Discriminant Classification Model Research Articles

Computerized Principal Component–Canonical Discriminant Analysis Classification Model for Pollen Grain Identification Using Surface-Enhanced Raman Spectroscopy

Pollen identification and quantification are employed in various applications, and studies have been conducted to achieve precise automated pollen recognition to minimize the manual labor and subjectivity associated with human identification. The objective of the study was to evaluate the potential of surface-enhanced Raman spectroscopy (SERS) properties of pollen grain spectra, specifically within the spectral range of 582–1563 cm–1, using the convective assembly combined principal component–canonical discriminant analysis (PC-CDA) classification model. This research aimed to explore the feasibility of using this approach for the automated identification of pollen in the future. This approach facilitates the optimal arrangement of pollen grains and colloidal silver nanoprobes, thereby enhancing SERS spectra reproducibility using the convective assembly sampling method. Pollen grains can be characterized at a microscopic level without the need for purification-based analysis procedures. The findings demonstrate the possibility of rapid and accurate classification of pollen grains with the PC-CDA detection model with a 99.3% sensitivity, 97.6% specificity, and 98.5% accuracy. Research study indicates that the SERS properties of pollen spectra could serve as a promising method for automated pollen identification using the PC–CDA classification model.

Sex estimation using long bones in the largest burial site of the Copper Age: Linear discriminant analysis and random forest

Sex estimation of the individuals in a sample is fundamental for any bioarchaeological study to define a particular demographic assemblage or to classify isolated remains. Long bones are an excellent alternative for sex estimation when the most dimorphic anatomical parts are not preserved or are highly altered. Here we propose a set of discriminant functions and classification models to estimate the sex of prehistoric individuals using linear discriminant analysis and machine learning approaches. Different osteometric variables were taken from the humeri, ulnae, radii, femurs and tibias of a sample of 109 articulated skeletons buried in the collective tomb of Camino del Molino (Region of Murcia, SE-Spain), dated to the 3rd millennium BC. Sex was estimated based on standard anthropological methods and ancient DNA analysis of a control sample. Fifty-two discriminant functions with prediction thresholds higher than 0.8 on the ROC curve were obtained using independent (22) and combined variables (30). The best LDA models for sex prediction were those based on proximal epiphyseal widths or their combination with other variables, reaching values close to 0.98 on the ROC curve. The random forest-based model obtained an accuracy of 0.94 and confirmed the importance of epiphyseal widths in sex classification. This analysis is more comprehensive than univariate LDA, as it allows for ranking the importance of bones in sex discrimination and considers correlations between long bones rather than treating them as independent observations. In contrast, applying LDA to each bone makes it easier to predict the sex of other coeval collections that do not have such a complete sample. This work aims to overcome the scarcity of methods that can be applied to sex estimation of the large volume of isolated remains from Camino del Molino and for other Mediterranean skeletal series from the Late Prehistory with high biological affinity and that share similar environmental conditions.

SSC and pH prediction and maturity classification of grapes based on hyperspectral imaging

Soluble solids content (SSC) and pH of red globe grapes are crucial measures of quality. In this paper, we used hyperspectral imaging technology to achieve nondestructive detection and distribution visualization of SSC and pH of red globe grapes. First, the hyperspectral images of samples were collected. Then, CARS, SPA, GA, IRIV were used to extract feature variables from raw spectral (RAW) information. The PLSR prediction models of samples were developed. By comparing the different prediction models, RAW-IRIV-PLSR was selected as the optimal model. Finally, the SSC and pH of the samples were calculated to obtain a grayscale image and perform a pseudo-color transformation to visualize the distribution of SSC and pH. By studying the classification of the maturity of samples, it was concluded that the best discriminant classification model of maturity was RAW-IRIV-ELM. Hyperspectral also provided a new method for maturity stage classification of red globe grapes.

