Principal Component Analysis Support Vector Machine Research Articles

Hybrid deep models for parallel feature extraction and enhanced emotion state classification

Emotions play a vital role in recognizing a person’s thoughts and vary significantly with stress levels. Emotion and stress classification have gained considerable attention in robotics and artificial intelligence applications. While numerous methods based on machine learning techniques provide average classification performance, recent deep learning approaches offer enhanced results. This research presents a hybrid deep learning model that extracts features using AlexNet and DenseNet models, followed by feature fusion and dimensionality reduction via Principal Component Analysis (PCA). The reduced features are then classified using a multi-class Support Vector Machine (SVM) to categorize different types of emotions. The proposed model was evaluated using the DEAP and EEG Brainwave datasets, both well-suited for emotion analysis due to their comprehensive EEG signal recordings and diverse emotional stimuli. The DEAP dataset includes EEG signals from 32 participants who watched 40 one-minute music videos, while the EEG Brainwave dataset categorizes emotions into positive, negative, and neutral based on EEG recordings from participants exposed to six different film clips. The proposed model achieved an accuracy of 95.54% and 97.26% for valence and arousal categories in the DEAP dataset, respectively, and 98.42% for the EEG Brainwave dataset. These results significantly outperform existing methods, demonstrating the model’s superior performance in terms of precision, recall, F1-score, specificity, and Mathew correlation coefficient. The integration of AlexNet and DenseNet, combined with PCA and multi-class SVM, makes this approach particularly effective for capturing the intricate patterns in EEG data, highlighting its potential for applications in human-computer interaction and mental health monitoring, marking a significant advancement over traditional methods.

Machine Learning Algorithms for Intelligent Decision Recognition and Quantification of Cr(III) in Chromium Speciation.

Cr(III) is a common oxidation state of chromium, and its presence in the environment can occur naturally or as a result of human activities, such as industrial processes, mining, and waste disposal. This article explores the application of machine learning algorithms for the intelligent decision recognition and quantification of Cr(III) in chromium speciation. Three different machine learning models, namely, the Decision Tree (DT) model, the PCA-SVM (Principal Component Analysis-Support Vector Machine) model, and the LDA (Linear Discriminant Analysis) model, were employed and evaluated for accurate and efficient classification of chromium concentrations based on their fluorescence responses. Furthermore, stepwise multiple linear regression analysis was utilized to achieve a more precise quantification of trivalent chromium concentrations through fluorescence visualization. The results demonstrate the potential of machine learning algorithms in accurately detecting and quantifying Cr(III) in chromium speciation with implications for environmental and industrial applications in chromium detection and quantification. The findings from this research pave the way for further exploration and implementation of these models in real-world scenarios, offering valuable insights into various environmental and industrial contexts.

A Method of Quantitative Detection of Fatigue Crack Depth in Bottom Rails by Ultrasonic Guided Waves Based on PCA-SVM

Abstract Ultrasonic guided wave is widely used to detect cracks in rail because of its long propagation distance and small attenuation. To effectively detect the fatigue crack in rail bottom through ultrasonic guided wave, an improved principal component analysis-support vector machine (PCA-SVM) intelligent algorithm based on grid search (GS) is proposed to detect the fatigue crack at different depths of rail bottom. The finite element method is used to establish the model of ultrasonic guided wave at different depths of the rail bottom, and the simulation model is compared with the experimental data to determine the effectiveness of the simulation model. Five main component features of the fatigue cracks at different depths are extracted by PCA. The GS method is used to optimize the penalty factor c and kernel function parameter g in the SVM, and the optimized SVM model is selected to identify the rail fatigue crack at different depths. The combination of theoretical simulation and experimental results shows that the accuracy of the training set and the test set of the improved PCA-SVM intelligent algorithm based on the GS method can reach 99.79 % and 99.73 %, respectively, which provides a basis and method for the detection of the fatigue crack depth of the rail bottom.

A Handheld Visible Resonance Raman Analyzer Used in Intraoperative Detection of Human Glioma.

