Quadratic Discriminant Analysis Models Research Articles

Construction of novel multi-epitope-based diagnostic biomarker HP16118P and its application in the differential diagnosis of Mycobacterium tuberculosis latent infection

Tuberculosis (TB) is an infectious disease that significantly threatens human health. However, the differential diagnosis of latent tuberculosis infection (LTBI) and active tuberculosis (ATB) remains a challenge for clinicians in early detection and preventive intervention. In this study, we developed a novel biomarker named HP16118P, utilizing 16 helper T lymphocyte (HTL) epitopes, 11 cytotoxic T lymphocyte (CTL) epitopes, and 8 B cell epitopes identified from 15 antigens associated with LTBI-RD using the IEDB database. We analyzed the physicochemical properties, spatial structure, and immunological characteristics of HP16118P using various tools, which indicated that it is a hydrophilic and relatively stable alkaline protein. Furthermore, HP16118P exhibited good antigenicity and immunogenicity, while being non-toxic and non-allergenic, with the potential to induce immune responses. We observed that HP16118P can stimulate the production of high levels of IFN-γ+ T lymphocytes in individuals with ATB, LTBI, and health controls. IL-5 induced by HP16118P demonstrated potential in distinguishing LTBI individuals and ATB patients (p=0.0372, AUC=0.8214, 95% CI [0.5843 to 1.000]) with a sensitivity of 100% and specificity of 71.43%. Furthermore, we incorporated the GM-CSF, IL-23, IL-5, and MCP-3 induced by HP16118P into 15 machine learning algorithms to construct a model. It was found that the Quadratic discriminant analysis model exhibited the best diagnostic performance for discriminating between LTBI and ATB, with a sensitivity of 1.00, specificity of 0.86, and accuracy of 0.93. In summary, HP16118P has demonstrated strong antigenicity and immunogenicity, with the induction of GM-CSF, IL-23, IL-5, and MCP-3, suggesting their potential for the differential diagnosis of LTBI and ATB.

Diagnostic Prediction of portal vein thrombosis in chronic cirrhosis patients using data-driven precision medicine model.

Portal vein thrombosis (PVT) is a significant issue in cirrhotic patients, necessitating early detection. This study aims to develop a data-driven predictive model for PVT diagnosis in chronic hepatitis liver cirrhosis patients. We employed data from a total of 816 chronic cirrhosis patients with PVT, divided into the Lanzhou cohort (n = 468) for training and the Jilin cohort (n = 348) for validation. This dataset encompassed a wide range of variables, including general characteristics, blood parameters, ultrasonography findings and cirrhosis grading. To build our predictive model, we employed a sophisticated stacking approach, which included Support Vector Machine (SVM), Naïve Bayes and Quadratic Discriminant Analysis (QDA). In the Lanzhou cohort, SVM and Naïve Bayes classifiers effectively classified PVT cases from non-PVT cases, among the top features of which seven were shared: Portal Velocity (PV), Prothrombin Time (PT), Portal Vein Diameter (PVD), Prothrombin Time Activity (PTA), Activated Partial Thromboplastin Time (APTT), age and Child-Pugh score (CPS). The QDA model, trained based on the seven shared features on the Lanzhou cohort and validated on the Jilin cohort, demonstrated significant differentiation between PVT and non-PVT cases (AUROC = 0.73 and AUROC = 0.86, respectively). Subsequently, comparative analysis showed that our QDA model outperformed several other machine learning methods. Our study presents a comprehensive data-driven model for PVT diagnosis in cirrhotic patients, enhancing clinical decision-making. The SVM-Naïve Bayes-QDA model offers a precise approach to managing PVT in this population.

Identification of Commercial Antimalarial Herbal Drugs Using Laser-Induced Autofluorescence Technique and Multivariate Algorithms.

