Class Prediction of High-Dimensional Data with Class Imbalance: Breast Cancer Gene Expression Data

Identification of molecular signatures that allow detection of the transition from normal breast epithelial cells to malignant invasive cells is a critical component in the development of diagnostic, therapeutic, and preventative strategies for human breast cancer. Substantial efforts have been devoted to deciphering breast cancer etiology at the genome level, but only a limited number of studies have appeared at the proteome level. In this work, we compared individual in situ proteome profiles of nonpatient matched nine noncancerous, normal breast epithelial (NBE) samples with nine estrogen receptor (ER)-positive (luminal subtype), invasive malignant breast epithelial (MBE) samples by combining laser capture microdissection (LCM) and quantitative shotgun proteomics. A total of 12,970 unique peptides were identified from the 18 samples, and 1623 proteins were selected for quantitative analysis using spectral index (SpI) as a measure of protein abundance. A total of 298 proteins were differentially expressed between NBE and MBE at 95% confidence level, and this differential expression correlated well with immunohistochemistry (IHC) results reported in the Human Protein Atlas (HPA) database. To assess pathway level patterns in the observed expression changes, we developed protein set enrichment analysis (PSEA), a modification of a well-known approach in gene expression analysis, Gene Set Enrichment Analysis (GSEA). Unlike single gene-based functional term enrichment analyses that only examines pathway overrepresentation of proteins above a given significance threshold, PSEA applies a weighted running sum statistic to the entire expression data to discover significantly enriched protein groups. Application of PSEA to the expression data in this study revealed not only well-known ER-dependent and cellular morphology-dependent protein abundance changes, but also significant alterations of downstream targets for multiple transcription factors (TFs), suggesting a role for specific gene regulatory pathways in breast tumorigenesis. A parallel GOMiner analysis revealed both confirmatory and complementary data to PSEA. The combination of the two annotation approaches yielded extensive biological feature mapping for in depth analysis of the quantitative proteomic data.

High-dimensional gene expression data often exhibit intricate correlation patterns as the result of coordinated genetic regulation. In practice, however, it is difficult to directly measure these coordinated underlying activities. Analysis of breast cancer survival data with gene expressions motivates us to use a two-stage latent factor approach to estimate these unobserved coordinated biologicalprocesses. Compared to existing approaches, our proposed procedure has several unique characteristics. In the first stage, an important distinction is that our procedure incorporates prior biological knowledge about gene-pathway membership into the analysis and explicitly model the effects of genetic pathways on the latent factors. Second, to characterize the molecular heterogeneity of breast cancer, our approach provides estimates specific to each cancer subtype. Finally, our proposed framework incorporates sparsity condition due to the fact that genetic networks are oftensparse. In the second stage, we investigate the relationship between latent factor activity levels and survival time with censoring using a general dimension reduction model in the survival analysis context. Combining the factor model and sufficient direction model provides an efficient way of analyzing high-dimensional data and reveals some interesting relations in the breast cancer gene expressiondata.

Background: Breast cancer is the most common malignancy among women worldwide, underscoring the importance of early detection and accurate prognostication. Machine learning (ML) has emerged as a promising approach, offering powerful tools for analyzing complex datasets in breast cancer prediction and diagnosis. Objective: This study evaluates the predictive performance of diverse ML algorithms for breast cancer classification using publicly available datasets, focusing on accuracy, interpretability, and generalizability. Methods: The dataset included clinical and demographic variables such as age, menopausal status, tumor size, and lymph node involvement. Data preprocessing addressed missing values and class imbalance, with the Synthetic Minority Oversampling Technique (SMOTE) applied to improve sensitivity for the minority class. Feature engineering involved interaction terms and scaling of numerical variables. Multiple ML models—Logistic Regression, Decision Tree, Random Forest, Gradient Boosting, Support Vector Machine (SVM), Naive Bayes, K-Nearest Neighbors (KNN), and Neural Networks—were trained and evaluated. Performance was measured using sensitivity, F1-score, and AUC-ROC. Model interpretability was enhanced with SHapley Additive exPlanations (SHAP). Results: Random Forest achieved the best performance with an AUC-ROC of 0.9751, followed by Gradient Boosting (0.9242) and Neural Networks (0.9254). Logistic Regression and SVM yielded comparable results (0.9005 and 0.9344). Ensemble models showed higher accuracy and generalizability, particularly on external validation. Tumor size and lymph node involvement emerged as key predictors. SMOTE improved sensitivity across models. Conclusion: This study demonstrates the potential of ML in breast cancer prediction, emphasizing the effectiveness of ensemble methods and interpretability tools. Future work should focus on integrating ML into clinical practice for earlier detection and personalized treatment.

