Risk Management of COVID-19 in the Residential Educational Setting: Lessons Learned and Implications for Moving Forward

Monitoring and responding to emerging infectious diseases in a university setting: A case study using COVID-19

Emerging infection diseases (EIDs) are an increasing threat to global public health, especially when the disease is newly emerging. Institutions of higher education (IHEs) are particularly vulnerable to EIDs because student populations frequently share high-density residences and strongly mix with local and distant populations. In fall 2020, IHEs responded to a novel EID, COVID-19. Here, we describe Quinnipiac University's response to SARS-CoV-2 and evaluate its effectiveness through empirical data and model results. Using an agent-based model to approximate disease dynamics in the student body, the University established a policy of dedensification, universal masking, surveillance testing via a targeted sampling design, and app-based symptom monitoring. After an extended period of low incidence, the infection rate grew through October, likely due to growing incidence rates in the surrounding community. A super-spreader event at the end of October caused a spike in cases in November. Student violations of the University's policies contributed to this event, but lax adherence to state health laws in the community may have also contributed. The model results further suggest that the infection rate was sensitive to the rate of imported infections and was disproportionately impacted by non-residential students, a result supported by the observed data. Collectively, this suggests that campus-community interactions play a major role in campus disease dynamics. Further model results suggest that app-based symptom monitoring may have been an important regulator of the University's incidence, likely because it quarantined infectious students without necessitating test results. Targeted sampling had no substantial advantages over simple random sampling when the model incorporated contact tracing and app-based symptom monitoring but reduced the upper boundary on 90% prediction intervals for cumulative infections when either was removed. Thus, targeted sampling designs for surveillance testing may mitigate worst-case outcomes when other interventions are less effective. The results' implications for future EIDs are discussed.

COVID-Transformer: Interpretable COVID-19 Detection Using Vision Transformer for Healthcare

In the recent pandemic, accurate and rapid testing of patients remained a critical task in the diagnosis and control of COVID-19 disease spread in the healthcare industry. Because of the sudden increase in cases, most countries have faced scarcity and a low rate of testing. Chest X-rays have been shown in the literature to be a potential source of testing for COVID-19 patients, but manually checking X-ray reports is time-consuming and error-prone. Considering these limitations and the advancements in data science, we proposed a Vision Transformer-based deep learning pipeline for COVID-19 detection from chest X-ray-based imaging. Due to the lack of large data sets, we collected data from three open-source data sets of chest X-ray images and aggregated them to form a 30 K image data set, which is the largest publicly available collection of chest X-ray images in this domain to our knowledge. Our proposed transformer model effectively differentiates COVID-19 from normal chest X-rays with an accuracy of 98% along with an AUC score of 99% in the binary classification task. It distinguishes COVID-19, normal, and pneumonia patient's X-rays with an accuracy of 92% and AUC score of 98% in the Multi-class classification task. For evaluation on our data set, we fine-tuned some of the widely used models in literature, namely, EfficientNetB0, InceptionV3, Resnet50, MobileNetV3, Xception, and DenseNet-121, as baselines. Our proposed transformer model outperformed them in terms of all metrics. In addition, a Grad-CAM based visualization is created which makes our approach interpretable by radiologists and can be used to monitor the progression of the disease in the affected lungs, assisting healthcare.