In malaria-prone developing countries the integrity of Anti-Malarial Herbal Drugs (AMHDs) which are easily preferred for treatment can be compromised. Currently, existing techniques for identifying AMHDs are destructive. We report on the use of non-destructive and sensitive technique, Laser-Induced-Autofluorescence (LIAF) in combination with multivariate algorithms for identification of AMHDs. The LIAF spectra were recorded from commercially prepared decoction AMHDs purchased from accredited pharmacy shop in Ghana. Deconvolution of the LIAF spectra revealed secondary metabolites belonging to derivatives of alkaloids and classes of phenolic compounds of the AMHDs. Principal Component Analysis (PCA) and Hierarchical Clustering Analysis (HCA) were able to discriminate the AMHDs base on their physicochemical properties. Based on two principal components, the PCA- QDA (Quadratic Discriminant Analysis), PCA-LDA (Linear Discriminant Analysis), PCA-SVM (Support Vector Machine) and PCA-KNN (K-Nearest Neighbour) models were developed with an accuracy performance of 99.0, 99.7, 100.0, and 100%, respectively, in identifying AMHDs. PCA-SVM and PCA-KNN provided the best classification and stability performance. The LIAF technique in combination with multivariate techniques may offer a non-destructive and viable tool for AMHDs identification.

Siberian and Russian sturgeon natal origin in South America: Fish farm or established population?

AbstractSturgeon (Acipenser baerii and A. gueldenstaedtii) specimens, native to the northern hemisphere, were reported in different environments of the La Plata Basin (South America). The objectives of this study were to provide the first insights into the natal origin and habitat use of these sturgeon species in the La Plata Basin through geochemical analysis (87Sr/86Sr) of fin spines and to review historical catch reports. Spine core‐to‐edge 87Sr/86Sr ratios were measured by LA‐MC‐ICPMS. A Quadratic Discriminant Analysis model based on water 87Sr/86Sr baseline of the La Plata Basin was run to infer the natal origin. The isotopic profiles indicate a common origin, compatible with the location of the fish farms in the Uruguay Basin. The A. baerii isotopic time series suggested that the fish moved towards the Paraná Basin months before capture, while A. gueldenstaedtii would have survived a longer time (perhaps years). Seventeen reports of sturgeons were recorded and preserved in museum collections between 1998 and 2020. Sturgeons were collected from Río de la Plata Estuary, Paraná and Uruguay basins and Atlantic coastal lagoons. It is recommended to closely monitor sturgeon catches, paying special attention to the appearance of specimens of reproductive age, in order to generate management and management plans if necessary.

오류 라벨을 포함한 분류모형의 강건성 모의실험연구

In order to increase the accuracy and performance of classification models, the quality of data must be guaranteed. There are various factors that affect the quality of these data, such as measurement accuracy, missing value, and sampling biases. This study focuses on the robustness of the classification model for mislabeled data. The reason we are interested in mislabeled data is that it significantly degrades the performance of the model more than other common measurement errors. In this study, the efficiency of various classification methods for data with mislabels is reviewed. Fisher's linear discriminant analysis model (LDA), quadratic discriminant analysis model (QDA), and logistic regression model were considered in the statistical model-based method, and SVM (support vector machine) and boosting are compared in the algorithm-based method. As a result of simulation study, it is found that the statistical model based approaches show better results in terms of robustness in the data assuming various situations of mislabels.

Discrimination of cancerous from benign pigmented skin lesions based on multispectral autofluorescence lifetime imaging dermoscopy and machine learning.

.SignificanceAccurate early diagnosis of malignant skin lesions is critical in providing adequate and timely treatment; unfortunately, initial clinical evaluation of similar-looking benign and malignant skin lesions can result in missed diagnosis of malignant lesions and unnecessary biopsy of benign ones.AimTo develop and validate a label-free and objective image-guided strategy for the clinical evaluation of suspicious pigmented skin lesions based on multispectral autofluorescence lifetime imaging (maFLIM) dermoscopy.ApproachWe tested the hypothesis that maFLIM-derived autofluorescence global features can be used in machine-learning (ML) models to discriminate malignant from benign pigmented skin lesions. Clinical widefield maFLIM dermoscopy imaging of 41 benign and 19 malignant pigmented skin lesions from 30 patients were acquired prior to tissue biopsy sampling. Three different pools of global image-level maFLIM features were extracted: multispectral intensity, time-domain biexponential, and frequency-domain phasor features. The classification potential of each feature pool to discriminate benign versus malignant pigmented skin lesions was evaluated by training quadratic discriminant analysis (QDA) classification models and applying a leave-one-patient-out cross-validation strategy.ResultsClassification performance estimates obtained after unbiased feature selection were as follows: 68% sensitivity and 80% specificity with the phasor feature pool, 84% sensitivity, and 71% specificity with the biexponential feature pool, and 84% sensitivity and 32% specificity with the intensity feature pool. Ensemble combinations of QDA models trained with phasor and biexponential features yielded sensitivity of 84% and specificity of 90%, outperforming all other models considered.ConclusionsSimple classification ML models based on time-resolved (biexponential and phasor) autofluorescence global features extracted from maFLIM dermoscopy images have the potential to provide objective discrimination of malignant from benign pigmented lesions. ML-assisted maFLIM dermoscopy could potentially assist with the clinical evaluation of suspicious lesions and the identification of those patients benefiting the most from biopsy examination.