This study investigates the adaptation of DeepInsight for cancer subtype classification using high-dimensional gene expression data. Originally designed for non-image data, DeepInsight has been adapted for cancer classification. We evaluated DeepInsight's performance against several models, including support vector machines, LightGBM, neural networks, and decision trees, with and without the application of the Synthetic Minority Oversampling Technique. The study utilized gene expression data from breast, lung, and colon cancers. A novel multi-class feature selection method was introduced, using modified aggregated class activation maps to identify key genes across different cancer subtypes. These critical genes were further analyzed through Gene Ontology to explore their roles in significant biological processes. DeepInsight consistently outperformed traditional models in terms of F1 score across breast, lung, and colon cancer datasets, effectively addressing multi-class classification challenges. Notably, several top genes were identified as significant across multiple methods. Furthermore, we conducted a Gene Ontology analysis on the critical genes, including the top genes identified by DeepInsight and the common genes recognized through multiple methods. The adaptation of DeepInsight provides an approach to cancer subtype classification by transforming high-dimensional gene expression data into image representations. Utilizing aggregated class activation maps, it effectively identifies critical pixels within these images, enabling the discovery of distinct genes that may not be highlighted by other methods. DeepInsight demonstrates potential as a valuable tool for classifying cancer subtypes and critical genes.

Cardiovascular diseases remain one of the leading causes of mortality worldwide, particularly in low- and middle-income countries. Early and accurate prediction of heart disease is essential for timely intervention and improved patient outcomes. Machine learning techniques offer promising solutions; however, challenges such as class imbalance, lack of interpretability, and limited real-world validation persist. In this study, a machine learning–based heart disease prediction framework was developed using a real-world clinical dataset comprising 5000 patient records collected from healthcare facilities in Bangladesh. Data preprocessing included cleaning, feature encoding, train–test splitting, and class imbalance handling using the Synthetic Minority Oversampling Technique (SMOTE). Multiple machine learning models—Logistic Regression, Decision Tree, Support Vector Machine, and Random Forest—were evaluated using 10-fold stratified cross-validation. Model performance was assessed using accuracy, precision, recall, and F1-score. SHAP (SHapley Additive exPlanations) was employed to enhance model interpretability. The best-performing model was deployed as a web-based decision support system. Among the evaluated models, the Random Forest classifier achieved the best performance, with an accuracy of 98%, recall of 96%, and F1-score of 96%. Ablation studies demonstrated the effectiveness of SMOTE, feature integration, and ensemble modeling. SHAP analysis identified clinically relevant features contributing to heart disease prediction, enhancing transparency and trust in model decisions. The proposed framework provides an accurate, interpretable, and practical solution for heart disease prediction using real-world clinical data. The integration of explainable machine learning and web-based deployment highlights its potential for clinical decision support. Future work will focus on multi-center prospective validation and adaptive model updating to further improve generalizability and real-world applicability. • A large real-world clinical dataset of 5000 patients was used for robust heart disease prediction. • A Random Forest model with SMOTE achieved high performance (98% accuracy, 96% recall). • Comprehensive ablation and model comparison analyses validate methodological robustness. • SHAP-based explainability provides transparent and clinically interpretable predictions. • The proposed system is deployed as a web-based clinical decision support tool.

