Predicting Infections in the Covid-19 Pandemic—Lessons Learned

Throughout the Covid-19 pandemic, a significant amount of effort had been put into developing techniques that predict the number of infections under various assumptions about the public policy and non-pharmaceutical interventions. While both the available data and the sophistication of the AI models and available computing power exceed what was available in previous years, the overall success of prediction approaches was very limited. In this paper, we start from prediction algorithms proposed for XPrize Pandemic Response Challenge and consider several directions that might allow their improvement. Then, we investigate their performance over medium-term predictions extending over several months. We find that while augmenting the algorithms with additional information about the culture of the modeled region, incorporating traditional compartmental models and up-to-date deep learning architectures can improve the performance for short term predictions, the accuracy of medium-term predictions is still very low and a significant amount of future research is needed to make such models a reliable component of a public policy toolbox. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Predicting Infections in the Covid-19 Pandemic—Lessons Learned

Throughout the Covid-19 pandemic, a significant amount of effort had been put into developing techniques that predict the number of infections under various assumptions about the public policy and non-pharmaceutical interventions. While both the available data and the sophistication of the AI models and available computing power exceed what was available in previous years, the overall success of prediction approaches was very limited. In this paper, we start from prediction algorithms proposed for XPrize Pandemic Response Challenge and consider several directions that might allow their improvement. Then, we investigate their performance over medium-term predictions extending over several months. We find that while augmenting the algorithms with additional information about the culture of the modeled region, incorporating traditional compartmental models and up-to-date deep learning architectures can improve the performance for short term predictions, the accuracy of medium-term predictions is still very low and a significant amount of future research is needed to make such models a reliable component of a public policy toolbox. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Do Hemodialysis Patients Need Immune Boosting with Vitamin, Mineral, and Probiotic Supplementation during COVID-19 Pandemic?

Coronavirus disease 2019 (COVID-19) has been accepted as a global pandemic, and poses a greater risk to the elderly and those with comorbidities. Comorbid diseases (particularly end-stage kidney disease with hemodialysis) and impaired immunity place patients in the high-risk group for COVID-19. In recent studies, it was also mentioned that exaggerated inflammation and a cytokine storm were the underlying causes related to the high mortality in COVID-19 patients. Currently, treatment modalities to balance the immune system of such vulnerable patient groups are essential, to protect them from the disease. Several vitamins (like vitamins C, D, and E), trace elements like zinc, and probiotics have been proposed as immune boosters to protect and combat infectious conditions. It is well known that these vitamins and elements are insufficient in hemodialysis patients. In this review, we aimed to evaluate the immune-boosting mechanisms of vitamins C, D, E, zinc, and probiotics, the studies related to their beneficial effects against infections, and their possible benefits for hemodialysis patients during the COVID-19 pandemic.

Bridging Predictive Analytics and Mobile Crowdsensing for Future Risk Maps of Communities against COVID-19

Crowd monitoring and management is an important application of Mobile Crowdsensing (MCS). The emergence of COVID-19 pandemic has made the modeling and simulation of community infection spread a vital activity in the battle against the disease. This paper provides insights for the utility of MCS to inform the decision support systems combating the pandemic. We present an MCS-driven community risk modeling solution against COVID-19 pandemic with the support of smart mobile device users (i.e., MCS participants), who opt-in to crowdsensing campaigns and grant access to their mobile device's built-in sensors (including GPS). Each community is defined by the spatio-temporal instances of MCS participants that are clustered based on the projected future movements of these participants. The MCS platform keeps track of the mobility patterns of the participants and utilizes unsupervised machine learning (ML) algorithms, more specifically k-means, Hidden Markov Model (HMM), and Expectation Maximization (EM) to predict a risk score of COVID-19 community spread for each community ahead of time. Through numerical results from simulating a metropolitan area (e.g., Paris), it is shown that communities? COVID-19 risk scores at the end of a set of MCS campaign can be predicted 20% ahead of time (i.e., upon completion of 80% of the MCS time commitments) with a dependability score up to 0.96 and an average of 0.93. Further tests with a larger population of participants show that community risk scores can be predicted 20% ahead of time with a dependability score up to 0.99 and an average of 0.98. © 2020 ACM.