Ligand and structure-based discovery of phosphorus-containing compounds as potential metalloproteinase inhibitors

ABSTRACT In this study, a methodology is proposed, combining ligand- and structure-based virtual screening tools, for the identification of phosphorus-containing compounds as inhibitors of zinc metalloproteases. First, we use Dragon molecular descriptors to develop a Linear Discriminant Analysis classification model, which is widely validated according to the OECD principles. This model is simple, robust, stable and has good discriminating power. Furthermore, it has a defined applicability domain and it is used for virtual screening of the DrugBank database. Second, docking experiments are carried out on the identified compounds that showed good binding energies to the enzyme thermolysin. Considering the potential toxicity of phosphorus-containing compounds, their toxicological profile is evaluated according to Protox II. Of the five molecules evaluated, two show carcinogenic and mutagenic potential at small LD50, not recommended as drugs, while three of them are classified as non-toxic, and could constitute a starting point for the development of new vasoactive metalloprotease inhibitor drugs. According to molecular dynamics simulation, two of them show stable interactions with the active site maintaining coordination with the metal. A high agreement is evident between QSAR, docking and molecular dynamics results, demonstrating the potentialities of the combination of these tools.

Use of noisy labels as weak learners to identify incompletely ascertainable outcomes: A Feasibility study with opioid-induced respiratory depression

ObjectiveAssigning outcome labels to large observational data sets in a timely and accurate manner, particularly when outcomes are rare or not directly ascertainable, remains a significant challenge within biomedical informatics. We examined whether noisy labels generated from subject matter experts’ heuristics using heterogenous data types within a data programming paradigm could provide outcomes labels to a large, observational data set. We chose the clinical condition of opioid-induced respiratory depression for our use case because it is rare, has no administrative codes to easily identify the condition, and typically requires at least some unstructured text to ascertain its presence. Materials and methodsUsing de-identified electronic health records of 52,861 post-operative encounters, we applied a data programming paradigm (implemented in the Snorkel software) for the development of a machine learning classifier for opioid-induced respiratory depression. Our approach included subject matter experts creating 14 labeling functions that served as noisy labels for developing a probabilistic Generative model. We used probabilistic labels from the Generative model as outcome labels for training a Discriminative model on the source data. We evaluated performance of the Discriminative model with a hold-out test set of 599 independently-reviewed patient records. ResultsThe final Discriminative classification model achieved an accuracy of 0.977, an F1 score of 0.417, a sensitivity of 1.0, and an AUC of 0.988 in the hold-out test set with a prevalence of 0.83% (5/599). DiscussionAll of the confirmed Cases were identified by the classifier. For rare outcomes, this finding is encouraging because it reduces the number of manual reviews needed by excluding visits/patients with low probabilities. ConclusionApplication of a data programming paradigm with expert-informed labeling functions might have utility for phenotyping clinical phenomena that are not easily ascertainable from highly-structured data.

The Strategy of Discriminating False Comments on the Internet by Fusing Probabilistic Topic and Word Vector Models

With the acceleration of the social process of "Internet+", e-commerce has entered an era of rapid development. In response to the subjective judgments in current false information detection methods and the problem of low classification accuracy caused by the failure to extract semantic information hidden in text, this study proposes Latent Dirichlet Allocation -Word vector-Edge Convolutional Neural Network (LW-ECNN) Model. First, the probabilistic topic model is organically combined with the word vector model, and then the edge path structure is proposed to optimize the Convolutional Neural Network (CNN) discriminative classification model. Finally, the data-side optimization module is combined with the edge CNN classifier to form a web-based false comment discrimination model based on the probabilistic topic and word vector models. The results show that the probabilistic topic and word vector-based online false comment discrimination model has good results in detecting false comments, and can provide data reference for false comment detection by false detection-related departments, which is of great practical significance to assist the network environment cleaning work.

Discriminating milk storage time with mid-infrared spectra combined with machine learning

Classification models for rapid discrimination of milk storage time were developed. Buffalo raw milk (BuR) and bovine raw (BoR), pasteurised (BoP) and ultra-high temperature sterilised (BUHT) milk were analysed by mid-infrared (MIR) spectroscopy at different storage times. A total of 175 (44), 96 (24), 252 (64) and 383 (97) samples were used in the developed (validated) models of BuR, BoR, BoP and BUHT milk, respectively. Seven, two, five, and three spectral regions were selected for BuR, BoR, BoP and BUHT, respectively, to develop their models, which reflect changes in milk protein, fat, lactose and urea nitrogen. Accuracies of the optimised models for BuR, BoR, BoP and BUHT milk were 0.95, 1.00, 0.89, and 0.93, respectively, in the validation dataset. All optimal models were obtained by machine learning methods. This preliminarily study shows that MIR combined with machine learning can rapidly identify the storage time of these four types of milk.

