Search Results (3)

Predicting length of stay (LoS) and understanding its underlying factors is essential to minimizing the risk of hospital-acquired conditions, improving financial, operational, and clinical outcomes, and better managing future pandemics. The purpose of this study was to forecast patients’ LoS using a deep learning model and to analyze cohorts of risk factors reducing or prolonging LoS. We employed various preprocessing techniques, SMOTE-N to balance data, and a TabTransformer model to forecast LoS. Finally, the Apriori algorithm was applied to analyze cohorts of risk factors influencing hospital LoS. The TabTransformer outperformed the base machine learning models in terms of F1 score (0.92), precision (0.83), recall (0.93), and accuracy (0.73) for the discharged dataset and F1 score (0.84), precision (0.75), recall (0.98), and accuracy (0.77) for the deceased dataset. The association mining algorithm was able to identify significant risk factors/indicators belonging to laboratory, X-ray, and clinical data, such as elevated LDH and D-dimer levels, lymphocyte count, and comorbidities such as hypertension and diabetes. It also reveals what treatments have reduced the symptoms of COVID-19 patients, leading to a reduction in LoS, particularly when no vaccines or medication, such as Paxlovid, were available. Full article

E-commerce has increased online credit card usage nowadays. Similarly, credit card transactions have increased for physical sales and purchases. This has increased the risk of credit card fraud (CCF) and made payment networks more vulnerable. Therefore, there is a need to develop a precise CCF detector to control such online fraud. Previously, many studies have been presented on CCF detection and gave good results and performance. However, these solutions still lack performance, and most of them have ignored the outlier problem before applying feature selection and oversampling techniques to give solutions for classification. The class imbalance problem is most prominent in available datasets of credit card transactions. Therefore, the proposed study applies preprocessing to clean the feature set at first. Then, outliers are detected and normalized using the IQR method. This outlier normalizes data fed to the Shapiro method for feature ranking and the 20 most prominent features are selected. This selected feature set is then fed to the SMOTEN oversampling method, which increases the minority class instances and equalizes the positive and negative instances. Next, this cleaned feature set is then fed to five ML classifiers, and four different splits of holdout validation are applied. There are two experiments conducted in which, firstly, the original data are fed to five ML classifiers and the holdout validation technique is used, in which the AUC reaches a maximum of 0.971. In Experiment 2, outliers are normalized, features are selected using the Shapiro method, and oversampling is performed using the SMOTEN method. This normalized and processed feature set is fed to five ML classifiers via holdout validation methods. The experimental results show a 1.00 AUC compared with state-of-the-art studies, which proves that the proposed study achieves better results using this specific framework. Full article

Machine Learning (ML) methods have become important for enhancing the performance of decision-support predictive models. However, class imbalance is one of the main challenges for developing ML models, because it may bias the learning process and the model generalization ability. In this paper, we consider oversampling methods for generating synthetic categorical clinical data aiming to improve the predictive performance in ML models, and the identification of risk factors for cardiovascular diseases (CVDs). We performed a comparative study of several categorical synthetic data generation methods, including Synthetic Minority Oversampling Technique Nominal (SMOTEN), Tabular Variational Autoencoder (TVAE) and Conditional Tabular Generative Adversarial Networks (CTGANs). Then, we assessed the impact of combining oversampling strategies and linear and nonlinear supervised ML methods. Lastly, we conducted a post-hoc model interpretability based on the importance of the risk factors. Experimental results show the potential of GAN-based models for generating high-quality categorical synthetic data, yielding probability mass functions that are very close to those provided by real data, maintaining relevant insights, and contributing to increasing the predictive performance. The GAN-based model and a linear classifier outperform other oversampling techniques, improving the area under the curve by 2%. These results demonstrate the capability of synthetic data to help with both determining risk factors and building models for CVD prediction. Full article