Raman spectroscopy coupled with chemometric modeling approaches for authentication of different paprika varieties at physiological maturity

Five Balkan paprika varieties at physiological maturity were investigated by means of Raman spectroscopy in order to discriminate the differences which stemmed from their genetic variability since the plants were grown under the same experimental conditions. The spectra were obtained using the 532 nm wavelength. In an effort to find the best classification power, several pre-processing methods were applied: 1) baseline correction, unit vector normalization; 2) baseline correction, unit vector normalization and first Savitzky-Golay derivative; 3) baseline correction, unit vector normalization and second Savitzky-Golay derivative; 4) baseline correction, unit vector normalization and third Savitzky-Golay derivative. All of the pre-processing methods were followed by making PCA-LDA (Principal Component Analysis-Linear Discriminant Analysis), QDA (Quadratic Discriminant Analysis), and PLS-DA (Partial Least Square - Discriminant Analysis) classification models. QDA showed the best discrimination power (83.87–100% and 89.47–100% for the training and the test data, respectively), then PCA-LDA (0.00–100 and 0.00–100% for the training and the test data, respectively) and PLS-DA (19.35–100% and 0.00–100.00% for the training and the test data, respectively). The results pointed out the applicability of chemometric modeling associated with Raman spectroscopy in the assessment of nutritionally similar samples, such as the studied red paprika varieties.

Effective differentiation between depressed patients and controls using discriminative eye movement features

BackgroundDepression is a common debilitating mental disorder caused by various factors. Identifying and diagnosing depression are challenging because the clinical evaluation of depression is mainly subjective, lacking objective and quantitative indicators. The present study investigated the value and significance of eye movement measurements in distinguishing depressed patients from controls. MethodsNinety-five depressed patients and sixty-nine healthy controls performed three eye movement tests, including fixation stability, free-viewing, and anti-saccade tests, and eleven eye movement indexes were obtained from these tests. The independent t-test was adopted for group comparisons, and multiple logistic regression analysis was employed to identify diagnostic biomarkers. Support vector machine (SVM), quadratic discriminant analysis (QDA), and Bayesian (BYS) algorithms were applied to build the classification models. ResultsDepressed patients exhibited eye movement anomalies, characterized by increased saccade amplitude in the fixation stability test; diminished saccade velocity in the anti-saccade test; and reduced saccade amplitude, shorter scan path length, lower saccade velocity, decreased dynamic range of pupil size, and lower pupil size ratio in the free-viewing test. Four features mentioned above entered the logistic regression equation. The classification accuracies of SVM, QDA, and BYS models reached 86.0%, 81.1%, and 83.5%, respectively. ConclusionsDepressed patients exhibited abnormalities across multiple tests of eye movements, assisting in differentiating depressed patients from healthy controls in a cost-effective and non-invasive manner.

Performance of the supervised learning algorithms in sex estimation of the proximal femur: A comparative study in contemporary Egyptian and Turkish samples

Sex estimation standards are population specific however, we argue that machine learning techniques (ML) may enhance the biological sex determination on trans-population application. Linear discriminant analysis (LDA) versus nine ML including quadratic discriminant analysis (QDA), support vector machine (SVM), Decision Tree (DT), Gaussian process (GPC), Naïve Bayesian (NBC), K-Nearest Neighbor (KNN), Random Forest (RFM) and Adaptive boosting (Adaboost) were compared. The experiments involve two contemporary populations: Turkish (n = 300) and Egyptian populations (n = 100) for training and validation, respectively. Base models were calibrated using isotonic and sigmoid calibration schemes. Results were analyzed at posterior probabilities (pp) thresholds >0.95 and >0.80. At pp = 0.5, ML algorithms yielded comparable accuracies in the training (90% to 97%) and test sets (81% to 88%) which are not modified after employing the calibration techniques. At pp >0.95, the raw RFM, LDA, QDA, and SVM models have shown the best performance however, calibration techniques improved the performance of various classifier especially NBC and Adaboost. By contrast, the performance of GPC, KNN, QDA models worsened by calibration. RFM has shown the best performance among all models at both thresholds whereas LDA benefited the best from using both calibration methods at pp >0.80. Complex ML models are not necessarily achieving better performance metrics. LDA and QDA remain the fastest and simplest classifiers. We demonstrated the capability of enhancing sex estimation using ML on an independent population sample however, differences in the underlying probability distribution generated by models were detected which warranted more cautious application by forensic practitioners.

