Earticle

Silicone hydrogel contact lenses provide high oxygen permeability, but the hydrophobic nature of silicone can reduce surface wettability during extended wear. In this study, DMA, NVP, HEMA, and TRIS based silicone hydrogel lenses were prepared with nanochitosan at 0.03–0.10 wt% and evaluated for optical properties, wettability, mechanical strength, elution stability, and antibacterial activity. UV–Vis measurements showed that UV-B transmittance decreased from 76.22% in the control lens to 36.21% at 0.10 wt%, while the refractive index remained stable at 1.4261~1.4263. Water content increased from 42.2% to 48.0%, and the static contact angle decreased from 104.9° to 87.2°. Tensile strength increased from 0.120 to 0.310 kgf/mm². Extract tests showed pH values of 7.0–7.5 and UV absorbance below 0.18, with a 2.3 mL difference in KMnO₄ consumption. Antibacterial tests against Escherichia coli and Staphylococcus aureus showed reduced colony formation with increasing nanochitosan content.

This study evaluates the impact of Korea’s major Free Trade Agreements (FTAs), particularly the Korea–China Free Trade Agreement (KCFTA) and the Korea–United States Free Trade Agreement (KORUS), on Korea’s export performance using econometric and machine learning approaches. Quarterly trade and macroeconomic data were collected from KOSIS, the IMF, the World Bank, and the National Bureau of Statistics of China (NBS). A Difference-in-Differences (DiD) framework was used to estimate relative treatment effects before and after FTA implementation, while Stepwise Regression, Ridge, Lasso, Elastic Net, and Random Forest models were applied for robustness and predictive analysis. The findings do not provide strong evidence that the FTAs generated additional positive export effects during the study period. Instead, GDP and exchange rates emerged as more significant determinants across most models. Among the predictive models, Random Forest demonstrated the strongest predictive performance with the lowest prediction errors. The study highlights the value of combining causal econometric analysis with machine learning techniques to evaluate trade agreement effectiveness from multiple analytical perspectives.

To investigate the effects of polyol-based surfactant types and processing conditions on the structure and physicochemical properties of emulsions, various emulsions were prepared using different surfactants and systematically evaluated. In addition, processing methods incorporating primary emulsification at elevated temperature followed by quenching and secondary emulsification were applied to compare the resulting microstructure and stability. Despite having identical compositions, either O/W or W/O emulsions were formed depending on the hydrophilic–lipophilic characteristics of the surfactants. Microscopic observations revealed that emulsions prepared by primary emulsification alone exhibited heterogeneous internal structures, whereas those subjected to quenching followed by secondary emulsification showed more uniform dispersion and improved structural stability. These differences in internal structure significantly affected viscosity and hysteresis behavior, with the secondary emulsification process leading to higher viscosity and more stable rheological properties. Furthermore, formulations prepared with sucrose- and sorbitan-based surfactants exhibited superior phase stability compared to those containing glycerol- and xylose-based surfactants. The crystallinity of the surfactants was also found to play a crucial role in determining the viscosity and textural properties of the emulsions.

Gaze estimation is essential in human–computer interaction, autonomous driving, and medical diagnosis, yet it suffers from susceptibility to environmental variations and poor generalization to unseen users. To address these challenges, we employ mixed-effects neural networks that capture both population-level (fixed) and individual-specific (random) patterns. However, existing models often rely on Gaussian assumptions, rendering them sensitive to outliers and incapable of inferring random effects for unseen clusters. We propose a t-distribution–based robust adversarially-regularized mixed effects neural network (RARMENN). The t-distribution enhances robustness against extreme values, while an adversarial classifier removes cluster-specific information from fixed effects, thereby enabling random-effect inference for new users. Experimental results on the MPIIGaze and UTMultiview datasets demonstrate up to a 16% improvement in within-subject accuracy and a 75% enhancement in LOOCV compared to Gaussian baselines, confirming significantly improved generalization and reliability for real-world gaze estimation.