A rapid and non-destructive identification for paper cup evidence based on shifted-excitation Raman difference spectroscopy and SOM clustering.

In order to achieve rapid, non-destructive, efficient, and accurate classification of paper cup samples, we propose a classification model that integrates shifted-excitation Raman difference spectroscopy (SERDS) with self-organizing map (SOM) and Bayesian optimization-support vector machine (BO-SVM). We collected differential Raman data from 52 paper cup samples using SERDS, with an excitation wavelength range of 784-785 nm, a laser power of 440 mW, an integration time of 10 s, and a spectral range spanning from 280 to 2700 cm-1 . Subsequently, principal component analysis (PCA) was applied to reduce the dimensionality of the data. The SOM clustering outcomes were utilized as the foundation for constructing the discriminant analysis (FDA) and BO-SVM classification models. The primary constituent of the paper cup samples was identified as cellulose, while additional fillers such as talc, calcium carbonate, and kaolin were also present. The SOM clustering categorized the samples into seven distinct groups. The FDA model achieved a classification accuracy of 92.3%, and the BO-SVM model reached a classification accuracy of 96.2%. The SOM clustering effectively discerned samples with different fillers, as evidenced by distinct peak numbers and shapes in the differential Raman spectra, thereby underscoring the practical significance of SOM clustering. In comparison with FDA, BO-SVM exhibited enhanced classification accuracy and exceptional performance in handling outliers and linearly inseparable data, indicating its superior generalization capabilities.

HSI combined with CNN model detection of heavy metal Cu stress levels in apple rootstocks

In recent years, the prolonged application of copper-based fungicides, ripening agents, and fertilizers in agricultural activities has resulted in copper (Cu) levels in the soil exceeding safe levels. Excess Cu has a negative impact on plant growth and crop yields. The practice of grafting horticultural crops onto appropriate rootstocks is frequently employed as a preventive measure against copper accumulation and its associated toxicity. Therefore, the selection of high-quality rootstocks that are not contaminated with heavy metal Cu is important for the growth of grafted crops and the breeding of grafted crops. This study proposed a technique based on convolutional neural network (CNN) and hyperspectral imaging (HSI) for the rapid, non-destructive, and accurate identification of copper contamination levels in apple rootstocks. The effect of different heavy metal Cu contamination levels on leaves was obtained by measuring chlorophyll content in leaves using the ultraviolet spectrophotometer method. A hybrid strategy for spectral variable selection was utilized, involving competitive adaptive reweighted sampling and successive projection algorithms. A traditional partial least squares discriminant analysis linear classification model, and least squares support vector machine, random forest, and extreme learning machine nonlinear models were developed. The results showed that the accuracy of the CNN model and Macro-F1 were 99.6% and 99.2%, respectively, which were better than the traditional linear and nonlinear models. Principal component analysis and t-distributed stochastic neighbor embedding were used for visual analysis to investigate the clustering patterns among sample classes. The results indicated that the combination of CNN and HSI techniques has great potential for classifying copper contamination levels in apple rootstocks.

Grinding burn classification with surface Barkhausen noise measurements

Industrial Barkhausen noise (BN) measurements are commonly utilized for final quality control after machining operations such as grinding to point out grinding burns. Grinding burns might compromise the final use and fatigue lifetime of the ground component. The industrial BN method itself is based on a pre-determined threshold value of the BN rootmean-square value (RMS). Elevated RMS values indicate detrimental changes in the component. Usually, the evaluation of grinding burn severity is not carried out. In this study, real ground cylindrical samples were collected that were rejected based on an industrial quality control with a BN unit. A more detailed BN analysis was carried out for 41 individual grinding burn locations followed by X-ray diffraction based residual stress (RS) surface measurements and residual stress and diffraction peak full-width-at-halfmaximum (FWHM) depth profiles. K-means clustering was applied to profiles to label the data points related to grinding burns of different severity. Three classes of grinding burns were identified and verified by micrographs and hardness. A linear discriminant classification model was then identified between the surface BN measurement features and labeled data points. The classification results were reasonable with about 80 % classification accuracy at worst. They showed that the classes identified can be detected with the surface BN measurements. Thus, the approach presented in this paper shows great potential in the practical use of BN measurement where grinding burns can be detected and evaluated with a surface BN measurement.