Abstract 13830: Prediction of Cardiac Rejection Trajectories via Machine Learning Derived Features From Digital Endomyocardial Biopsy Images

Introduction: Cellular rejection occurs in 20-40% of transplant recipients with an increased risk of graft failure. Previously, we developed a model for predicting ISHLT rejection grades via the automated extraction of morphologic features in Endomyocardial biopsy (EMB) images. Considering the frequent discordance between conventional rejection grade and clinical rejection severity, in this work, we sought to identify morphologic features that predict the clinical trajectory of rejection events. Methods: Our study comprised 299 EMBs with grade and trajectory labels. Trajectory labels are based on the development of overt clinical signs of allograft injury as “clinically evident” or “clinically silent”. Morphologic features describing number, spatial arrangement of lymphocytes, and shape, orientation of interstitial fibers were computationally extracted. To identify the top features associated with clinically evident disease, the T-test method was applied across 500 iterations of 3-fold cross validation. In each iteration, a quadratic discriminant analysis model was trained with top 10 features on 2 folds of dataset using trajectory labels (M trj ) and was validated on one hold out test set. This model was trained using grade labels (M grd ) to predict “high” (2R + 3R) or “low” (0R + 1R). To assess feature importance, in each model, the frequency of every feature appearing in the classifier was measured through iterations. Results: The mean area under the receiver operating curve of M trj and M grd was 0.80±0.04 and 0.84±0.02 correspondingly. The top features for predicting grades differ substantially from those for predicting clinical trajectory (Figure 1). Conclusions: The additional features required to predict clinical trajectory vs. rejection grade may explain the discordance between conventional histology and rejection syndrome observed in clinical practice, and highlights the translational potential of computer-assisted histologic analysis of EMBs.

Microbiological Quality Assessment of Chicken Thigh Fillets Using Spectroscopic Sensors and Multivariate Data Analysis.

Fourier transform infrared spectroscopy (FT-IR) and multispectral imaging (MSI) were evaluated for the prediction of the microbiological quality of poultry meat via regression and classification models. Chicken thigh fillets (n = 402) were subjected to spoilage experiments at eight isothermal and two dynamic temperature profiles. Samples were analyzed microbiologically (total viable counts (TVCs) and Pseudomonas spp.), while simultaneously MSI and FT-IR spectra were acquired. The organoleptic quality of the samples was also evaluated by a sensory panel, establishing a TVC spoilage threshold at 6.99 log CFU/cm2. Partial least squares regression (PLS-R) models were employed in the assessment of TVCs and Pseudomonas spp. counts on chicken’s surface. Furthermore, classification models (linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), support vector machines (SVMs), and quadratic support vector machines (QSVMs)) were developed to discriminate the samples in two quality classes (fresh vs. spoiled). PLS-R models developed on MSI data predicted TVCs and Pseudomonas spp. counts satisfactorily, with root mean squared error (RMSE) values of 0.987 and 1.215 log CFU/cm2, respectively. SVM model coupled to MSI data exhibited the highest performance with an overall accuracy of 94.4%, while in the case of FT-IR, improved classification was obtained with the QDA model (overall accuracy 71.4%). These results confirm the efficacy of MSI and FT-IR as rapid methods to assess the quality in poultry products.

Chemometrics-assisted isotope ratio fingerprinting based on gas chromatography/combustion/isotope ratio mass spectrometry for saffron authentication

