Discriminative Ridge Machine: A Classifier for High-Dimensional Data or Imbalanced Data

Detection of late gadolinium enhancement in patients with hypertrophic cardiomyopathy using machine learning

BACKGROUND

Late gadolinium enhancement (LGE) on cardiac magnetic resonance (CMR) in hypertrophic cardiomyopathy (HCM) typically represents myocardial fibrosis and may lead to fatal ventricular arrhythmias. However, CMR is resource-intensive and sometimes contraindicated. Thus, in patients with HCM, we aimed to detect LGE on CMR by applying machine learning (ML) algorithm to clinical parameters.

METHODS AND RESULTS

In this trans-Pacific multicenter study of HCM, a ML model was developed to distinguish the presence or absence of LGE on CMR by ridge classification method using 22 clinical parameters including 9 echocardiographic data. Among 742 patients in this cohort, the ML model was constructed in 2 institutions in the United States (training set, n = 554) and tested using data from an institution in Japan (test set, n = 188). LGE was detected in 299 patients (54%) in the training set and 76 patients (40%) in the test set. In the test set, the area under the receiver-operating-characteristic curve (AUC) of the ML model derived from the training set was 0.77 (95% confidence interval [CI] 0.70-0.84). When compared with a reference model constructed with 3 conventional risk factors for LGE on CMR (AUC 0.69 [95% CI 0.61-0.77]), the ML model outperformed the reference model (DeLong's test P = 0.01).

CONCLUSIONS

This trans-Pacific study demonstrates that ML analysis of clinical parameters can distinguish the presence of LGE on CMR in patients with HCM. Our ML model would help physicians identify patients with HCM in whom the pre-test probability of LGE is high, and therefore CMR will have higher utility.

Representation of Imprecision in Deep Neural Networks for Image Classification

Quantification and reduction of uncertainty in deep-learning techniques have received much attention but ignored how to characterize the imprecision caused by such uncertainty. In some tasks, we prefer to obtain an imprecise result rather than being willing or unable to bear the cost of an error. For this purpose, we investigate the representation of imprecision in deep-learning (RIDL) techniques based on the theory of belief functions (TBF). First, the labels of some training images are reconstructed using the learning mechanism of neural networks to characterize the imprecision in the training set. In the process, a label assignment rule is proposed to reassign one or more labels to each training image. Once an image is assigned with multiple labels, it indicates that the image may be in an overlapping region of different categories from the feature perspective or the original label is wrong. Second, those images with multiple labels are rechecked. As a result, the imprecision (multiple labels) caused by the original labeling errors will be corrected, while the imprecision caused by insufficient knowledge is retained. Images with multiple labels are called imprecise ones, and they are considered to belong to meta-categories, the union of some specific categories. Third, the deep network model is retrained based on the reconstructed training set, and the test images are then classified. Finally, some test images that specific categories cannot distinguish will be assigned to meta-categories to characterize the imprecision in the results. Experiments based on some remarkable networks have shown that RIDL can improve accuracy (AC) and reasonably represent imprecision both in the training and testing sets.

Artificial intelligence-driven predictive framework for early detection of still birth

Predictive modeling is becoming increasingly popular in the context of early disease detection. The use of machine learning approaches for predictive modeling can help early detection of diseases thereby enabling medical experts to appropriate medical treatments. Stillbirth prediction is a similar domain where artificial intelligence-based predictive modeling can alleviate this significant global health challenge. Despite advancements in prenatal care, the prevention of stillbirths remains a complex issue requiring further research and interventions. The cardiotocography (CTG) dataset is used in this research work from the UCI machine learning (ML) repository to investigate the efficiency of the proposed approach regarding stillbirth prediction. This research work adopts the Tabular Prior Data Fitted Network (TabPFN) model which was originally designed to solve small tabular classification. TabPFN is used to predict the still or live birth during pregnancy with 97.91% accuracy. To address this life-saving problem with more accurate results and in-depth analysis of ML models, this research work makes use of 13 renowned ML models for performance comparison with the proposed model. The proposed model is evaluated using precision, recall, F-score, Mathews Correlation Coefficient (MCC), and the area under the curve evaluation parameters and the results are 97.87%, 98.26%, 98.05%, 96.42%, and 98.88%, respectively. The results of the proposed model are further evaluated using k-fold cross-validation and its performance comparison with other state-of-the-art studies indicating the superior performance of TabPFN model.