Non-destructive recognition of copy paper based on advanced spectral fusion and feature optimization

In order to provide more clues for ongoing investigations and case handling, as well as achieve fast, non-destructive, and accurate identification of copy paper found at crime scenes, this study aims to utilize advanced spectral fusion technology to characterize and identify the three-dimensional features of the “origin-manufacturer-brand” of copy paper. Confocal Raman Microscopic and Fourier transform infrared spectroscopy were employed to collect spectral data from 200 samples from four regions (Shandong, Henan, Shaanxi, Jiangsu). The effects of different preprocessing methods, such as Hilbert transformation and deconvolution, on the model's ability to distinguish were compared. Feature variables were extracted using principal component analysis, and Bayesian discriminant classification models were constructed based on single infrared spectroscopy, Raman spectroscopy, and three types of spectral fusion datasets. By comparing the classification accuracy of different models, the primary fusion based on the full spectrum dataset was selected as the optimal model for the three-dimensional feature classification of copy paper. The accuracy achieved for origin (96%), manufacturer (100%), and brand (100%) was satisfactory, and the classification results were highly accurate. This study provides valuable insights and serves as a reference for its application in forensic science research.

A Study of the Ordinal Scale Classification Algorithm for Cyber Threat Intelligence Based on Deception Technology

Cyber deception technology plays an important role in monitoring attackers’ activities and detecting new attack types. However, in a deceptive environment, low-risk attack traffic, such as scanning, is included in large quantities and acts as noise. Therefore, even though high-risk traffic is actually present, it may be overlooked, or the analysis algorithm’s accuracy regarding traffic may be reduced, causing significant difficulties in intrusion detection and analysis processes. In this study, we propose a model that can identify and filter the ordinal scale risk of the source IP in deceptive environment-generated traffic. This model aims to quickly classify low-risk attacks, including information gathering and scanning, which are widely and repeatedly performed, as well as high-risk attacks, rather than classifying specific types of attacks. Most existing deceptive technology-based Cyber Threat Intelligence (CTI) generation studies have been limited in their applicability to real-world environments because data labeling, learning, and detection processes using AI algorithms that consume significant amounts of time and computing resources. Here, the Naive Bayes discriminant analysis-based ordinary scale classification model showed higher accuracy for low-risk attack classification, while consuming significantly fewer resources than the models presented in other studies do. The accuracy of the current active deceptive environment traffic analysis research may be enhanced by filtering low-risk traffic via preprocessing.

A Metabolomics-Based Study on the Discriminative Classification Models and Toxicological Mechanism of Estazolam Fatal Intoxication.

Fatal intoxication with sedative-hypnotic drugs is increasing yearly. However, the plasma drug concentration data for fatal intoxication involving these substances are not systematic and even overlap with the intoxication group. Therefore, developing a more precise and trustworthy approach to determining the cause of death is necessary. This study analyzed mice plasma and brainstem samples using the liquid chromatography-high resolution tandem mass spectrometry (LC-HR MS/MS)-based metabolomics method to create discriminative classification models for estazolam fatal intoxication (EFI). The most perturbed metabolic pathway between the EFI and EIND (estazolam intoxication non-death) was examined, Both EIND and EFI groups were administered 500 mg of estazolam per 100 g of body weight. Mice that did not die beyond 8 hours were treated with cervical dislocation and were classified into the EIND groups; the lysine degradation pathway was verified by qPCR (Quantitative Polymerase Chain Reaction), metabolite quantitative and TEM (transmission electron microscopy) analysis. Non-targeted metabolomics analysis with EFI were the experimental group and four hypoxia-related non-drug-related deaths (NDRDs) were the control group. Mass spectrometry data were analyzed with Compound Discoverer (CD) 3.1 software and multivariate statistical analyses were performed using the online software MetaboAnalyst 5.0. After a series of analyses, the results showed the discriminative classification model in plasma was composed of three endogenous metabolites: phenylacetylglycine, creatine and indole-3-lactic acid, and in the brainstem was composed of palmitic acid, creatine, and indole-3-lactic acid. The specificity validation results showed that both classification models distinguished between the other four sedatives-hypnotics, with an area under ROC curve (AUC) of 0.991, and the classification models had an extremely high specificity. When comparing different doses of estazolam, the AUC value of each group was larger than 0.80, and the sensitivity was also high. Moreover, the stability results showed that the AUC value was equal to or very close to 1 in plasma samples stored at 4 °C for 0, 1, 5, 10 and 15 days; the predictive power of the classification model was stable within 15 days. The results of lysine degradation pathway validation revealed that the EFI group had the highest lysine and saccharopine concentrations (mean (ng/mg) = 1.089 and 1.2526, respectively) when compared to the EIND and control group, while the relative expression of SDH (saccharopine dehydrogenase) showed significantly lower in the EFI group (mean = 1.206). Both of these results were statistically significant. Furthermore, TEM analysis showed that the EFI group had the more severely damaged mitochondria. This work gives fresh insights into the toxicological processes of estazolam and a new method for identifying EFI-related causes of mortality.

