Informed Deep Learning for Epidemics Forecasting

The SARS-CoV-2 pandemic has galvanized the interest of the scientific community toward methodologies apt at predicting the trend of the epidemiological curve, namely, the daily number of infected individuals in the population. One of the critical issues, is providing reliable predictions based on interventions enacted by policy-makers, which is of crucial relevance to assess their effectiveness. In this paper, we provide a novel data-driven application incorporating sub-symbolic knowledge to forecast the spreading of an epidemic depending on a set of interventions. More specifically, we focus on the embedding of classical epidemiological approaches, i.e., compartmental models, into Deep Learning models, to enhance the learning process and provide higher predictive accuracy. © 2022 The authors and IOS Press.

Does Time from COVID-19 Symptom Onset to Administration of Antispike Protein Monoclonal Antibody Predict Response?

Background. Casirivimab/imdevimab is a monoclonal antibody (mAb) cocktail with emergency use authorization for mild-to-moderate coronavirus disease 2019 (Covid-19) in patients at high risk for severe disease progression and/or hospitalization. Little is known about the importance of early administration of this product. The objective of this study was to determine if early administration (within 3 days of symptom onset) of casirivimab/imdevimab is associated with better outcomes. Methods. Single-center, retrospective cohort study including all consecutive patients who received casirivimab/imdevimab at our institution through May 2021. The primary outcome was 30-day post-infusion hospital admission rate in patients who received mAb ≥ 3 days (later) or < 3 days (early) in relation to patient reported symptom onset. Secondary outcomes included any hospital revisit within 30-days. Adverse events were also captured. Chi-square and independent samples t-test were used to compare categorical and continuous data, respectively. Multivariable logistic regression was used to adjust for confounders. Results. 270 patients met the inclusion criteria and were included in the analysis. There were 80 patients with early administration and 190 with later administration. Baseline characteristics for both groups were similar. Mean age was approximately 64 years and BMI 31 mg/m2. Table 1 provides a summary of patient characteristics. Late and early administration of casirivimab/imdevimab were similar in terms of hospital admission for any therapy related failure within 30 days of mAb administration after adjusting for age and Charlson comorbidity index (3.7% vs. 7.5%;adjusted odds ratio 0.69, 95% confidence interval, 0.20 -2.39;p=0.561). Similarly, there were no significant differences in any hospital revisit. Conclusion. We did not find any difference in outcomes between early and late administration of casirivimab/imdevimab.

CCTV: A new network-based methodology for the analysis and visualization of COVID-19 data

The novel COVID-19 pandemic has posed unprecedented challenges to the society and the health sector all over the globe. Here, we present a new network-based methodology to analyze COVID-19 data measures and its application on a real dataset. The goal of the methodology is to analyze set of homogeneous datasets (i.e. COVID-19 data in several regions) using a statistical test to find similar/dissimilar dataset, mapping such similarity information on a graph and then using community detection algorithm to visualize and analyze the initial dataset. The methodology and its implementation as R function are publicly available at https://github.com/mmilano87/analyzeC19D. We evaluated diverse Italian COVID-19 data made publicly available by the Italian Protezione Civile Department at https://github.com/pcm-dpc/COVID-19/ © 2021 IEEE.

A novel network-based methodology for analysis of COVID-19 data

The novel COVID-19 pandemic has posed unprecedented challenges to the society and the health sector all over the globe. Here, we present a new network-based methodology to analyze COVID-19 data measures and its application on a real dataset. The goal of the methodology is to analyze set of homogeneous datasets (i.e. COVID-19 data in several regions) using a statistical test to find similar/dissimilar dataset, mapping such similarity information on a graph and then using community detection algorithm to visualize and analyze the initial dataset. The methodology and its implementation as R function are publicly available at https://github.com/mmilano87/analyzeC19D. We evaluated diverse Italian COVID-19 data made publicly available by the Italian Protezione Civile Department at https://github.com/pcm-dpc/COVID-19/. We considered the data provided for each Italian region in two periods February 24-April 26, 2020 (1st wave), and September 28-November 29, 2020 (2nd wave) and then we compared two periods. Similarity matrices of Italian regions for ten COVID-19 data measures are built by using statistical analysis;then they are mapped to undirected networks. Each node represents an Italian region and an edge connects statistically similar regions. Finally, clusters of regions with similar behaviour were found using network-based community detection algorithms. Experiments depict the communities formed by Italian regions over time and the communities change with respect to the ten data measures and time. © 2021 Copyright for this paper by its authors.