In the present contribution, the capability of isotopic ratio mass spectrometry (IRMS) for saffron authentication and detection of four common plant-derived adulterants (marigold flower, safflower, rubia, and saffron style) was investigated. For this purpose, 62 authentic saffron samples were analyzed by elemental analyzer-IRMS (EA-IRMS) and gas chromatography-combustion-IRMS (GC-C-IRMS). In this regard, EA-IRMS and GC-C-IRMS isotope fingerprints of carbon-13 and nitrogen-15 isotopes of saffron components were provided and then analyzed by chemometric methods. Principal component analysis (PCA) showed two different behaviors regarding two main regions. Then, a representative saffron sample was provided to study adulteration. On this matter, binary mixtures of saffron and adulterants were prepared at five different weight percentages (5%, 10%, 15%, 25%, and 35%) and analyzed by EA-IRMS and GC-C-IRMS. Data-driven soft independent modeling of class analogy (DD-SIMCA) was used to model authentic saffron samples and find a boundary between authentic and adulterated samples with a sensitivity of 100% by GC-C-IRMS. After that, discriminant models of linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), and partial least squares-discriminant analysis (PLS-DA) were tested to find the best discrimination line and also detection of the lowest level of adulterants. Among different models, the QDA model outperformed other methods and showed the ability to predict adulterants at 5% w/w level with 100% accuracy and precision. Finally, the developed QDA model was successfully used to discriminate a set of mixed samples of saffron and four adulterants as well as some commercial samples.

Detection and Classification of Multi-Type Cells by Using Confocal Raman Spectroscopy.

Tumor cells circulating in the peripheral blood are the prime cause of cancer metastasis and death, thus the identification and discrimination of these rare cells are crucial in the diagnostic of cancer. As a label-free detection method without invasion, Raman spectroscopy has already been indicated as a promising method for cell identification. This study uses a confocal Raman spectrometer with 532 nm laser excitation to obtain the Raman spectrum of living cells from the kidney, liver, lung, skin, and breast. Multivariate statistical methods are applied to classify the Raman spectra of these cells. The results validate that these cells can be distinguished from each other. Among the models built to predict unknown cell types, the quadratic discriminant analysis model had the highest accuracy. The demonstrated analysis model, based on the Raman spectrum of cells, is propitious and has great potential in the field of biomedical for classifying circulating tumor cells in the future.

Determination of phosphorus status in bread wheat leaves by visible and near-infrared spectral discriminant analysis

This study developed a quadratic discriminant analysis (QDA) model from the spectroradiometer reflections (400 to 1000 nm) and phosphorus (P) uptake in wheat under varying rates of P dosages (0, 25, and 50 ppm P) in the tillering (GS25) and heading (GS55) stages. Seventy-two experimental plants were grown under controlled greenhouse conditions. Stepwise multiple regression analysis was used to determine the wavelengths associated with different periods and P doses. Principal component analysis was employed to select the five wavelengths (418, 563, 639, 756, and 1000 nm) that best encompassed the total variance amongst the different reflection values. The QDA model assigned the training data to their real classes (0 ppm P: 79%, 25 ppm P: 50%, and 50 ppm: 83%) with 71% accuracy. For validation of the model, 36 randomly selected test data were used (0 ppm P: 75%, 25 ppm P: 42%, and 50 ppm P; 92%) and resulted in 69% accuracy. Results concluded that wheat P demand during different vegetation stages can be determined from the spectral wavelengths input into a QDA model; for future research, however, we suggest the nutrient dosage ranges are broad enough to provide sufficient variability. Nevertheless, discriminant modeling is a viable method of determining plant nutritional status by spectral data.

Novel Quantitative Analysis Using Optical Imaging (VELscope) and Spectroscopy (Raman) Techniques for Oral Cancer Detection.

