Alessandro Vespignani’s research while affiliated with Northeastern University and other places


Ad

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (256)


Fig. 1. Epidemiological situation of COVID-19 in Mexico. A) Map of cumulative cases per 100,000 people, as of 2020 September 1. B) Timeline of new cases per 100,000 population at the state level (7-day rolling average), highlighting the 15 states with the most severe cumulative outbreaks. C) Number of municipalities that reported confirmed cases of COVID-19 through time. D) Age and sex distributions of confirmed COVID-19 cases across Mexico, highlighting "early" and "late" periods during which the relative risk of infections were calculated. E) Age and sex relative risk ratios of infection, comparing the early vs. late periods from (D).
Spatial scales of COVID-19 transmission in Mexico
  • Article
  • Full-text available

September 2024

·

84 Reads

PNAS Nexus

Brennan Klein

·

Harrison Hartle

·

Munik Shrestha

·

[...]

·

Moritz U G Kraemer

During outbreaks of emerging infectious diseases, internationally connected cities often experience large and early outbreaks, while rural regions follow after some delay. This hierarchical structure of disease spread is influenced primarily by the multiscale structure of human mobility. However, during the COVID-19 epidemic, public health responses typically did not take into consideration the explicit spatial structure of human mobility when designing nonpharmaceutical interventions (NPIs). NPIs were applied primarily at national or regional scales. Here, we use weekly anonymized and aggregated human mobility data and spatially highly resolved data on COVID-19 cases at the municipality level in Mexico to investigate how behavioral changes in response to the pandemic have altered the spatial scales of transmission and interventions during its first wave (March–June 2020). We find that the epidemic dynamics in Mexico were initially driven by exports of COVID-19 cases from Mexico State and Mexico City, where early outbreaks occurred. The mobility network shifted after the implementation of interventions in late March 2020, and the mobility network communities became more disjointed while epidemics in these communities became increasingly synchronized. Our results provide dynamic insights into how to use network science and epidemiological modeling to inform the spatial scale at which interventions are most impactful in mitigating the spread of COVID-19 and infectious diseases in general.

Download

Figure 3: Models performance in time and by horizon. A) Comparison of the average WIS of different models to the baseline across different forecast rounds. The ensemble model is highlighted in red. The background displays the reported ILI incidence for the corresponding weeks. B) On the left, we show the absolute WIS values of the Ensemble for different forecasting horizons (from 1 to 4 weeks ahead). In the inset, the figure also shows the median WIS relative to the baseline model by horizons. On the right, we repeat the analysis considering the absolute error of the median as a performance metric. The box boundaries represent the interquartile range (IQR), the line inside the box indicates the median and the whiskers extend to 1.5 times the IQR from the quartiles.
Collaborative forecasting of influenza-like illness in Italy: the Influcast experience

September 2024

·

16 Reads

Collaborative hubs that integrate multiple teams to generate ensemble projections and forecasts for shared targets are now regarded as state-of-the-art in epidemic predictive modeling. In this paper, we introduce Influcast, Italy's first epidemic forecasting hub for influenza-like illness. During the 2023/2024 winter season, Influcast provided 20 rounds of forecasts, involving five teams and eight models to predict influenza-like illness incidence up to four weeks in advance at the national and regional administrative level. The individual forecasts were synthesized into an ensemble and benchmarked against a baseline model. The ensemble forecasts consistently outperformed both individual models and baseline forecasts, demonstrating superior accuracy at national and sub-national levels across various metrics. Despite a decline in absolute performance over longer horizons, the ensemble model outperformed the baseline in all considered time frames. These findings underscore the importance of multimodel forecasting hubs in producing consistent short-term influenza-like illnesses forecasts that can inform public health preparedness and mitigation strategies.


Figure 3: Changing the transmission dynamics predictably affects the time to first detection. We use the same baseline WWSN as in Fig. 1 and the same detectable period. Unless specified, we keep an average reproduction number of 2, a generation time of 4 days, and a 16% detection rate at sentinels. All prediction intervals are obtained with n = 3244 subpopulations. (A) T fd from all origins, with varying reproduction number, generation time, and detection rate. The center line of the box plot indicates the median, the box covers the interquartile range and the whiskers cover the 90% central prediction interval; the outliers outside the interval are not shown. (B-C) We vary the generation time between 4 and 36 days, resulting in doubling times between 3.4 and 26.2 days. (B) T fd and T fd /T 2 + log 2 T 2 as a function of the doubling time. Circles indicate the median and the error bars cover the interquartile range. The dashed lines are there to guide the eyes. (C) Distributions of T fd and T fd /T 2 + log 2 T 2 over all origins for different doubling times. We use kernel density estimates (KDE) for the distributions to improve the visualization.
Optimization and performance analytics of global aircraft-based wastewater surveillance networks

