{{announcement.body}}
{{announcement.title}}

Can Smart Thermometer Data Predict the COVID-19 Pandemic?

DZone 's Guide to

Can Smart Thermometer Data Predict the COVID-19 Pandemic?

See how, using sensors and geospatial data, one company is using IoT to help predict and combat flu-like illnesses like novel coronavirus.

· IoT Zone ·
Free Resource

Rapidly identifying emerging epidemics remains a massive challenge that limits our ability to effectively curtail outbreaks, such as COVID-19. In response, smart thermometer company Kinsa has developed a method to identify anomalous influenza-like illness incidence (ILI) outbreaks in real-time using their county-level illness signals, developed from real-time geospatial thermometer data and highly accurate 12-week illness forecasts. Through their analysis, Kinsa flags anomalously high incidence data by comparing real-time ILI to expected seasonal influenza trends, where these expectations are generated from geo-specific influenza forecasts made from a point prior to potential outbreaks.

Creation of The U.S Health Weather Map


The U.S. Health Weather Map is a visualization of seasonal illness linked to influenza-like illness. The aggregate, anonymized data visualized is derived from Kinsa’s network of Smart Thermometers and the accompanying mobile applications.

The mapping was Created in collaboration with Benjamin Dalziel, Oregon State University. It shows two key data points: The illness levels currently being observing, and the degree to which those levels are higher than the typical levels we expect to see at this point in the flu season. Kinsa believes these “atypical illness”, may in some cases be connected to the COVID-19 pandemic.

Kinsa was able to generate a range of expected influenza trends per county by using a highly accurate, long-lead influenza forecast trained on their county-level ILI data. This forecasting approach leverages geo-specific data to estimate daily, seasonal transmissivity patterns of influenza per city, allowing us to learn the ‘fingerprint’ of each region’s past influenza outbreaks. This method builds upon the findings outlined in academic research that demonstrates that individual cities have unique epidemic intensity curves driven by climate and population structure. These patterns are used to build highly accurate long-lead illness forecasts at the county-scale. For example, small cities are known to have sharp epidemics, whereas larger cities tend to have ‘flattened’ incidence curves due to herd immunity effects.

According to Kinsa’s CEO, Inder Singh, 

"We’ve built an early warning system showing where there are unusually high and growing levels of fever. This system is meant to guide public health first-responders to the areas needing further investigation and resources, because something outside of the ordinary is happening. 

We’ve done our best to make this tool easily accessible. The data is displayed as a map showing current illness levels throughout the United States. We’ve also created another view that highlights areas where illness levels are higher than typically expected at this point in the flu season." 

While not predicting COVID-19, the company notes that a very strong correlation since March 1 between cumulative atypical illness incidence and positive COVID-19 tests (at the state level) in terms of geographies affected and timing within affected geographies, which suggests that their data provides a useful indication of where COVID-19 may likely be occurring.

Brooklyn, NY - Late Season Detection

Example plot for how this method would be used to identify potential COVID-19 outbreaks in real-time. Expected influenza forecasts are generated from a point prior to outbreak, and anomalies falling outside of the 95% confidence limit are identified each day.

Kinsa recommends interpreting their data as complementary to other available data sources including explicit COVID-19 testing and ER Admissions data, etc. when determining where to allocate scarce resources. Clearly the data cannot be accurate in areas where there are not many smart thermometers sold.

The Value of Geospatial Analysis

The map was made using Open map via CARTO's location intelligence platform. CARTO have made its platform available to public and private sector organizations fighting against the coronavirus outbreak via their grants program and it has been used by a number of organisations to assist with mapping and geospatial analysis such as the Humanistic GIS Lab at the University of Washington - Seattle, NBC New York News Coverage. Hocelot in Spain are trying to avoid emergency calls overloading the system and to identify infection focus, while Colten Care Limited in the UK is using the platform to visually display and manage the affect of COVID-19 on their care home residents over 21 locations, their interaction with families/friends and their employees.

Topics:
mapping ,geospatial data ,geospatial analysis ,smart thermometer ,health tech ,covid-19 ,iot

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}