Prediction of the stage of cancer plays an important role in planning the course of treatment and has been largely reliant on imaging tools which do not capture molecular events that cause cancer progression. Gene-expression data-based analyses are able to identify these events, allowing RNA-sequence and microarray cancer data to be used for cancer analyses. Breast cancer is the most common cancer worldwide, and is classified into four stages - stages 1, 2, 3, and 4 [2]. While machine learning models have previously been explored to perform stage classification with limited success, multi-class stage classification has not had significant progress. There is a need for improved multi-class classification models, such as by investigating deep learning models. Gene-expression-based cancer data is characterised by the small size of available datasets, class imbalance, and high dimensionality. Class balancing methods must be applied to the dataset. Since all the genes are not necessary for stage prediction, retaining only the necessary genes can improve classification accuracy. The breast cancer samples are to be classified into 4 classes of stages 1 to 4. Invasive ductal carcinoma breast cancer samples are obtained from The Cancer Genome Atlas (TCGA) and Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) datasets and combined. Two class balancing techniques are explored, synthetic minority oversampling technique (SMOTE) and SMOTE followed by random undersampling. A hybrid feature selection pipeline is proposed, with three pipelines explored involving combinations of filter and embedded feature selection methods: Pipeline 1 - minimum-redundancy maximum-relevancy (mRMR) and correlation feature selection (CFS), Pipeline 2 - mRMR, mutual information (MI) and CFS, and Pipeline 3 - mRMR and support vector machine-recursive feature elimination (SVM-RFE). The classification is done using deep learning models, namely deep neural network, convolutional neural network, recurrent neural network, a modified deep neural network, and an AutoKeras generated model. Classification performance post class-balancing and various feature selection techniques show marked improvement over classification prior to feature selection. The best multiclass classification was found to be by a deep neural network post SMOTE and random undersampling, and feature selection using mRMR and recursive feature elimination, with a Cohen-Kappa score of 0.303 and a classification accuracy of 53.1%. For binary classification into early and late-stage cancer, the best performance is obtained by a modified deep neural network (DNN) post SMOTE and random undersampling, and feature selection using mRMR and recursive feature elimination, with an accuracy of 81.0% and a Cohen-Kappa score (CKS) of 0.280. This pipeline also showed improved multiclass classification performance on neuroblastoma cancer data, with a best area under the receiver operating characteristic (auROC) curve score of 0.872, as compared to 0.71 obtained in previous work, an improvement of 22.81%. The results and analysis reveal that feature selection techniques play a vital role in gene-expression data-based classification, and the proposed hybrid feature selection pipeline improves classification performance. Multi-class classification is possible using deep learning models, though further improvement particularly in late-stage classification is necessary and should be explored further.

Criminal activities in the Kaduna-Abuja Highway have been an enormous security risk to the commuters, residents, and economic growth in Northern Nigeria. Conventional law enforcement approaches are still reactive and lack data, which restricts the anticipatory and preemptive actions. This paper describes Data Mining Analytics Approach-Based System of forecasting the occurrence of crime along the Kaduna-Abuja highway, using developed models of machine learning. The analysis used three supervised learning algorithms which include Support Vector Machine (SVM), K-Nearest Neighbor (KNN) and Random Forest (RF) to represent crime patterns and predict potential types of crime, time and location. These algorithms were selected because they are widely used for classification problems, handle and have shown strong performance in similar prediction studies. Artificial data of 1,500 entries that cited five years of criminal incidences was created and prepared. Class imbalance was dealt with with the help of SMOTE (Synthetic Minority Oversampling Technique), hyperparameters optimization with the help of the GridSearchCV. The experimental outcomes showed that the KNN classifier performed best having an accuracy of 71% then SVM and Random Forest with 66.5% respectively. Analysis of the importance of features showed that the most significant predictors were month, day of the week, number of victims, year, and time of day. Study evidences that predictive analytics based on data can complement the proactive policing practices and help the security agencies to predict the high-risk times and locations along the transport corridors in Nigeria that are critical.

Machine learning classifiers are increasingly used to create predictive models for pathological complete response (pCR) in breast cancer after neoadjuvant therapy (NAT). Few studies have compared the effectiveness of different ML classifiers. This study evaluated radiomics models based on pre- and post-contrast first-phase T1 weighted images (T1WI) in predicting breast cancer pCR after NAT and compared the performance of ML classifiers. This retrospective study enrolled 281 patients undergoing NAT from the Duke-Breast-Cancer-MRI dataset. Radiomic features were extracted from pre- and post-contrast first-phase T1WI images. The Synthetic Minority Oversampling Technique (SMOTE) was applied, then the dataset was randomly divided into training and validation groups (7:3). The radiomics model was built using selected optimal features. Support vector machine (SVM), random forest (RF), decision tree (DT), k-nearest neighbor (KNN), extreme gradient boosting (XGBoost), and light gradient boosting machine (LightGBM) were classifiers. Receiver operating characteristic curves were used to assess predictive performance. LightGBM performed best in predicting pCR [area under the curve (AUC): 0.823, 95% confidence interval (CI) [0.743-0.902], accuracy 74.0%, sensitivity 85.0%, specificity 67.2%]. During subgroup analysis, RF was most effective in pCR prediction in luminal breast cancers (AUC: 0.914, 95% CI [0.847-0.981], accuracy 87.0%, sensitivity 85.2%, specificity 88.1%). In triple-negative breast cancers, LightGBM performed best (AUC: 0.836, 95% CI [0.708-0.965], accuracy 78.6%, sensitivity 68.2%, specificity 90.0%). The LightGBM-based radiomics model performed best in predicting pCR in patients with breast cancer. RF and LightGBM showed promising results for luminal and triple-negative breast cancers, respectively.

