combining particle swarm optimization (PSO) and support vector machine (SVM) algorithms is proposed for estimating these parameters. According to the engineering geological background and support scheme, a three-dimensional model of the water diversion surge shaft is established by FLAC3D. An orthogonal test is designed to verify the accuracy of the numerical model. Then, the surrounding rock mechanical parameter database is established. The PSO-SVM intelligent inversion algorithm is used to invert the optimal values of the mechanical parameters of the surrounding rock. The support for excavating the next layer depends on the mechanical parameters of the current rock layer. An optimized design scheme is then compared and analyzed with the original support scheme by considering deformation and plastic characteristics. The research results demonstrate that the PSO-SVM intelligent inversion algorithm can effectively improve the accuracy and efficiency of the inversion of rock mechanical parameters. Under the influence of excavation, the surrounding rock in the plastic zone mainly fails in shear, with maximum deformation occurring in the middle and lower parts of the excavation area. The maximum deformation of the surrounding rock under support with long anchor cables is 0.6 cm less than that of support without long anchor cables and 4.07 cm less than that of support without an anchor. In the direction of the maximum and minimum principal stress, the maximum depth of the plastic zone under the support with long anchor cables is 1.3 m to 2.6 m less than that of the support without long anchor cables and the support without an anchor. Compared with the support without long anchor cables and support without an anchor, the support with long anchor cables can effectively control the deformation of the surrounding rock and limit the development of the plastic zone.

Spatiotemporal trends and drivers of forest cover change in Metekel Zone forest areas, Northwest Ethiopia.

The spatiotemporal dynamics of forest cover are essential for understanding the patterns and processes of forest change over time and space. This research focused on the spatiotemporal trends and drivers of forest cover change in the Metekel Zone of Northwest Ethiopia. Landsat 5, Landsat 7, and Landsat 8 imagery, covering the period from 1986 to 2019, were used for land use/cover classification. Land use/cover classification was performed using random forest (RF) and support vector machine (SVM) algorithms in the Google Earth Engine (GEE) platform, with training samples obtained through visual image interpretation. Spectral indices, such as the normalized difference vegetation index, soil-adjusted vegetation index, leaf area index, and normalized difference water index, were analyzed to examine forest cover dynamics over time. In addition, key informant interviews (KIIs) and focus group discussions (FGDs) were conducted. Findings revealed that forest cover decreased significantly from 51.37% in 1986 to 17.20% in 2019, driven largely by human activities such as agricultural expansion, increased demand for firewood, and urban expansion. Findings from spectral indices further corroborated the finding that forest cover in the study region (mainly in the southwestern part) substantially decreased from 1986 to 2019. Concerning forest depletion, the lack of local community awareness has become a key challenge. This problem is attributed to communities prioritizing immediate needs such as fuel and land for agriculture over long-term forest conservation. To combat ongoing deforestation, the Metekel Zone Administration, in collaboration with the land administration office and other stakeholders, revisited and strengthened existing forest policies and control systems. It is also suggested that community awareness, chiefly among youth, should be enhanced through the strategic expansion of formal and nonformal educational initiatives, which empower the youth as agents of change and promote the dissemination of knowledge throughout the community.

Prediction of building HVAC energy consumption based on least squares support vector machines

Prediction of building HVAC energy consumption based on least squares support vector machines

Application of CT-based foundational AI and radiomics models for prediction of outcome for non-small-cell lung cancer patients treated on the NRG/RTOG 0617 clinical trial

Abstract Objectives To apply CT-based foundational artificial intelligence (AI) and radiomics models for predicting overall survival (OS) for patients with locally advanced non-small cell lung cancer (NSCLC). Methods Data for 449 patients retrospectively treated on the NRG Oncology/Radiation Therapy Oncology Group (RTOG) 0617 clinical trial were analyzed. Foundational AI, radiomics and clinical features were evaluated using univariate cox regression and correlational analyses to determine independent predictors of survival. Several models were fit using these predictors and model performance was evaluated using nested cross-validation and unseen independent test datasets via area under receiver-operator-characteristic curves, AUC’s. Results For all patients, the combined foundational AI and clinical models achieved AUCs of 0.67 for the Random Forest (RF) model. The combined radiomics and clinical models achieved RF AUCs of 0.66. In the low-dose arm, foundational AI alone achieved AUC of 0.67, while AUC for the ensemble radiomics and clinical models was 0.65 for the support vector machine (SVM). In the high-dose arm, AUC values were 0.67 for combined radiomics and clinical models and 0.66 for the foundational AI model. Conclusions This study demonstrated encouraging results for application of foundational AI and radiomics models for prediction of outcomes. More research is warranted to understand the value of ensemble models toward improving performance via complementary information. Advances in knowledge Using foundational AI and radiomics-based models we were able to identify significant signatures of outcomes for NSCLC patients retrospectively treated on a national cooperative group clinical trial. Associated models will be important for application toward prospective patients.

Feature Selection and Machine Learning Approaches in Prediction of Current E-Cigarette Use Among U.S. Adults in 2022

Feature selection is essentially the process of picking informative and relevant features from a larger collection of features. Few studies have focused on predictors for current e-cigarette use among U.S. adults using feature selection and machine learning (ML) approaches. This study aimed to perform feature selection and develop ML approaches in prediction of current e-cigarette use using the 2022 Health Information National Trends Survey (HINTS 6). The Boruta algorithm and the least absolute shrinkage and selection operator (LASSO) were used to perform feature selection of 71 variables. The random oversampling example (ROSE) method was utilized to deal with imbalance data. Five ML tools including support vector machines (SVMs), logistic regression (LR), random forest (RF), gradient boosting machine (GBM), and extreme gradient boosting (XGBoost) were applied to develop ML models. The overall prevalence of current e-cigarette use was 4.3%. Using the overlapped 15 variables selected by Boruta and LASSO, the RF algorithm provided the best classifier with an accuracy of 0.992, sensitivity of 0.985, F1 score of 0.991, and AUC of 0.999. Weighted logistic regression further confirmed that age, education level, smoking status, belief in the harm of e-cigarette use, binge drinking, belief in alcohol increasing cancer, and the Patient Health Questionnaire-4 (PHQ4) score were associated with e-cigarette use. This study confirmed the strength of ML techniques in survey data, and the findings will guide inquiry into behaviors and mentalities of substance users.

Discrete Fourier Transform in Unmasking Deepfake Images: A Comparative Study of StyleGAN Creations

This study proposes a novel forgery detection method based on the analysis of frequency components of images using the Discrete Fourier Transform (DFT). In recent years, face manipulation technologies, particularly Generative Adversarial Networks (GANs), have advanced to such an extent that their misuse, such as creating deepfakes indistinguishable to human observers, has become a significant societal concern. We reviewed two GAN architectures, StyleGAN and StyleGAN2, generating synthetic faces that were compared with real faces from the FFHQ and CelebA-HQ datasets. The key results demonstrate classification accuracies above 99%, with F1 scores of 99.94% for Support Vector Machines and 97.21% for Random Forest classifiers. These findings underline the fact that performing frequency analysis presents a superior approach to deepfake detection compared to traditional spatial detection methods. It provides insight into subtle manipulation cues in digital images and offers a scalable way to enhance security protocols amid rising digital impersonation threats.