Simple SummaryOur study aims to develop a novel quantitative analysis method that can increase the oral cancer detection rate for screening oral cancer. We used two different optical techniques, a light-based detection technique (VELScope) and a vibrational spectroscopic technique (Raman spectroscopy). First, we analyzed and evaluated the performance of these two techniques individually using PCA–LDA, and PCA–QDA classifiers. The PCA–LDA of Raman spectroscopy had 82.9% accuracy, 80% sensitivity, and 85.7% specificity, while the region of interests on the autofluorescence images were differentiated with 90% accuracy, 100% sensitivity, and 80% specificity. Afterward, we combined both techniques and evaluated their performance. The combination of two optical techniques can differentiate the cancer and normal groups with 97.14% accuracy, 100% sensitivity, and 94.3% specificity. The main advantage of our study is that we can confirm our results by using two different techniques that are completely independent of each other. That is the reason that the combination of two techniques can increase the sensitivity and specificity.In this study, we developed a novel quantitative analysis method to enhance the detection capability for oral cancer screening. We combined two different optical techniques, a light-based detection technique (visually enhanced lesion scope) and a vibrational spectroscopic technique (Raman spectroscopy). Materials and methods: Thirty-five oral cancer patients who went through surgery were enrolled. Thirty-five cancer lesions and thirty-five control samples with normal oral mucosa (adjacent to the cancer lesion) were analyzed. Thirty-five autofluorescence images and 70 Raman spectra were taken from 35 cancer and 35 control group cryopreserved samples. The normalized intensity and heterogeneity of the 70 regions of interest (ROIs) were calculated along with 70 averaged Raman spectra. Linear discriminant analysis (LDA) and quadratic discriminant analysis (QDA) were used with principal component analysis (PCA) to differentiate the cancer and control groups (normal). The classifications rates were validated using two different validation methods, leave-one-out cross-validation (LOOCV) and k-fold cross-validation. Results: The cryopreserved normal and tumor tissues were differentiated using the PCA–LDA and PCA–QDA models. The PCA–LDA of Raman spectroscopy (RS) had 82.9% accuracy, 80% sensitivity, and 85.7% specificity, while ROIs on the autofluorescence images were differentiated with 90% accuracy, 100% sensitivity, and 80% specificity. The combination of two optical techniques differentiated cancer and normal group with 97.14% accuracy, 100% sensitivity, and 94.3% specificity. Conclusion: In this study, we combined the data of two different optical techniques. Furthermore, PCA–LDA and PCA–QDA quantitative analysis models were used to differentiate tumor and normal groups, creating a complementary pathway for efficient tumor diagnosis. The error rates of RS and VELcope analysis were 17.10% and 10%, respectively, which was reduced to 3% when the two optical techniques were combined.

Flow Regime Identification in Vertical Upward Gas–Liquid Flow Using an Optical Sensor With Linear and Quadratic Discriminant Analysis

Abstract The accurate identification of gas–liquid flow regimes in pipes remains a challenge for the chemical process industries. This paper proposes a method for flow regime identification that combines responses from a nonintrusive optical sensor with linear discriminant analysis (LDA) and quadratic discriminant analysis (QDA) for vertical upward gas–liquid flow of air and water. A total of 165 flow conditions make up the dataset, collected from an experimental air–water flow loop with a transparent test section (TS) of 27.3 mm inner diameter and 5 m length. Selected features extracted from the sensor response are categorized into feature group 1, average sensor response and standard deviation, and feature group 2 that also includes percentage counts of the calibrated responses for water and air. The selected features are used to train, cross validate, and test four model cases (LDA1, LDA2, QDA1, and QDA2). The LDA models produce higher average test classification accuracies (both 95%) than the QDA models (80% QDA1 and 45% QDA2) due to misclassification associated with the slug and churn flow regimes. Results suggest that the LDA1 model case is the most stable with the lowest average performance loss and is therefore considered superior for flow regime identification. In future studies, a larger dataset may improve stability and accuracy of the QDA models, and an extension of the conditions and parameters would be a useful test of applicability.

Preliminary study on discriminating HER2 2+ amplification status of breast cancers based on texture features semi-automatically derived from pre-, post-contrast, and subtraction images of DCE-MRI.

ObjectiveTo investigate whether texture features extracted from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) are associated with human epidermal growth factor receptor type 2 (HER2) 2+ status of breast cancer.Materials and methods92 MRI cases including 52 HER2 2+ positive and 40 negative patients confirmed by fluorescence in situ hybridization were retrospectively selected. The lesion area was semi-automatically delineated, and a total of 488 texture features were respectively extracted from precontrast, postcontrast, and subtraction images. The Student’s t-test or Mann-Whitney U test was performed to identify statistically significant features between different HER2 2+ amplification groups. Least absolute shrinkage and selection operator (LASSO) was used to search for the optimal feature subsets. Three machine learning classifiers, logistic regression analysis (LRA), quadratic discriminant analysis (QDA), and support vector machine (SVM), were used with a leave-one-out cross validation method to establish the classification models of HER2 2+ status. Classification performance was evaluated by receiver operating characteristic (ROC) analysis.ResultsBased on the texture analysis with SVM model, the areas under the ROC curve (AUCs) were 0.890 for subtraction images, 0.736 for postcontrast images, and 0.672 for precontrast images, respectively. For LRA model, the AUCs were 0.884, 0.733, and 0.623, respectively. For QDA model, the AUCs were 0.831, 0.726, and 0.568, respectively. LRA and the SVM model with subtraction images reached significantly better performance than the QDA model (P = 0.0227 and P = 0.0088, respectively).ConclusionTexture features of breast cancer extracted from DCE-MRI are associated with HER2 2+ status. Additional studies are necessary to confirm the present preliminary findings.