August 2024

·

26 Reads

·

1 Citation

Aircraft wastewater surveillance has been proposed as a novel approach to monitor the global spread of pathogens. Here we develop a computational framework to provide actionable information for designing and estimating the effectiveness of global aircraft-based wastewater surveillance networks (WWSNs). We study respiratory diseases of varying transmission potentials and find that networks of 10 to 20 strategically placed wastewater sentinel sites can provide timely situational awareness and function effectively as an early warning system. The model identifies potential blind spots and suggests optimization strategies to increase WWSNs effectiveness while minimizing resource use. Our findings highlight that increasing the number of sentinel sites beyond a critical threshold does not proportionately improve WWSNs capabilities, stressing the importance of resource optimization. We show through retrospective analyses that WWSNs can significantly shorten the detection time for emerging pathogens. The presented approach offers a realistic analytic framework for the analysis of WWSNs at airports.


National incident weekly hospital admissions and select forecasts
National weekly observed hospitalizations (black points) along with FluSight ensemble forecasts for four weeks of submissions in the 2021–22 season (a) and seven weeks of submissions in the 2022-23 season (b). The median FluSight ensemble forecast values (blue points) are shown with the corresponding 50%, 80%, and 95% prediction intervals (blue shaded regions). c–e Show national incident weekly hospital admissions (black points) from the 2022-23 season and predictions from all models submitted on November 11, 2022 (c), December 05, 2022 (d) and February 27, 2023 (e). Colored bands indicate 95% prediction intervals for each model. Team forecasts for additional weeks are available in an interactive dashboard¹².
Standardized rank by season
Standardized rank of weighted interval score (WIS) over all forecast jurisdictions and horizons (1- to 4-week ahead), for the FluSight ensemble and each team submitting at least 75% of the forecast targets (see Table 1 for qualifying teams and season metrics) for the 2021–22 (a) and 2022–23 (b) seasons.
Relative WIS by state and model. State-level WIS values for each team relative to the FluSight baseline model
The range of Relative WIS values below 1, in blue, indicate better performance than the FluSight baseline (white). Relative WIS values above 1, in red, indicate poor performance relative to the FluSight baseline. Teams are ordered on horizontal axis from lowest to highest Relative WIS values for each season, 2021–22 (a) and 2022–23 (b). Analogous jurisdiction-specific relative WIS scores on log transformed counts are displayed in Supplementary Fig. 7.
WIS by model
Time series of log transformed absolute WIS for state and territory targets. Note that the forecast evaluation period translates to 1-week ahead forecast target end dates from February 26–June 25, 2022 (a), and October 22, 2022, to May 20, 2023 (b), and 4-week ahead forecast target end dates from March 19–July 16, 2022 (c), and November 5, 2022–June 10, 2023 (d). Weekly results for the FluSight baseline and ensemble models are shown in red and blue respectively. Results for individual contributing models are shown in light gray.
Coverage by model
1 and 4-week ahead 95% coverage for state and territory targets. Note that the forecast evaluation period translates to 1-week ahead forecast target end dates from February 26–June 25, 2022 (a), and October 22, 2022–May 20, 2023 (b), and 4-week ahead forecast target end dates from March 19–July 16, 2022 (c), and November 5, 2022–June 10, 2023 (d). Weekly results for the FluSight baseline and ensemble models are shown in red and blue, respectively. Results for individual contributing models are shown in light gray.
Evaluation of FluSight influenza forecasting in the 2021–22 and 2022–23 seasons with a new target laboratory-confirmed influenza hospitalizations

July 2024

·

186 Reads

·

5 Citations

Nature Communications

Accurate forecasts can enable more effective public health responses during seasonal influenza epidemics. For the 2021–22 and 2022–23 influenza seasons, 26 forecasting teams provided national and jurisdiction-specific probabilistic predictions of weekly confirmed influenza hospital admissions for one-to-four weeks ahead. Forecast skill is evaluated using the Weighted Interval Score (WIS), relative WIS, and coverage. Six out of 23 models outperform the baseline model across forecast weeks and locations in 2021–22 and 12 out of 18 models in 2022–23. Averaging across all forecast targets, the FluSight ensemble is the 2nd most accurate model measured by WIS in 2021–22 and the 5th most accurate in the 2022–23 season. Forecast skill and 95% coverage for the FluSight ensemble and most component models degrade over longer forecast horizons. In this work we demonstrate that while the FluSight ensemble was a robust predictor, even ensembles face challenges during periods of rapid change.