NMR-Metabolomics Reveals a Metabolic Shift after Surgical Resection of Non-Small Cell Lung Cancer

Lung cancer can be detected by measuring the patient's plasma metabolomic profile using nuclear magnetic resonance (NMR) spectroscopy. This NMR-based plasma metabolomic profile is patient-specific and represents a snapshot of the patient's metabolite concentrations. The onset of non-small cell lung cancer (NSCLC) causes a change in the metabolite profile. However, the level of metabolic changes after complete NSCLC removal is currently unknown. Fasted pre- and postoperative plasma samples of 74 patients diagnosed with resectable stage I-IIIA NSCLC were analyzed using 1H-NMR spectroscopy. NMR spectra (s = 222) representing two preoperative and one postoperative plasma metabolite profile at three months after surgical resection were obtained for all patients. In total, 228 predictors, i.e., 228 variables representing plasma metabolite concentrations, were extracted from each NMR spectrum. Two types of supervised multivariate discriminant analyses were used to train classifiers presenting a strong differentiation between the pre- and postoperative plasma metabolite profiles. The validation of these trained classification models was obtained by using an independent dataset. A trained multivariate discriminant classification model shows a strong differentiation between the pre- and postoperative NSCLC profiles with a specificity of 96% (95% CI [86-100]) and a sensitivity of 92% (95% CI [81-98]). Validation of this model results in an excellent predictive accuracy of 90% (95% CI [77-97]) and an AUC value of 0.97 (95% CI [0.93-1]). The validation of a second trained model using an additional preoperative control sample dataset confirms the separation of the pre- and postoperative profiles with a predictive accuracy of 93% (95% CI [82-99]) and an AUC value of 0.97 (95% CI [0.93-1]). Metabolite analysis reveals significantly increased lactate, cysteine, asparagine and decreased acetate levels in the postoperative plasma metabolite profile. The results of this paper demonstrate that surgical removal of NSCLC generates a detectable metabolic shift in blood plasma. The observed metabolic shift indicates that the NSCLC metabolite profile is determined by the tumor's presence rather than donor-specific features. Furthermore, the ability to detect the metabolic difference before and after surgical tumor resection strongly supports the prospect that NMR-generated metabolite profiles via blood samples advance towards early detection of NSCLC recurrence.

Identification of markers for tea authenticity assessment: Non-targeted metabolomics of highly similar oolong tea cultivars (Camellia sinensis var. sinensis)

Tea authenticity assessment remains a challenge because of various cultivars and similar chemical profiles among tea cultivars. Unique oolong tea cultivars (YaShiXiang (YSX)) were used as models in the present study to identify potential markers for tea authenticity assessment via non-targeted metabolomics. With chemometric analysis, a total of ten markers were found to differ significantly among YSX and the high similarity cultivar tea samples, which were up-regulated in YSX. Based on the ten markers, classification model for discrimination of YSX were developed with different combinations of markers. All models were able to classify YSX samples correctly (93.0%–100% for internal cross-validation; 80.5%–99.0% for external validation). The simultaneous determination of metabolites in this study suggested that a combination of non-targeted metabolomics and chemometric methods could provide high resolution markers for tea authenticity assessment, and which can serve as a basis for further quality evaluation and breeding of tea cultivars in the future. • Highly similar oolong tea cultivars were analyzed for tea authenticity assessment. • Chemical composition of tea were analyzed by UHPLC Q-Exative Orbitrap MS. • Ten markers were identified to evaluate the authenticity of oolong tea.

