Automated facial expression recognition (FER) is an active research area due to its practical importance in a wide range of applications. In recent years, deep learning-based approaches have delivered promising performances in FER, leveraging the latest advances in computer vision. However, mask-wearing after the onset of the COVID-19 pandemic has posed challenges for the existing models when the salient features from the masked region are unavailable. This study investigates what effects facial masks will bring to expression detection using state-of-the-art deep learners. Specifically, we evaluate three deep neural networks in recognizing six emotional categories on masked facial images and compare the results to unmasked counterparts reported by prior studies. We based our work on the FER2013 dataset and augmented regular face images with artificial masks utilizing the Dlib and OpenCV libraries. Our experimental results indicate that deep learning models can be effective in recognizing some masked expressions (e.g., “Happy”, “Surprise”, and “Neural”) but fall short on the others (e.g., “Angry”, “Fear”, “Sad”). Furthermore, with the presence of facial masks, angry faces are most likely to be misclassified as neural, and fear is the most challenging emotion to detect.

The COVID-19 pandemic has caused unprecedented challenges to public health and disruption to everyday life. The news in 2020 was dominated by the worldwide spread of COVID-19, overwhelming healthcare providers and drastically changing people's lives. In 2021, the release of vaccines from multiple pharmaceutical companies changed the focus to ending the pandemic through mass inoculation. Nevertheless, the vaccine acceptance rate differs significantly across US counties, ranging from 99% to 0.1%. Our study investigates the principal risk factors in predicting COVID-19 infection and mortality rates at the county level during the early vaccination era. We are particularly interested in the role of vaccination in curbing the exacerbation of COVID-19. To this end, we first compare the efficacy of six established machine learning algorithms to predict county-level infection and mortality rates. Next, we perform risk factor analysis by identifying common principal predictors revealed by the models. Our experimental results suggest that vaccination plays an essential role in limiting COVID-19 infection and mortality. Furthermore, socioeconomic factors (e.g., severe housing problems and median household income) are more predictive of county-level mortality rate than intuitive features such as availability of healthcare resources (e.g., total numbers of hospitals/ICU beds/MDs). Our findings could provide additional insights to assist in COVID-19 resource allocation and priority setting.