Fig. 1 Users' choices on online platforms generate data used to train recommenders. These recommenders then offer suggestions to users, influencing their choices, which in turn generate more data for re-training recommenders. This iterative process creates a potentially endless feedback loop.
Human-AI Coevolution

May 2024

·

864 Reads

·

1 Citation

Human-AI coevolution, defined as a process in which humans and AI algorithms continuously influence each other, increasingly characterises our society, but is understudied in artificial intelligence and complexity science literature. Recommender systems and assistants play a prominent role in human-AI coevolution, as they permeate many facets of daily life and influence human choices on online platforms. The interaction between users and AI results in a potentially endless feedback loop, wherein users' choices generate data to train AI models, which, in turn, shape subsequent user preferences. This human-AI feedback loop has peculiar characteristics compared to traditional human-machine interaction and gives rise to complex and often "unintended" social outcomes. This paper introduces Coevolution AI as the cornerstone for a new field of study at the intersection between AI and complexity science focused on the theoretical, empirical, and mathematical investigation of the human-AI feedback loop. In doing so, we: (i) outline the pros and cons of existing methodologies and highlight shortcomings and potential ways for capturing feedback loop mechanisms; (ii) propose a reflection at the intersection between complexity science, AI and society; (iii) provide real-world examples for different human-AI ecosystems; and (iv) illustrate challenges to the creation of such a field of study, conceptualising them at increasing levels of abstraction, i.e., technical, epistemological, legal and socio-political.


(A) Daily maximum temperature (°C) for each day from December 1, 2017, to May 30, 2018, the seasonal trend of the temperatures, daily variation between the maximum temperature, and seasonal trend. (B) Estimated daily number of total contacts for each week from December 1, 2017, to May 30, 2018. The line and shaded area represent the mean and 95% PrI of the mean daily values for each season, respectively.
(A) Estimated weekly incidence of ILI⁺ infections during the 2017–2018 influenza season in Shanghai, China. Dots represent the observed data; line and shaded area represent the mean and 95% CI of model simulations. ILI⁺ refers to the estimated weekly number of ILI cases that tested positive for influenza A(H1N1)pdm09. (B) Estimated posterior distribution of weekly net reproduction number. The line and shaded area represent the mean and 95% CI, respectively. (C) Posterior distribution of the final infection attack rate for the 2017–2018 influenza season. The boxplot reports quantiles 0.025, 0.25, 0.5, 0.75, and 0.975 of the distribution.
Evaluating Seasonal Variations in Human Contact Patterns and Their Impact on the Transmission of Respiratory Infectious Diseases

May 2024

·

44 Reads

·

3 Citations

Background Human contact patterns are a key determinant driving the spread of respiratory infectious diseases. However, the relationship between contact patterns and seasonality as well as their possible association with the seasonality of respiratory diseases is yet to be clarified. Methods We investigated the association between temperature and human contact patterns using data collected through a cross‐sectional diary‐based contact survey in Shanghai, China, between December 24, 2017, and May 30, 2018. We then developed a compartmental model of influenza transmission informed by the derived seasonal trends in the number of contacts and validated it against A(H1N1)pdm09 influenza data collected in Shanghai during the same period. Results We identified a significant inverse relationship between the number of contacts and the seasonal temperature trend defined as a spline interpolation of temperature data (p = 0.003). We estimated an average of 16.4 (95% PrI: 15.1–17.5) contacts per day in December 2017 that increased to an average of 17.6 contacts (95% PrI: 16.5–19.3) in January 2018 and then declined to an average of 10.3 (95% PrI: 9.4–10.8) in May 2018. Estimates of influenza incidence obtained by the compartmental model comply with the observed epidemiological data. The reproduction number was estimated to increase from 1.24 (95% CI: 1.21–1.27) in December to a peak of 1.34 (95% CI: 1.31–1.37) in January. The estimated median infection attack rate at the end of the season was 27.4% (95% CI: 23.7–30.5%). Conclusions Our findings support a relationship between temperature and contact patterns, which can contribute to deepen the understanding of the relationship between social interactions and the epidemiology of respiratory infectious diseases.