Dynamic Changes in Plasma Metabolic Profiles Reveal a Potential Metabolite Panel for Interpretation of Fatal Intoxication by Chlorpromazine or Olanzapine in Mice.

Diagnosing the cause of fatal intoxication by antipsychotic agents is an important task in forensic practice. In the 2020 Annual Report of the American Association of Poison Control Centers, among 40 deaths caused by antipsychotics, 21 cases were diagnosed as "probably responsible", thereby indicating that more objective diagnostic tools are needed. We used liquid chromatography-mass spectrometry-based integrated metabolomics analysis to measure changes in metabolic profiles in the plasma of mice that died from fatal intoxication due to chlorpromazine (CPZ) or olanzapine (OLA). These results were used to construct a stable discriminative classification model (DCM) comprising L-acetylcarnitine, succinic acid, and propionylcarnitine between fatal intoxication caused by CPZ/OLA and cervical dislocation (control). Performance evaluation of the classification model in mice that suffered fatal intoxication showed relative specificity for different pharmacodynamic drugs and relative sensitivity in different life states (normal, intoxication, fatal intoxication). A stable level of L-acetylcarnitine and variable levels of succinic acid and propionylcarnitine between fatal-intoxication and intoxication groups revealed procedural perturbations in metabolic pathways related to fatal intoxication by CPZ/OLA. Additional stability studies revealed that decomposition of succinic acid in fatal-intoxication samples (especially in the OLA group) could weaken the prediction performance of the binary-classification model; however, levels of these three potential metabolites measured within 6 days in fresh samples kept at 4 °C revealed a good performance of our model. Our findings suggest that metabolomics analysis can be used to explore metabolic alterations during fatal intoxication due to use of antipsychotic agents and provide evidence for the cause of death.

Multi-class classification model for psychiatric disorder discrimination

BackgroundPhysicians follow-up a symptom-based approach in the diagnosis of psychiatric diseases. According to this approach, a process based on internationally valid diagnostic tools such as The Diagnostic and Statistical Manual of Mental Disorders (DSM) or International Classification of Diseases (ICD), patient reports and the observation and experience of the physician is monitored. As in other fields of medicine, the search for biomarkers that can be used in processes related to diseases continues in psychiatry and various researches are carried out in this field. ObjectivesWithin the scope of this study, a dataset containing electroencephalogram (EEG) measurements of individuals diagnosed with different psychiatric diseases were analyzed by machine learning methods and the diseases were differentiated/classified with the models obtained. Thus, it was investigated whether EEG data could be a biomarker for psychiatric diseases. Materials and methodsIn the dataset analyzed within the scope of the study, for 550 patients (81 bipolar disorder, 95 attention deficit and hyperactivity disorder - ADHD, 67 depression, 34 obsessive compulsive disorder - OCD, 75 opioid, 146 posttraumatic stress disorder - PTSD, 52 schizophrenia) and 84 healthy individuals, there are 634 samples (rows), 77 variables (columns) in total. 76 of the variables consist of absolute power values belonging to 4 frequency bands (alpha, beta, delta, theta) collected from 19 different electrodes. 80 % of the dataset was used for training the models and 20 % of the data was used for testing the performance of the models. The 5-fold cross validation (CV) method, which repeats 3 times in the training dataset, was used and with this method, the hyperparameters used in the models were also optimized. Different models have been established with the selected hyperparameters and the performance of these models has been tested with the test dataset. C5.0, random forest (RF), support vector machine (SVM) and artificial neural networks (ANN) were used to build the models. ResultsWithin the scope of the study, the absolute power values obtained from EEG measurements performed using 19 electrodes were analyzed by machine learning methods. It was concluded that classification between disease groups was feasible with a high accuracy (C5.0–0.841, SVM_radial − 0.841, RF − 0.762). It was observed that ADHD, depression and schizophrenia diseases can be differentiated better (F-score = 1, balanced accuracy = 1) once the results were evaluated on a class category basis according to the F- measure and balanced accuracy values. Discussion and conclusionThrough the medium of the analyzes made within the scope of this study, it was investigated whether EEG data could be used as a biomarker for the detection and diagnosis of psychiatric diseases. The findings obtained from this study revealed that by using EEG data as a biomarker, it can be highly predicted whether a person has a psychiatric disease or not. Once evaluated with broad strokes, it is feasible to assert that it is possible to analyze whether the person who consults a physician with a complaint is ranked among the psychiatric disease class with EEG measurement. When trying to differentiate between numerous and diverse disease categories, it may be claimed that some diseases (ADHD, depression, schizophrenia) can be distinguished better by coming to the fore on a model basis. Considering the findings, it is anticipated that the analyzes obtained as a result of this study will contribute to the studies to be conducted using machine learning in the field of psychiatry.

