Predictive Algorithm for Infectious Disease Outbreak

Identifies Geographic Locations Susceptible to Potential Disease Outbreaks at a Future Date

This predictive algorithm accounts for multiple variables, such as climate sensitivities, population density, social practices, and human movement, to predict the level of disease outbreak and whether the outbreak is waxing, waning, or static. The impact of pathogens on human populations typically lead to outbreaks of massive diseases and these outbreaks will continue to occur over time. Conventional epidemiological models are simulative, generating “what if” scenarios, and requires disease prevalence or incidence. In the epidemiological domain, there is less familiarity with the concept of using forecasted risk of disease outbreaks to take early action. This is mainly due to the lack of disciplinary focus on the integration of weather and climate information with epidemiological data, attributing to the absence of predictive tools for air-, water-, and vector-borne diseases, such as Cholera and COVID-19.

Researchers at the University of Florida have developed a predictive intelligence system to identify when and where infectious disease outbreaks are likely to occur weeks in advance. This algorithm accounts for variables differently than current technologies, utilizing a disease risk model (DRM) approach. By calculating the risk of the trigger and transmission of a particular infectious disease, it allows governments, world health organizations, and companies to develop mitigation and intervention plans that could reduce the impacts of an outbreak.

Application

Develops a prediction of when and where infectious disease outbreaks are likely to occur in the future

Advantages

• Predicts infectious disease outbreaks in advance, allowing governments, world health organizations, and companies to develop mitigation and intervention plans to prevent disease spread
• Integrative techniques account for environmental factors that could spread disease, allowing for predictions on outbreaks
• It can be applied to local, regional, and global scales with little to no tweaking of the model parameters
• Utilizes a risk score approach, minimizing the disease surveillance data availability requirements, such as prevalence or incidence time series
• Insurance companies remain informed on worldwide disease risks, allowing them to easily update policies
• Predictive algorithm encompasses symptomatic and holistic approaches in infectious disease spread predictions

Technology

The predictive algorithm identifies the occurrence of an infectious disease in a particular geographic location and gives risk scores to determine the likelihood of an outbreak. The disease risk model (DRM) integrates a multitude of social and environmental factors, such as climate, geographic location, human mobility, demographics, access to sanitation, and more, to produce this risk score. It also utilizes data from newer transmission pathways of pathogens for the predictive algorithm. The algorithm can also determine whether the number of cases is increasing, decreasing, or staying the same. This prediction system comprises a computer, an operating system, and a processor.