Challenges of COVID-19 Case Forecasting in the US, 2020–2021

May 2024

·

110 Reads

·

9 Citations

During the COVID-19 pandemic, forecasting COVID-19 trends to support planning and response was a priority for scientists and decision makers alike. In the United States, COVID-19 forecasting was coordinated by a large group of universities, companies, and government entities led by the Centers for Disease Control and Prevention and the US COVID-19 Forecast Hub (https://covid19forecasthub.org). We evaluated approximately 9.7 million forecasts of weekly state-level COVID-19 cases for predictions 1–4 weeks into the future submitted by 24 teams from August 2020 to December 2021. We assessed coverage of central prediction intervals and weighted interval scores (WIS), adjusting for missing forecasts relative to a baseline forecast, and used a Gaussian generalized estimating equation (GEE) model to evaluate differences in skill across epidemic phases that were defined by the effective reproduction number. Overall, we found high variation in skill across individual models, with ensemble-based forecasts outperforming other approaches. Forecast skill relative to the baseline was generally higher for larger jurisdictions (e.g., states compared to counties). Over time, forecasts generally performed worst in periods of rapid changes in reported cases (either in increasing or decreasing epidemic phases) with 95% prediction interval coverage dropping below 50% during the growth phases of the winter 2020, Delta, and Omicron waves. Ideally, case forecasts could serve as a leading indicator of changes in transmission dynamics. However, while most COVID-19 case forecasts outperformed a naïve baseline model, even the most accurate case forecasts were unreliable in key phases. Further research could improve forecasts of leading indicators, like COVID-19 cases, by leveraging additional real-time data, addressing performance across phases, improving the characterization of forecast confidence, and ensuring that forecasts were coherent across spatial scales. In the meantime, it is critical for forecast users to appreciate current limitations and use a broad set of indicators to inform pandemic-related decision making.


Projected weekly COVID-19 hospitalizations in the United States across scenarios, April 2023–April 2025
Ensemble projections from the COVID-19 SMH of national COVID-19 hospitalization for the period April 2023–April 2025 are shown by scenario. Dots indicate the observed weekly hospitalizations between December 1, 2022 and December 16, 2023. Shading from lightest to darkest represents 90%, 80%, and 50% projection intervals. Red dashed lines correspond to the CDC-designated COVID-19 community-level indicators: medium (10–19 weekly hospitalizations per 100,000) and high (>20 weekly hospitalizations per 100,000) levels. The vertical line on April 15, 2023, marks the start of the projection period. COVID-19, Coronavirus Disease 2019; SMH, Scenario Modeling Hub.
Percent and total prevented COVID-19 hospitalizations and deaths by annual vaccination recommendation with reformulated vaccines
Relative and absolute differences in cumulative hospitalizations and deaths over the next 2 years (April 2023–April 2025) between different vaccination recommendations. Red and blue dots and error bars represent the median and 95% CI of percent prevented outcomes in high and low immune escape scenarios (50% per year and 20% per year), respectively. CI, confidence interval; COVID-19, Coronavirus Disease 2019.
Relationship between prevented COVID-19 hospitalizations and assumed vaccine coverage in individuals aged 65 and above across US states
The relationship between the cumulative difference in COVID-19 hospitalizations for the next 2 years (April 2023–April 2025) under different vaccination recommendations and assumed vaccine uptake among those aged 65 and above (65+) in each US state: (A and B) vaccination of all compared to no vaccination and (C and D) vaccination of 65+, compared to no vaccination. The x-axis represents the assumed vaccine coverage among 65+ at saturation considering the higher severity in 65+ (likely to have the most significant contribution to decreasing hospitalizations). Dots in each panel correspond to individual US states. COVID-19, Coronavirus Disease 2019.
Comparison between the projected COVID-19 mortality by scenario and the 10 leading causes of pre-pandemic mortality in the United States
Projected COVID-19 mortality by scenario and by period (April 2023–April 2024 and April 2024–April 2025) are compared with the 10 leading causes of mortality in the United States, which were obtained from the CDC age-adjusted disease burden rates in the pre-pandemic period [28]. COVID-19, Coronavirus Disease 2019.
Projected national peak timing and peak size of hospitalizations across scenarios
Potential impact of annual vaccination with reformulated COVID-19 vaccines: Lessons from the US COVID-19 scenario modeling hub