There is still a lack of reliable intraoperative tools for glioma diagnosis and to guide the maximal safe resection of glioma. We report continuing work on the optical biopsy method to detect glioma grades and assess glioma boundaries intraoperatively using the VRR-LRRTM Raman analyzer, which is based on the visible resonance Raman spectroscopy (VRR) technique. A total of 2220 VRR spectra were collected during surgeries from 63 unprocessed fresh glioma tissues using the VRR-LRRTM Raman analyzer. After the VRR spectral analysis, we found differences in the native molecules in the fingerprint region and in the high-wavenumber region, and differences between normal (control) and different grades of glioma tissues. A principal component analysis-support vector machine (PCA-SVM) machine learning method was used to distinguish glioma tissues from normal tissues and different glioma grades. The accuracy in identifying glioma from normal tissue was over 80%, compared with the gold standard of histopathology reports of glioma. The VRR-LRRTM Raman analyzer may be a new label-free, real-time optical molecular pathology tool aiding in the intraoperative detection of glioma and identification of tumor boundaries, thus helping to guide maximal safe glioma removal and adjacent healthy tissue preservation.

Label-free detection and discrimination of respiratory pathogens based on electrochemical synthesis of biomaterials-mediated plasmonic composites and machine learning analysis

Seasonal outbreaks of respiratory viral infections remain a global concern, with increasing morbidity and mortality rates recorded annually. Timely and false responses contribute to the widespread of respiratory pathogenic diseases owing to similar symptoms at an early stage and subclinical infection. The prevention of emerging novel viruses and variants is also a big challenge. Reliable point-of-care diagnostic assays for early infection diagnosis play a critical role in the response to threats of epidemics or pandemics. We developed a facile method for specifically identifying different viruses based on surface-enhanced Raman spectroscopy (SERS) with pathogen-mediated composite materials on Au nanodimple electrodes and machine learning (ML) analyses. Virus particles were trapped in three-dimensional plasmonic concave spaces of the electrode via electrokinetic preconcentration, and Au films were simultaneously electrodeposited, leading to the acquisition of intense and in-situ SERS signals from the Au–virus composites for ultrasensitive SERS detection. The method was useful for rapid detection analysis (<15 min), and the ML analysis for specific identification of eight virus species, including human influenza A viruses (i.e., H1N1 and H3N2 strains), human rhinovirus, and human coronavirus, was conducted. The highly accurate classification was achieved using the principal component analysis-support vector machine (98.9%) and convolutional neural network (93.5%) models. This ML-associated SERS technique demonstrated high feasibility for direct multiplex detection of different virus species for on-site applications.

Rapid Identification of Foodborne Pathogens in Limited Resources Settings Using a Handheld Raman Spectroscopy Device

Rapid and precise methods to detect pathogens are paramount in ensuring food safety and selecting appropriate disinfection treatments. Raman spectrometry is a promising technology being investigated for detecting pathogens and achieving rapid, culture-free, and label-free methods. Nonetheless, previous Raman techniques require additional steps, including the preparation of slides that could introduce significant variability. In this study, we investigated the capability of a Raman handheld device for rapid identification of monocultures of Listeria monocytogenes, Salmonella Typhimurium, Escherichia coli O157:H7, and Staphylococcus aureus, and the combination of co-cultures in BHI broth suspension by utilising principal component analysis (PCA) and support vector machine (SVM) classification of Raman spectra. The detection method accurately identified monocultures (0.93 ± 0.20), achieving good discrimination after 24 h of bacterial growth. However, the PCA–SVM system was less accurate for classifying co-cultures (0.67 ± 0.35). These results show that this method requires an isolation step followed by biomass enrichment (&gt;8 log10 CFU/mL) for accurate identification. The advantage of this technology is its simplicity and low-cost preparation, achieving high accuracy in monocultures in a shorter time than conventional culture-dependent methods.