BackgroundPhysical activity is a key focus in the field of public health, and subjective life expectancy is closely associated with individuals’ physical and psychological well-being. This study aimed to identify the risk factors for subjective life expectancy among middle-aged and older adults with active and inactive physical activity levels, and to provide an evidence base for developing differentiated health intervention strategies. MethodsBased on data from the China Health and Retirement Longitudinal Study (CHARLS) 2018 survey, a total of 10,945 participants were included. Five machine learning models, including Random Forest (RF), Logistic Regression (LR), Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), and Light Gradient Boosting Machine (LightGBM), were separately constructed for the active and inactive groups. To reduce bias caused by class imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) was applied to generate synthetic samples for the minority class. The dataset was split into a training set (70%) and a testing set (30%), and ten-fold cross-validation combined with grid search was employed to optimize hyperparameters, ensuring both robustness and generalizability of the models. Model performance was evaluated using the Area Under the Receiver Operating Characteristic Curve (AUC), accuracy, sensitivity, specificity, and F1-score.ResultsThe active group (4,707 men and 4,885 women) had a mean age of 59.76 years, while the inactive group (662 men and 691 women) had a mean age of 63.00 years. The Support Vector Machine (SVM) model achieved the best performance in the inactive group (AUC: 0.797; accuracy: 0.722; sensitivity: 0.747), whereas the Light Gradient Boosting Machine (LightGBM) model achieved the best performance in the active group (AUC: 0.775; accuracy: 0.745; specificity: 0.814). Feature importance analysis indicated that “age” was the most important variable in the Support Vector Machine (SVM) model, while “perceived health” was the most important variable in the Light Gradient Boosting Machine (LightGBM) model.ConclusionMachine learning methods can effectively identify key risk factors influencing subjective life expectancy among middle-aged and older adults, and provide valuable guidance for targeted health management strategies tailored to populations with different levels of physical activity.

oan default poses a significant threat to the sustainability of financial institutions, necessitating the development of intelligent, data-driven systems for early risk detection. This research presents a robust and interpretable machine learning framework for predicting loan default risk using a real-world financial dataset comprising 255,347 anonymized loan records with a pronounced class imbalance (11.6% default rate). To address the skewed class distribution, the Synthetic Minority Oversampling Technique (SMOTE) was applied to the training set, enhancing the model's sensitivity to defaulters. Four supervised learning algorithms Logistic Regression, Support Vector Machine (SVM), Random Forest, and Extreme Gradient Boosting (XGBoost) were implemented and rigorously evaluated using stratified 5-fold cross-validation. Performance metrics included Accuracy, Precision, Recall, F1-Score, and Area Under the Receiver Operating Characteristic Curve (AUC-ROC), with particular emphasis on metrics sensitive to class imbalance. Among the models tested, the Random Forest classifier achieved the best overall performance, attaining a test accuracy of 96.26%, F1-Score of 0.8014, and AUC-ROC of 0.9215, thereby offering a balanced and reliable prediction of default risk. To ensure model transparency and support regulatory compliance, HasMortgage, EmploymentType, and LoanPurpose as key drivers of default risk, aligning with domain knowledge and enhancing stakeholder trust. The study concludes that combining ensemble machine learning models with class imbalance handling and explainable AI techniques offers a practical and effective solution for credit risk assessment. Recommendations were made for financial institutions, data scientists, and policymakers to adopt interpretable, fair, and performance-optimized predictive systems. This research contributes to the growing body of literature on responsible AI in finance and lays a foundation for future advancements in ethical and data-driven credit decision-making.

Diagnosing a parasitic disease is a very difficult job in clinical practice. In this study, we constructed a machine learning model for diagnosis prediction using patient information. First, we diagnosed whether a patient has a parasitic disease. Next, we predicted the proper diagnosis method among the six types of diagnostic terms (biopsy, endoscopy, microscopy, molecular, radiology, and serology) if the patient has a parasitic disease. To make the datasets, we extracted patient information from PubMed abstracts from 1956 to 2019. We then used two datasets: the prediction for parasite-infected patient dataset (N = 8748) and the prediction for diagnosis method dataset (N = 3780). We then compared four machine learning models: support vector machine, random forest, multi-layered perceptron, and gradient boosting. To solve the data imbalance problem, the synthetic minority over-sampling technique and TomekLinks were used. In the parasite-infected patient dataset, the random forest, random forest with synthetic minority over-sampling technique, gradient boosting, gradient boosting with synthetic minority over-sampling technique, and gradient boosting with TomekLinks demonstrated the best performances (AUC: 79%). In predicting the diagnosis method dataset, gradient boosting with synthetic minority over-sampling technique was the best model (AUC: 87%). For the class prediction, gradient boosting demonstrated the best performances in biopsy (AUC: 88%). In endoscopy (AUC: 94%), molecular (AUC: 90%), and radiology (AUC: 88%), gradient boosting with synthetic minority over-sampling technique demonstrated the best performance. Random forest demonstrated the best performances in microscopy (AUC: 82%) and serology (AUC: 85%). We calculated feature importance using gradient boosting; age was the highest feature importance. In conclusion, this study demonstrated that gradient boosting with synthetic minority over-sampling technique can predict a parasitic disease and serve as a promising diagnosis tool for binary classification and multi-classification schemes.