April 2024

·

150 Reads

·

3 Citations

Background Coronavirus Disease 2019 (COVID-19) continues to cause significant hospitalizations and deaths in the United States. Its continued burden and the impact of annually reformulated vaccines remain unclear. Here, we present projections of COVID-19 hospitalizations and deaths in the United States for the next 2 years under 2 plausible assumptions about immune escape (20% per year and 50% per year) and 3 possible CDC recommendations for the use of annually reformulated vaccines (no recommendation, vaccination for those aged 65 years and over, vaccination for all eligible age groups based on FDA approval). Methods and findings The COVID-19 Scenario Modeling Hub solicited projections of COVID-19 hospitalization and deaths between April 15, 2023 and April 15, 2025 under 6 scenarios representing the intersection of considered levels of immune escape and vaccination. Annually reformulated vaccines are assumed to be 65% effective against symptomatic infection with strains circulating on June 15 of each year and to become available on September 1. Age- and state-specific coverage in recommended groups was assumed to match that seen for the first (fall 2021) COVID-19 booster. State and national projections from 8 modeling teams were ensembled to produce projections for each scenario and expected reductions in disease outcomes due to vaccination over the projection period. From April 15, 2023 to April 15, 2025, COVID-19 is projected to cause annual epidemics peaking November to January. In the most pessimistic scenario (high immune escape, no vaccination recommendation), we project 2.1 million (90% projection interval (PI) [1,438,000, 4,270,000]) hospitalizations and 209,000 (90% PI [139,000, 461,000]) deaths, exceeding pre-pandemic mortality of influenza and pneumonia. In high immune escape scenarios, vaccination of those aged 65+ results in 230,000 (95% confidence interval (CI) [104,000, 355,000]) fewer hospitalizations and 33,000 (95% CI [12,000, 54,000]) fewer deaths, while vaccination of all eligible individuals results in 431,000 (95% CI: 264,000–598,000) fewer hospitalizations and 49,000 (95% CI [29,000, 69,000]) fewer deaths. Conclusions COVID-19 is projected to be a significant public health threat over the coming 2 years. Broad vaccination has the potential to substantially reduce the burden of this disease, saving tens of thousands of lives each year.


The decline of the 2022 Italian mpox epidemic: Role of behavior changes and control strategies

March 2024

·

60 Reads

·

7 Citations

In 2022, a global outbreak of mpox occurred, predominantly impacting men who have sex with men (MSM). The rapid decline of this epidemic is yet to be fully understood. We investigated the Italian outbreak by means of an individual-based mathematical model calibrated to surveillance data. The model accounts for transmission within the MSM sexual contact network, in recreational and sex clubs attended by MSM, and in households. We indicate a strong spontaneous reduction in sexual transmission (61-87%) in affected MSM communities as the possible driving factor for the rapid decline in cases. The MSM sexual contact network was the main responsible for transmission (about 80%), with clubs and households contributing residually. Contact tracing prevented about half of the potential cases, and a higher success rate in tracing contacts could significantly amplify its effectiveness. Notably, immunizing the 23% of MSM with the highest sexual activity (10 or more partners per year) could completely prevent new mpox resurgences. This research underscores the importance of augmenting contact tracing, targeted immunization campaigns of high-risk groups, and fostering reactive behavioral changes as key strategies to manage and prevent the spread of emerging sexually transmitted pathogens like mpox within the MSM community.



Ad

Citations (80)


... The feasibility of wastewater surveillance, especially near-source surveillance in places such as schools and long-haul aircraft, hinges on the accurate characterization of shedding of the relevant analytes 11, 98 . For example, in the case of aircraft wastewater surveillance, whether or not passengers shed SARS-CoV-2 RNA in urine makes a significant difference in both the number of aircraft that must be tested and in the likelihood of detecting a novel infectious agent via such a system 17,99 . Whether or not such a system is fiscally and technically feasible could hinge entirely on the question of urine. ...

Reference:

Revisiting the Potential Role of Urine in Wastewater Surveillance: COVID-19 and Beyond
Optimization and performance analytics of global aircraft-based wastewater surveillance networks

... Although using historical data to inform influenza forecasts is more common for statistical/ machine learning models than for mechanistic models [3,13,[16][17][18][19][20][21]24,42,43] other studies also made use of past data to improve standard SIR models. For instance, the study by Ben-Nun et al [44] used data augmentation to make maximum use of prior data within a mechanistic framework. ...