Parallel Network Analysis and Communities Detection (PANC) Pipeline for the Analysis and Visualization of COVID-19 Data

A new coronavirus, causing a severe acute respiratory syndrome (COVID-19), was started at Wuhan, China, in December 2019. The epidemic has rapidly spread across the world becoming a pandemic that, as of today, has affected more than 70 million people causing over 2 million deaths. To better understand the evolution of spread of the COVID-19 pandemic, we developed PANC (Parallel Network Analysis and Communities Detection), a new parallel preprocessing methodology for network-based analysis and communities detection on Italian COVID-19 data. The goal of the methodology is to analyze set of homogeneous datasets (i.e. COVID-19 data in several regions) using a statistical test to find similar/dissimilar behaviours, mapping such similarity information on a graph and then using community detection algorithm to visualize and analyze the initial dataset. The methodology includes the following steps: (i) a parallel methodology to build similarity matrices that represent similar or dissimilar regions with respect to data;(ii) an effective workload balancing function to improve performance;(iii) the mapping of similarity matrices into networks where nodes represent Italian regions, and edges represent similarity relationships;(iv) the discovering and visualization of communities of regions that show similar behaviour. The methodology is general and can be applied to world-wide data about COVID-19, as well as to all types of data sets in tabular and matrix format. To estimate the scalability with increasing workloads, we analyzed three synthetic COVID-19 datasets with the size of 90.0MB, 180.0MB, and 360.0MB. Experiments was performed on showing the amount of data that can be analyzed in a given amount of time increases almost linearly with the number of computing resources available. Instead, to perform communities detection, we employed the real data set. © 2021 World Scientific Publishing Company.

Influence of bamlanivimab administration timing relative to symptoms on hospital revisit

Introduction: Bamlanivimab is a monoclonal antibody treatment for mild-to-moderate coronavirus disease 2019 (Covid-19) in patients at high risk for severe disease progression and/or hospitalization. Despite emerging evidence that bamlanivimab plus etesevimab decreases viral load more than monotherapy, there is insufficient evidence of bamlanivimab monotherapy's effects on 28-day all-cause hospital revisit and adverse drug reactions (ADRs) considering its widespread use. Research Question or Hypothesis: Does bamlanivimab administration within 3 days of symptom onset have a lower 30-day revisit rate versus later administration? Study Design: Single-center, retrospective cohort study Methods: The electronic medical record was queried for all consecutive patients who received bamlanivimab in a 2-month period. The primary outcome was 30-day post infusion revisit rate in patients who presented in < 3 days (early) versus ≥ 3 days (later) of symptom onset. Secondary outcomes included Covid-19- and ADR-related rates of revisit and 30-day hospital admission rate between groups. Chi-square and independent samples t-test were used to compare categorical and continuous data, respectively. Results: 183 patients met the inclusion criteria and were included in the analysis. There were 70 patients with early administration and 113 with later administration. Baseline characteristics for both groups were similar. The average age was 67 years and BMI 30 mg/m2;proportions of active smokers was roughly 4.5% and patients with diabetes were 30%. Early and late administration of bamlanivimab were alike in terms of any hospital revisit (21.4% vs. 22.1%;p=0.912). Similarly, there was no significant between group difference for COVID-19 or ADR related revisits as well as for Covid-19 hospital admission within 30 days. No variables predictive of 30-day hospital revisit were identified. Conclusion: We did not find any difference in outcomes between early and late administration of bamlanivimab. The hospital admission rate was similar to previous studies.