<p>Classifying high-dimensional data are a challenging task in data mining. Gene expression data is a type of high-dimensional data that has thousands of features. The study was proposing a method to extract knowledge from high-dimensional gene expression data by selecting features and classifying. Lasso was used for selecting features and the classification and regression tree (CART) algorithm was used to construct the decision tree model. To examine the stability of the lasso decision tree, we performed bootstrap aggregating (Bagging) with 50 replications. The gene expression data used was an ovarian tumor dataset that has 1,545 observations, 10,935 gene features, and binary class. The findings of this research showed that the lasso decision tree could produce an interpretable model that theoretically correct and had an accuracy of 89.32%. Meanwhile, the model obtained from the majority vote gave an accuracy of 90.29% which showed an increase in accuracy of 1% from the single lasso decision tree model. The slightly increasing accuracy shows that the lasso decision tree classifier is stable.</p>

Anomaly detection in medical imaging is pivotal for early diagnosis and treatment planning. However, the inherent class imbalance in medical datasets poses significant challenges, often leading to biased models that underperform on minority classes. This study investigates the integration of the Synthetic Minority Over-sampling Technique (SMOTE) with various machine learning and deep learning models to enhance anomaly detection in medical images. By applying SMOTE to balance datasets and evaluating its impact across multiple models, we demonstrate improved detection accuracy, sensitivity, and specificity. The findings underscore the efficacy of SMOTE in addressing class imbalance, thereby enhancing the reliability of anomaly detection systems in medical imaging.

Diabetes is a chronic metabolic disorder characterized by persistent hyperglycemia. The globally rising prevalence of diabetes has made early and accurate diagnosis imperative to avoid long-term sequelae and decrease the cost burden on health facilities. Machine Learning (ML) has emerged as a highly effective tool of clinical science because of its capability of discovering complex patient data patterns and diagnostic performance optimization. This paper is a comparative performance evaluation of five ML models—Decision Tree (DT), Random Forest (RF), Logistic Regression (LR), K-Nearest Neighbors (KNN), and Multilayer Perceptron (MLP)—utilized for diabetes prediction. Data were processed using the “Healthcare Diabetes Dataset,” made up of eight commonly used clinical parameters. For making the models trustworthy, a strong data preprocessing pipeline was utilized, made up of outlier detection using the Interquartile Range (IQR), normalization of data, and class balancing using the Synthetic Minority Over-sampling Technique (SMOTE). Results reveal that the RF and DT models achieved the highest performance based on accuracy rates of 98.15% and 97.51%, respectively, though more moderate outcomes were recorded by LR and MLP. They reveal the remarkable potential of ML models, particularly ensemble-based ML models such as RF, at supporting early diagnosis of diabetes. When implemented complementarily to clinical decision-making processes, these models can serve a cost-effective and effective replacement of conventional diagnostic methods.

The rise of IoT technologies has increased the need for robust threat detection to address growing cyber-physical risks. Traditional machine learning (ML) models often struggle with imbalanced datasets, particularly in detecting rare threats. This study tackles this challenge using the "IoT_Modbus" dataset, a benchmark for multiclass cybersecurity classification, and explores advanced resampling techniques to enhance model performance. By combining methods like Synthetic Minority Oversampling Technique (SMOTE) and Adaptive Synthetic Sampling (ADASYN) with ensemble classifiers, Random Forest (RF), Gradient Boosting, XGBoost, LightGBM, and CatBoost, significant improvements in accuracy were achieved. RF outperformed others, reaching 99.06% accuracy with SMOTE and 99.00% with ADASYN. Among boosting-based models, XGBoost led with 86.21% (SMOTE) and 86.63% (ADASYN), while CatBoost followed closely. These results highlight the effectiveness of resampling in addressing class imbalance and improving predictive performance. Visual tools like heatmaps and accuracy charts further clarify these trends. This research provides cybersecurity professionals with a practical framework to enhance IoT threat detection and underscores the importance of resampling strategies for better classification outcomes. By guiding model selection for imbalanced datasets, it offers valuable insights for securing IoT systems in an interconnected world.