Evaluation of FluSight influenza forecasting in the 2021–22 and 2022–23 seasons with a new target laboratory-confirmed influenza hospitalizations

... In complex social systems, individuals' behaviour and interactions affect the collective outcomes, but the relationship can be fundamentally different from a simple sum or average [67][68][69][70] . The collective outcomes in H-M social systems differ from those in human-only systems because machines behave differently from humans and H-M and M-M the ecological approach to studying machine behaviour 16 , the hybrid collective intelligence perspective 17 and the budding field of social AI 18 . We aim to offer a conceptual overview of the topic. ...

Human-AI Coevolution

... The immunisation status arms also have a W compartment to allow for waning of immunity after vaccination or mAbs. Individuals move between age groups, states, levels, and arms at rates specified in Additional file 1: Tables S9-S12 [2,4,7,[9][10][11][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30]. The levels refer to the number of RSV infection in an individual. ...

Evaluating Seasonal Variations in Human Contact Patterns and Their Impact on the Transmission of Respiratory Infectious Diseases

... Notwithstanding, even the most accurate case forecasts exhibited unreliability in critical phases, underscoring the imperative for enhanced leading indicator forecasts. 6 Classical epidemic models, such as 'Susceptible, Infectious' (SI), 'Susceptible, Infectious, Susceptible' (SIS), and 'Susceptible, Infectious, Recovered' (SIR) models, found application in simulating the spread of COVID-19 and prognosticating its dynamics. 7 Expanding the scope, another study applied traditional predictive models to forecast the total number of COVID-19 cases and risk factors associated with the virus, emphasizing a multifaceted approach encompassing epidemiological dynamics and contextual risk factors. ...

Challenges of COVID-19 Case Forecasting in the US, 2020–2021

... Round 14 also provided evidence to support the White House's push for manufacturers to have bivalent vaccines ready sooner, moving up the availability of a COVID-19 vaccine booster by two months, from November 2022 to September 2022. Further, Round 17 projections demonstrated the potential of broad repeated booster vaccination to reduce the burden of disease at a two-year projection period (Jung et al., 2023). ...

Potential impact of annual vaccination with reformulated COVID-19 vaccines: Lessons from the US COVID-19 scenario modeling hub

... Indeed, the rapid infection of highly active MSM could have already led to infection-induced immunity at the population level 4,5 ; or individuals may have spontaneously avoided risk in response to public health messaging 6,7 . The interplay between these factors is complex and context-specific, with timing and impact varying across countries 5,[8][9][10][11][12][13][14] , depending on factors like the time of first importations 2 , vaccination campaign onset, and resource availability. In France, vaccine postexposure prophylaxis (PEP) of contacts of confirmed cases began on May 27, shortly after the first confirmed case on May 19, but was initially limited to 802 doses over 45 days. ...

The decline of the 2022 Italian mpox epidemic: Role of behavior changes and control strategies

... The Global Epidemic and Mobility model. The Global Epidemic and Mobility model (GLEAM) is a computational platform used for modeling epidemic spread, combining stochastic elements and spatial data in an age-structured, metapopulation framework [23][24][25]. GLEAM divides the world into distinct geographic subpopulations using a Voronoi tessellation of the Earth's surface, with each subpopulation centered around major transportation hubs such as airports. These subpopulations are detailed with high-resolution data about population demographics, age-specific contact patterns, health infrastructure, . ...

A multiscale modeling framework for Scenario Modeling: Characterizing the heterogeneity of the COVID-19 epidemic in the US
  • Citing Article
  • March 2024

Epidemics

... Performance-wise, the trimmed LOP generally produces the highest performing ensemble model for COVID-19 and outperforms the individual models on average (Howerton et al., 2023a). While much progress has been made in ensembling and performance evaluation of infectious disease projections in recent years (Bay et al., 2023;Bracher et al., 2021;Cramer et al., 2022b;Keeling et al., 2022;Sherratt et al., 2023), more work is needed to further improve these multi-model efforts. ...

Ensemble 2 : Scenarios ensembling for communication and performance analysis
  • Citing Article
  • February 2024

Epidemics

... It is well known that reductions in mobility (a proxy for reductions in population mixing) have reduced the transmission of COVID-19 within a location especially during the early phase of a disease outbreak (61,62). However, it remains unclear how structural changes to the mobility network (shifts in the frequency and intensity of mobility within and among regions) have impacted COVID-19 dynamics empirically (56,57,(63)(64)(65). Our underlying hypothesis is that more tightly connected communities exhibit more synchronized epidemic dynamics and, conversely, that more disjointed individual communities have less synchronized epidemics and their epidemics are more likely to fade out (16,18,19). ...

Characterizing collective physical distancing in the U.S. during the first nine months of the COVID-19 pandemic