Grouped-sampling technique to deal with unbalance in Raman spectral data modeling

Due to limitations in disease prevalence and hospital specificity, spectral data are often collected with unbalanced sample size. To solve this problem, a new sampling method – grouped-sampling was proposed in this research, which is shown to be effective for unbalanced data. It avoids over-fitting of over-sampling and overcomes under-sampling utilization of under-sampling. In this study, we applied grouped-sampling to two unbalanced datasets where the sample proportions are 199:40 and 75:225. And then verified from two classic models: PCA-SVM (Principal Component Analysis-Support Vector Machine) and the deep learning algorithm GoogLeNet. The accuracy of these two datasets were 85.11% and 96.15% in PCA-SVM and 85.10% and 84.61% on GoogLeNet. Also, the F1-score were evaluated to measure the classification balance of sampling method, and result shows that F1-score of grouped-sampling is always the highest compared to over-sampling and under-sampling. In summary, compared to traditional sampling methods, grouped-sampling performs better on prediction for classes with smaller sample size, which means grouped-sampling can improve the balance of classification results and the potential of practical application. Therefore, we develop a group sampling method that distinguishes between under- and over-sampling, which greatly improves the accuracy and balance of predictions for unbalanced samples.

Detection of Ofloxacin and Norfloxacin in Duck Meat Using Surface-Enhanced Raman Spectroscopy (SERS) Coupled with Multivariate Analysis

Excess use of ofloxacin (OFL) and norfloxacin (NOR) in duck breeding is a significant food safety issue due to its residues. Therefore, a surface-enhanced Raman spectroscopy (SERS) method was developed to detect these analytes in duck meat to protect consumer health. The SERS conditions for OFL and NOR including the adsorption time, the volume ratio of gold nanoparticles to NaCl solution, and the volume of enhancement solution, were optimized by single factor experiments, and their values were 6 min, 3:1, and 20 μL, respectively. Furthermore, a total of 396 samples were used to establish a principal component analysis–support vector machine (PCA-SVM) model. The performance was evaluated using three pretreatment methods: adaptive iterative–penalty least squares (air-PLS) and standard normal variate (SNV), air-PLS and first derivative coupled with SNV, and air-PLS and second derivative coupled with SNV. Air-PLS and second derivative coupled with SNV was selected to be the optimal pretreatment. The sensitivity and specificity of PCA-SVM for the classification of the meat samples were 85 to 100% and 95 to 100% with an accuracy of 93%. The results show that SERS was an effective and rapid approach for the identification of OFL and NOR in duck meat.

New Insights into the Multivariate Analysis of SER Spectra Collected on Blood Samples for Prostate Cancer Detection: Towards a Better Understanding of the Role Played by Different Biomolecules on Cancer Screening: A Preliminary Study.

Simple SummaryIn recent years, research on biofluids using Raman and SERS has expanded dramatically, indicating the enormous promise of this technology as a high-throughput tool for identifying cancer and other disorders. In the investigations thus far, researchers have concentrated on a specific illness or condition, but the techniques employed to acquire experimental spectra prevent direct comparison of the data. This necessitates comparative research of a variety of diseases and an increase in scientific cooperation to standardize experimental conditions. In our study, positive results were reached by applying a combined SERS multivariate analysis (MVA) to the urgent problem of prostate cancer diagnosis that was directly linked to real-world settings in healthcare. Moreover, in comparison to the prostate-specific antigen (PSA) test, which has a high sensitivity but limited specificity, our combined SERS-MVA method has greater specificity, which may assist in preventing the overtreatment of patients.It is possible to obtain diagnostically relevant data on the changes in biochemical elements brought on by cancer via the use of multivariate analysis of vibrational spectra recorded on biological fluids. Prostate cancer and control groups included in this research generated almost similar SERS spectra, which means that the values of peak intensities present in SERS spectra can only give unspecific and limited information for distinguishing between the two groups. Our diagnostic algorithm for prostate cancer (PCa) differentiation was built using principal component analysis and linear discriminant analysis (PCA-LDA) analysis of spectral data, which has been widely used in spectral data management in many studies and has shown promising results so far. In order to fully utilize the entire SERS spectrum and automatically determine the most meaningful spectral features that can be used to differentiate PCa from healthy patients, we perform a multivariate analysis on both the entire and specific spectral intervals. Using the PCA-LDA model, the prostate cancer and control groups are clearly distinguished in our investigation. The separability of the following two data sets is also evaluated using two alternative discrimination techniques: principal least squares discriminant analysis (PLS-DA) and principal component analysis—support vector machine (PCA-SVM).