Discriminative algorithm approach to forecast Cd threshold exceedance probability for rice grain based on soil characteristics

The relationship between cadmium (Cd) concentration in rice grains and the soil that they are cultivated in is highly uncertain due to the influence of soil properties, rice varieties, and other undetermined factors. In this study, we introduce the probability of exceeding the threshold to characterize this uncertainty and then, build a probabilistic forewarning model. Additionally, a number of associated factors have been used as parameters to improve model performance. Considering that the physicochemical properties and Cd concentration in the soil (Cdsoil) do not follow a normal distribution, and are not independent of each other, a discriminative algorithm, represented by a logistic regression (LR), performed better than generative algorithms, such as the naive Bayes and quadratic discriminant analysis models. The performance of the LR based model was found to be 0.5% better in the case of the univariate model (Cdsoil) and 4.1% better with a multivariate model (soil properties used as additional factors) (p < 0.01). The output of the LR based model predicted probabilities that were positively correlated to the true exceedance rate (R2 = 0.949,p < 0.01), within an exceedance threshold range of 0.1–0.4 mg kg−1 and a mean deviation of 5.75%. A sensitivity analysis showed that the effect of soil properties on the exceedance probability weakens with an increase in Cd concentration in rice grains. When the threshold is below 0.15 mg kg−1, soil pH strongly influences the exceedance probability. As the threshold increases, the influence of pH on the exceedance probability is gradually superseded. By quantifying the uncertainty regarding the relationship between Cd concentration in rice grains and soil, the discriminative algorithm-based probabilistic forecasting model offers a new way to assess Cd pollution in rice grown in contaminated paddy fields.

Machine Learning for the Prediction of New-Onset Diabetes Mellitus during 5-Year Follow-up in Non-Diabetic Patients with Cardiovascular Risks.

PurposeMany studies have proposed predictive models for type 2 diabetes mellitus (T2DM). However, these predictive models have several limitations, such as user convenience and reproducibility. The purpose of this study was to develop a T2DM predictive model using electronic medical records (EMRs) and machine learning and to compare the performance of this model with traditional statistical methods.Materials and MethodsIn this study, a total of available 8454 patients who had no history of diabetes and were treated at the cardiovascular center of Korea University Guro Hospital were enrolled. All subjects completed 5 years of follow up. The prevalence of T2DM during follow up was 4.78% (404/8454). A total of 28 variables were extracted from the EMRs. In order to verify the cross-validation test according to the prediction model, logistic regression (LR), linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), and K-nearest neighbor (KNN) algorithm models were generated. The LR model was considered as the existing statistical analysis method.ResultsAll predictive models maintained a change within the standard deviation of area under the curve (AUC) <0.01 in the analysis after a 10-fold cross-validation test. Among all predictive models, the LR learning model showed the highest prediction performance, with an AUC of 0.78. However, compared to the LR model, the LDA, QDA, and KNN models did not show a statistically significant difference.ConclusionWe successfully developed and verified a T2DM prediction system using machine learning and an EMR database, and it predicted the 5-year occurrence of T2DM similarly to with a traditional prediction model. In further study, it is necessary to apply and verify the prediction model through clinical research.

Efficient Integration of Sufficient Dimension Reduction and Prediction in Discriminant Analysis

ABSTRACTSufficient dimension reduction (SDR) methods are popular model-free tools for preprocessing and data visualization in regression problems where the number of variables is large. Unfortunately, reduce-and-classify approaches in discriminant analysis usually cannot guarantee improvement in classification accuracy, mainly due to the different nature of the two stages. On the other hand, envelope methods construct targeted dimension reduction subspaces that achieve dimension reduction and improve parameter estimation efficiency at the same time. However, little is known about how to construct envelopes in discriminant analysis models. In this article, we introduce the notion of the envelope discriminant subspace (ENDS) as a natural inferential and estimative object in discriminant analysis that incorporates these considerations. We develop the ENDS estimators that simultaneously achieve sufficient dimension reduction and classification. Consistency and asymptotic normality of the ENDS estimators are established, where we carefully examine the asymptotic efficiency gain under the classical linear and quadratic discriminant analysis models. Simulations and real data examples show superb performance of the proposed method. Supplementary materials for this article are available online.