Herding towards pygmalion: Examining the cultural dimension of market and bank based systems

The study aims to not only detect the presence of herd behavior in the countries studied, but also to examine the effect of cultural dimensions and market/bank-based systems on the herding behavior of financial market investors. The study employs the Cross-Sectional Standard Deviation and Cross-Sectional Absolute Deviation methods to analyze daily data from public companies traded in the capital markets in Emerging Seven and Group of Seven economies. The results suggest that being a member of E7-G7, a Future Oriented (FO), and a Performance Oriented (PO) cultures are the most important factors in explaining herd behavior. Additionally, the study found that the Ridge Classifier and CatBoost Classifier algorithms arethe most superior model for estimating herd behavior periods determined by the CSSD and CASD models, respectively. The feature selection results show that the Assertiveness (A) in-group collectivism (GC) are the three most important explanatory factors of the herd behavior.

A Comprehensive Framework for Long-Tailed Learning via Pretraining and Normalization

Data in the visual world often present long-tailed distributions. However, learning high-quality representations and classifiers for imbalanced data is still challenging for data-driven deep learning models. In this work, we aim at improving the feature extractor and classifier for long-tailed recognition via contrastive pretraining and feature normalization, respectively. First, we carefully study the influence of contrastive pretraining under different conditions, showing that current self-supervised pretraining for long-tailed learning is still suboptimal in both performance and speed. We thus propose a new balanced contrastive loss and a fast contrastive initialization scheme to improve previous long-tailed pretraining. Second, based on the motivative analysis on the normalization for classifier, we propose a novel generalized normalization classifier that consists of generalized normalization and grouped learnable scaling. It outperforms traditional inner product classifier as well as cosine classifier. Both the two components proposed can improve recognition ability on tail classes without the expense of head classes. We finally build a unified framework that achieves competitive performance compared with state of the arts on several long-tailed recognition benchmarks and maintains high efficiency.

Optimizing HCV Disease Prediction in Egypt: The hyOPTGB Framework

The paper focuses on the hepatitis C virus (HCV) infection in Egypt, which has one of the highest rates of HCV in the world. The high prevalence is linked to several factors, including the use of injection drugs, poor sterilization practices in medical facilities, and low public awareness. This paper introduces a hyOPTGB model, which employs an optimized gradient boosting (GB) classifier to predict HCV disease in Egypt. The model's accuracy is enhanced by optimizing hyperparameters with the OPTUNA framework. Min-Max normalization is used as a preprocessing step for scaling the dataset values and using the forward selection (FS) wrapped method to identify essential features. The dataset used in the study contains 1385 instances and 29 features and is available at the UCI machine learning repository. The authors compare the performance of five machine learning models, including decision tree (DT), support vector machine (SVM), dummy classifier (DC), ridge classifier (RC), and bagging classifier (BC), with the hyOPTGB model. The system's efficacy is assessed using various metrics, including accuracy, recall, precision, and F1-score. The hyOPTGB model outperformed the other machine learning models, achieving a 95.3% accuracy rate. The authors also compared the hyOPTGB model against other models proposed by authors who used the same dataset.

Global and Local Similarity Learning in Multi-Kernel Space for Nonnegative Matrix Factorization

Most of existing nonnegative matrix factorization (NMF) methods do not fully exploit global and local similarity information from data. In this paper, we propose a novel local similarity learning approach in the convex NMF framework, which encourages inter-class separability that is desired for clustering. Thus, the new model is capable of enhancing intra-class similarity and inter-class separability with simultaneous global and local learning. Moreover, the model learns the factor matrices in an augmented kernel space, which is a convex combination of pre-defined kernels with auto-learned weights. Thus, the learnings of cluster structure, representation factor matrix, and the optimal kernel mutually enhance each other in a seamlessly integrated model, which leads to informative representation. Multiplicative updating rules are developed with theoretical convergence guarantee. Extensive experimental results have confirmed the effectiveness of the proposed model.