A four-step protection strategy to improve stability of high voltage transmission lines

Opportune detection of transient faults and restoration of the faulty phase in the least possible time plays a key role in improving the stability of high voltage transmission lines (HVTLs). This paper presents a novel four-step strategy for the protection of HVTLs so that the stability and reliability of power systems are enhanced. Firstly, the cumulative sum (CUSUM) algorithm is used to quickly detect the transient and determine the faulty phase. In the second step, the settings of the static var compensator (SVC) are changed to reduce the reclosing dead time as much as possible. Afterward, the transient or permanent type of the fault is identified based on the proposed principal component analysis-support vector machine (PCA-SVM) algorithm. In the fourth step, if a transient fault is recognized, the successful single-phase reclosing time is identified by the PCA-SVM and the reclosing command is issued to high voltage circuit breakers (HVCBs). Simulation results obtained by using EMTP-RV and MATLAB for a sample 400 kV power system illustrate that the proposed protection strategy can increase the reliability of the power system via reduction of the dead time, correct detection of transient faults from permanent ones, and accurate detection of the reclosing time.

Hybrid Machine Learning Model for Face Recognition Using SVM

Face recognition systems have enhanced human-computer interactions in the last ten years. However, the literature reveals that current techniques used for identifying or verifying faces are not immune to limitations. Principal Component Analysis-Support Vector Machine (PCA-SVM) and Principal Component Analysis-Artificial Neural Network (PCA-ANN) are among the relatively recent and powerful face analysis techniques. Compared to PCA-ANN, PCA-SVM has demonstrated generalization capabilities in many tasks, including the ability to recognize objects with small or large data samples. Apart from requiring a minimal number of parameters in face detection, PCA-SVM minimizes generalization errors and avoids overfitting problems better than PCA-ANN. PCA-SVM, however, is ineffective and inefficient in detecting human faces in cases in which there is poor lighting, long hair, or items covering the subject's face. This study proposes a novel PCA-SVM-based model to overcome the recognition problem of PCA-ANN and enhance face detection. The experimental results indicate that the proposed model provides a better face recognition outcome than PCA-SVM.

An efficient P300 detection algorithm based on Kernel Principal Component Analysis-Support Vector Machine

Human-machine interaction using brain signals has been made possible by the advent of a technology popularly known as a brain-computer interface (BCI). P300 is the most studied event related potentials (ERP) and is used in many BCI systems. The existing multi-trial P300 detection methods suffer drawbacks of being time-consuming and computationally complex, whereas, the existing single-trial methods are apt at achieving only moderate accuracy levels. In this paper, a novel approach to achieve a high level of accuracy for a single trial P300 signal detection amidst noise and artifacts. In this method, features were obtained from wavelet coefficients; subsequently, feature dimensions were reduced, thereby enhancing the speed of classification along with a manifold improvement in the accuracy of P300 signal classification. An accuracy of 98.53% was achieved for Subject S1 and 99.25% for Subject S2 using the proposed method. A high level of accuracy was obtained, compared to many existing techniques. Moreover, the speed of classification improved with the use of reduced feature dimensions.

Label-free detection of living cervical cells based on microfluidic device with terahertz spectroscopy.

Early diagnosis of cervical cancer is essential for a good prognosis. Terahertz wave detection technology is a nondestructive and label-free physical detection technology, which can detect and monitor the cancer cells in real time, especially for patients with deep or inaccessible tumors. In this study, a single-cell-layer microfluidic device was developed. After replacing the optical clearing agent, the characteristics of H8, HeLa and SiHa cell lines in adherent and suspended states were detected. Additionally, the absorption increased with increasing cell density. For the mixed suspension cell samples, principal component analysis-support vector machine method was used to identify benign and malignant cell component. After living cells formaldehyde, changes in cell membrane permeability were evaluated to identify the cell survival status (i.e., dead or living) based on terahertz spectroscopy amplitude differences. Therefore, extending the terahertz spectrum detection to the molecular level can characterize the life essence of cells and tissues.

