A Bayesian convolutional neural network-based generalized linear model.

Convolutional neural networks (CNNs) provide flexible function approximations for a wide variety of applications when the input variables are in the form of images or spatial data. Although CNNs often outperform traditional statistical models in prediction accuracy, statistical inference, such as estimating the effects of covariates and quantifying the prediction uncertainty, is not trivial due to the highly complicated model structure and overparameterization. To address this challenge, we propose a new Bayesian approach by embedding CNNs within the generalized linear models (GLMs) framework. We use extracted nodes from the last hidden layer of CNN with Monte Carlo (MC) dropout as informative covariates in GLM. This improves accuracy in prediction and regression coefficient inference, allowing for the interpretation of coefficients and uncertainty quantification. By fitting ensemble GLMs across multiple realizations from MC dropout, we can account for uncertainties in extracting the features. We apply our methods to biological and epidemiological problems, which have both high-dimensional correlated inputs and vector covariates. Specifically, we consider malaria incidence data, brain tumor image data, and fMRI data. By extracting information from correlated inputs, the proposed method can provide an interpretable Bayesian analysis. The algorithm can be broadly applicable to image regressions or correlated data analysis by enabling accurate Bayesian inference quickly.

Technology-based interactive communication simulation addressing challenging communication situations for nursing students.

BACKGROUND: Simulation-based interventions for nursing students addressing challenging communication situations involving geriatric patients and end-of-life care are limited. PURPOSE: This study evaluated the effects of technology-based interactive communication simulations on nursing students' communication knowledge, self-efficacy, skills, compassion, and program satisfaction. METHOD: A randomized controlled repeated-measures design was used with third- and fourth-year nursing students enrolled in five nursing colleges located in five regions in Korea as participants. Participants were randomly assigned to either a technology-based interactive communication simulation or an attention control group. Changes in communication knowledge, self-efficacy, skills, compassion, and program satisfaction were assessed using three self-reported measures and communication skills were measured by the raters. Statistical analyses included descriptive analyses, chi-square tests, t-tests, and a generalized estimating equation model. RESULTS: Eighty students participated in one of the two programs, and 77 in the four-week follow-up test. The intervention group indicated significant improvements in communication knowledge, self-efficacy, skills, and compassion, as well as higher program satisfaction compared with the attention control group. Communication skills as assessed by raters also showed significant change at all assessment time points. CONCLUSION: The technology-based interactive communication simulation program is effective in improving communication skills among nursing students managing geriatric patients and end-of-life care.