A Linear Discriminant Analysis and Classification Model for Breast Cancer Diagnosis

Although most cases are identified at a late stage, breast cancer is the most public malignancy amongst women globally. However, mammography for the analysis of breast cancer is not routinely available at all general hospitals. Prolonging the period between detection and treatment for breast cancer may raise the likelihood of proliferating the disease. To speed up the process of diagnosing breast cancer and lower the mortality rate, a computerized method based on machine learning was created. The purpose of this investigation was to enhance the investigative accuracy of machine-learning algorithms for breast cancer diagnosis. The use of machine-learning methods will allow for the classification and prediction of cancer as either benign or malignant. This investigation applies the machine learning algorithms of random forest (RF) and the support vector machine (SVM) with the feature extraction method of linear discriminant analysis (LDA) to the Wisconsin Breast Cancer Dataset. The SVM with LDA and RF with LDA yielded accuracy results of 96.4% and 95.6% respectively. This research has useful applications in the medical field, while it enhances the efficiency and precision of a diagnostic system. Evidence from this study shows that better prediction is crucial and can benefit from machine learning methods. The results of this study have validated the use of feature extraction for breast cancer prediction when compared to the existing literature.

On robust risk-based active-learning algorithms for enhanced decision support

Classification models are a fundamental component of physical-asset management technologies such as structural health monitoring (SHM) systems and digital twins. Previous work introduced risk-based active learning, an online approach for the development of statistical classifiers that takes into account the decision-support context in which they are applied. Decision-making is considered by preferentially querying data labels according to expected value of perfect information (EVPI). Although several benefits are gained by adopting a risk-based active learning approach, including improved decision-making performance, the algorithms suffer from issues relating to sampling bias as a result of the guided querying process. This sampling bias ultimately manifests as a decline in decision-making performance during the later stages of active learning, which in turn corresponds to lost resource/utility.The current paper proposes two novel approaches to counteract the effects of sampling bias: semi-supervised learning, and discriminative classification models. These approaches are first visualised using a synthetic dataset, then subsequently applied to an experimental case study, specifically, the Z24 Bridge dataset. The semi-supervised learning approach is shown to have variable performance; with robustness to sampling bias dependent on the suitability of the generative distributions selected for the model with respect to each dataset. In contrast, the discriminative classifiers are shown to have excellent robustness to the effects of sampling bias. Moreover, it was found that the number of inspections made during a monitoring campaign, and therefore resource expenditure, could be reduced with the careful selection of the statistical classifiers used within a decision-supporting monitoring system.

Machine Learning Assisted Handheld Confocal Raman Micro-Spectroscopy for Identification of Clinically Relevant Atopic Eczema Biomarkers.

Atopic dermatitis (AD) is a common chronic inflammatory skin dermatosis condition due to skin barrier dysfunction that causes itchy, red, swollen, and cracked skin. Currently, AD severity clinical scores are subjected to intra- and inter-observer differences. There is a need for an objective scoring method that is sensitive to skin barrier differences. The aim of this study was to evaluate the relevant skin chemical biomarkers in AD patients. We used confocal Raman micro-spectroscopy and advanced machine learning methods as means to classify eczema patients and healthy controls with sufficient sensitivity and specificity. Raman spectra at different skin depths were acquired from subjects’ lower volar forearm location using an in-house developed handheld confocal Raman micro-spectroscopy system. The Raman spectra corresponding to the skin surface from all the subjects were further analyzed through partial least squares discriminant analysis, a binary classification model allowing the classification between eczema and healthy subjects with a sensitivity and specificity of 0.94 and 0.85, respectively, using stratified K-fold (K = 10) cross-validation. The variable importance in the projection score from the partial least squares discriminant analysis classification model further elucidated the role of important stratum corneum proteins and lipids in distinguishing two subject groups.