Validating Multivariate Classification Algorithms in Raman Spectroscopy-Based Osteosarcoma Cellular Analysis

Raman microspectroscopy has been widely demonstrated as an ideal analytical tool for preclinical drug development and clinical applications. However, it is still not easy to accurately identify the subtle spectral variations in different biological samples, which requires a feasible combination between novel spectra collection instrumentation and effective data mining algorithms. In this study, three distinct multivariate classification approaches, which were principal component analysis-linear discriminant analysis (PCA-LDA), support vector machine (SVM), and principal component analysis-support vector machine (PCA-SVM), were validated and compared for obtaining reliable and chemically significant results from the analysis of bio-spectral data. Their performances were evaluated by classifying the spectral characteristics of osteosarcoma cells treated with N-[N-(3,5-difluorophenacetyl)-L-alanyl]-S-phenylglycine t-butyl ester (DAPT) from untreated cells. Based on the discriminated spectral variations, the results indicate that PCA combined with the radial basis function (RBF) kernel SVM model achieved the highest classification accuracy. In general, this study confirms that PCA-SVM algorithm improves the automatic processing accuracy and efficiency of micro-Raman spectroscopy, which may be adopted in further cell screening and analysis applications.

Raman spectroscopy and machine learning for the classification of breast cancers

Breast cancer is a major health threat for women. The drug responses associated with different breast cancer subtypes have obvious effects on therapeutic outcomes; therefore, the accurate classification of breast cancer subtypes is critical. Breast cancer subtype classification has recently been examined using various methods, and Raman spectroscopy has emerged as an effective technique that can be used for noninvasive breast cancer analysis. However, the accurate and rapid classification of breast cancer subtypes currently requires a great deal of effort and experience with the processing and analysis of Raman spectra data. Here, we adopted Raman spectroscopy and machine learning techniques to simplify and accelerate the process used to distinguish normal from breast cancer cells and classify breast cancer subtypes. Raman spectra were obtained from cultured breast cancer cell lines, and the data were analyzed by two machine learning algorithms: principal component analysis (PCA)–discriminant function analysis (DFA) and PCA–support vector machine (SVM). The accuracies with which these two algorithms were able to distinguish normal breast cells from breast cancer cells were both greater than 97%, and the accuracies of breast cancer subtype classification for both algorithms were both greater than 92%. Moreover, our results showed evidence to support the use of characteristic Raman spectral features as cancer cell biomarkers, such as the intensity of intrinsic Raman bands, which increased in cancer cells. Raman spectroscopy combined with machine learning techniques provides a rapid method for breast cancer analysis able to reveal differences in intracellular compositions and molecular structures among subtypes.

Longitudinal Crack Detection Approach Based on Principal Component Analysis and Support Vector Machine for Slab Continuous Casting

Longitudinal cracks are the typical surface defects of continuous casting slabs, resulting in additional processing or even casting interruption. Monitoring longitudinal crack defects is of great significance in stabilizing and improving the slab quality. Herein, a monitoring model is developed to recognize the longitudinal crack defect of continuous casting slabs using principal component analysis (PCA) and support vector machine (SVM). First, the typical characteristics of the temperature patterns corresponding to the longitudinal crack defect are extracted, including the normal casting temperature with small and large fluctuations, as well as the longitudinal crack temperature. Then, PCA is used to reduce the dimension of these characteristics for removing the redundancy and reducing the computational burden. Subsequently, an SVM is applied to identify normal and longitudinal crack temperature patterns for training the monitoring model PCA–SVM. The monitoring performance is verified by the test data, in which the training accuracy and test accuracy are 100% and 96%, respectively. It is worth mentioning that the model can successfully predict all the real longitudinal crack defects, showing excellent detection performance. The established model is expected to provide a theoretical basis and a reliable way for online monitoring of slab surface cracks.

Non-invasive SERS serum detection technology combined with multivariate statistical algorithm for simultaneous screening of cervical cancer and breast cancer.