Clustering of Multilayer Networks Using Joint Learning Algorithm With Orthogonality and Specificity of Features

Complex systems in nature and society consist of various types of interactions, where each type of interaction belongs to a layer, resulting in the so-called multilayer networks. Identifying specific modules for each layer is of great significance for revealing the structure-function relations in multilayer networks. However, the available approaches are criticized undesirable because they fail to explicitly the specificity of modules, and balance the specificity and connectivity of modules. To overcome these drawbacks, we propose an accurate and flexible algorithm by joint learning matrix factorization and sparse representation (jMFSR) for specific modules in multilayer networks, where matrix factorization extracts features of vertices and sparse representation discovers specific modules. To exploit the discriminative latent features of vertices in multilayer networks, jMFSR incorporates linear discriminant analysis (LDA) into non-negative matrix factorization (NMF) to learn features of vertices that distinguish the categories. To explicitly measure the specificity of features, jMFSR decomposes features of vertices into common and specific parts, thereby enhancing the quality of features. Then, jMFSR jointly learns feature extraction, common-specific feature factorization, and clustering of multilayer networks. The experiments on 11 datasets indicate that jMFSR significantly outperforms state-of-the-art baselines in terms of various measurements.

Assessing Biomaterial-Induced Stem Cell Lineage Fate by Machine Learning-Based Artificial Intelligence

Current functional assessment of biomaterial-induced stem cell lineage fate in vitro mainly relies on biomarker-dependent methods with limited accuracy and efficiency. Here a "Mesenchymal stem cell Differentiation Prediction (MeD-P)" framework for biomaterial-induced cell lineage fate prediction is reported. MeD-P contains a cell-type-specific gene expression profile as a reference by integrating public RNA-seq data related to tri-lineage differentiation (osteogenesis, chondrogenesis, and adipogenesis) of human mesenchymal stem cells (hMSCs) and a predictive model for classifying hMSCs differentiation lineages using the k-nearest neighbors (kNN) strategy. It is shown that MeD-P exhibits an overall accuracy of 90.63% on testing datasets, which is significantly higher than the model constructed based on canonical marker genes (80.21%). Moreover, evaluations of multiple biomaterials show that MeD-P provides accurate prediction of lineage fate on different types of biomaterials as early as the first week of hMSCs culture. In summary, it is demonstrated that MeD-P is an efficient and accurate strategy for stem cell lineage fate prediction and preliminary biomaterial functional evaluation.

An Accelerated Maximally Split ADMM for a Class of Generalized Ridge Regression

Ridge regression (RR) has been commonly used in machine learning, but is facing computational challenges in big data applications. To meet the challenges, this article develops a highly parallel new algorithm, i.e., an accelerated maximally split alternating direction method of multipliers (A-MS-ADMM), for a class of generalized RR (GRR) that allows different regularization factors for different regression coefficients. Linear convergence of the new algorithm along with its convergence ratio is established. Optimal parameters of the algorithm for the GRR with a particular set of regularization factors are derived, and a selection scheme of the algorithm parameters for the GRR with general regularization factors is also discussed. The new algorithm is then applied in the training of single-layer feedforward neural networks. Experimental results on performance validation on real-world benchmark datasets for regression and classification and comparisons with existing methods demonstrate the fast convergence, low computational complexity, and high parallelism of the new algorithm.

Hybrid Feature Extraction Model to Categorize Student Attention Pattern and Its Relationship with Learning

The increase of instructional technology, e-learning resources, and online courses has created opportunities for data mining and learning analytics in the pedagogical domain. A large amount of data is obtained from this domain that can be analyzed and interpreted so that educators can understand students’ attention. In a classroom where students have their own computers in front of them, it is important for instructors to understand whether students are paying attention. We collected on- and off-task data to analyze the attention behaviors of students. Educational data mining extracts hidden information from educational records, and we are using it to classify student attention patterns. A hybrid method is used to combine various techniques like classifications, regressions, or feature extraction. In our work, we combined two feature extraction techniques: principal component analysis and linear discriminant analysis. Extracted features are used by a linear and kernel support vector machine (SVM) to classify attention patterns. Classification results are compared with linear and kernel SVM. Our hybrid method achieved the best results in terms of accuracy, precision, recall, F1, and kappa. Also, we correlated attention with learning. Here, learning corresponds to tests and a final course grade. For determining the correlation between grades and attention, Pearson’s correlation coefficient and p-value were used.

Modeling and Numerical Validation for an Algorithm Based on Cellular Automata to Reduce Noise in Digital Images

Given the grid features of digital images, a direct relation with cellular automata can be established with transition rules based on information of the cells in the grid. This document presents the modeling of an algorithm based on cellular automata for digital images processing. Using an adaptation mechanism, the algorithm allows the elimination of impulsive noise in digital images. Additionally, the comparison of the cellular automata algorithm and median and mean filters is carried out to observe that the adaptive process obtains suitable results for eliminating salt and pepper type-noise. Finally, by means of examples, the result of the algorithm are shown graphically.