Surface-enhanced Raman scattering (SERS), as a rapid, reliable and non-destructive spectral detection technology, has made a series of breakthrough achievements in screening and pre-diagnosis of various cancerous tumors. In this paper, high-performance gold nanoparticles/785 porous silicon photonic crystals (Au NPs/785 PSi PhCs) active SERS substrates were specially designed for serum testing, and realized highly sensitive detection of serum from healthy people, patients with cervical cancer and breast cancer. Based on the SERS spectra of the three groups of serum, the significant differences between the healthy group and cancer group at 1030cm-1 and 1051cm-1 were analyzed, and the similar but different serum SERS spectra of cervical cancer and breast cancer patients were compared. In addition, the spectral difference detected by SERS technology combined with a multivariate statistical algorithm was used to distinguish three kinds of serum. The serum SERS spectral sensitive bands were extracted by recursive weighted partial least squares (rPLS), and the three classification diagnosis models were established by combining orthogonal partial least squares discriminant analysis (OPLS-DA), linear discriminant analysis (LDA) and principal component analysis support vector machine (PCA-SVM) for synchronous classification and discrimination of the three groups of serum. The diagnostic results showed that the overall screening accuracy of three models were 93.28%, 97.77% and 94.78%, respectively. These above results confirmed that the Au NPs/785 PSi PhCs can realize super-sensitive detection of serum, and the established diagnostic model has great potential for pre-diagnosis and simultaneous screening of cervical cancer and breast cancer.

Spectral detection technology of vegetable oil: Spectral analysis of porphyrins and terpenoids

In this paper, the existence of porphyrins and terpenoids in different vegetable oils and their spectral characterization techniques are reported. The classification of pure vegetable oils was realised by principal component analysis - support vector machine (PCA-SVM) model. The absorption spectra, Raman spectra, fluorescence spectra and supercontinuum spectra of 8 kinds of pure vegetable oils were studied, and the effects of oil types and processing technology on spectral differences were analysed. The results showed that the fingerprint information of 4 kinds of spectral techniques mainly came from chlorophyll and β-carotene in vegetable oil. The extra virgin olive oil (EVOO) got by physical cold pressing technology had the most porphyrins and terpenoids content and the strongest activity. Therefore, the spectral characterization of porphyrins and terpenoids in vegetable oil can guide the regulation of the processing technology of vegetable oil and realise the qualitative and quantitative analysis of vegetable oil.

Raman Microspectral Study and Classification of the Pathological Evolution of Breast Cancer Using Both Principal Component Analysis-Linear Discriminant Analysis and Principal Component Analysis-Support Vector Machine

To facilitate the enhanced reliability of Raman-based tumor detection and analytical methodologies, an ex vivo Raman spectral investigation was conducted to identify distinct compositional information of healthy (H), ductal carcinoma in situ (DCIS), and invasive ductal carcinoma (IDC). Then, principal component analysis-linear discriminant analysis (PCA-LDA) and principal component analysis-support vector machine (PCA-SVM) models were constructed for distinguishing spectral features among different tissue groups. Spectral analysis highlighted differences in levels of unsaturated and saturated lipids, carotenoids, protein, and nucleic acid between healthy and cancerous tissue and variations in the levels of nucleic acid, protein, and phenylalanine between DCIS and IDC. Both classification models were principal component analysis-linear discriminant analysis to be extremely efficient on discriminating tissue pathological types with 99% accuracy for PCA-LDA and 100%, 100%, and 96.7% for PCA-SVM analysis based on linear kernel, polynomial kernel, and radial basis function (RBF), respectively, while PCA-SVM algorithm greatly simplified the complexity of calculation without sacrificing performance. The present study demonstrates that Raman spectroscopy combined with multivariate analysis technology has considerable potential for improving the efficiency and performance of breast cancer diagnosis.

Rapid and real-time detection of black tea fermentation quality by using an inexpensive data fusion system

Intelligent identification of black tea fermentation quality is becoming a bottleneck to industrial automation. This study presents at-line rapid detection of black tea fermentation quality at industrial scale based on low-cost micro-near-infrared spectroscopy (NIRS) and laboratory-made computer vision system (CVS). High-performance liquid chromatography and a spectrophotometer were used for determining the content of catechins and theaflavins, and the color of tea samples, respectively. Hierarchical cluster analysis combined with sensory evaluation was used to group samples through different fermentation degrees. A principal component analysis–support vector machine (SVM) model was developed to discriminate the black tea fermentation degree using color, spectral, and data fusion information; high accuracy (calibration = 95.89%, prediction = 89.19%) was achieved using mid-level data fusion. In addition, SVM model for theaflavins content prediction was established. The results indicated that the micro-NIRS combined with CVS proved a portable and low-cost tool for evaluating the black tea fermentation quality.