By combining extensive private and public information, a small startup says it is able to predict COVID-19 hotspots at the neighborhood level each week, with an accuracy of 92%.
Leading the news: The start-up, Data Driven Health, on Wednesday released a version of its flu prediction model and COVID, which provides data down to the neighborhood level.
Why it’s important: Assuming the company’s fluid model is consistently accurate, governments and health systems could use this data to set policies and allocate resources, while people could use it to assess the personal risk of planned outings.
How it works: Fludemic includes a series of data on a daily basis using a combination of clinical data, socioeconomic indicators, and factors such as mask protection policy and mobility trends.
Yes, but: Although the model is said to have 92% accuracy in detecting hot spots, it is slightly less accurate when sparsely populated areas are contemplated or mortality rates are predicted.
Between lines: The company began work on a flu prediction model last year, but quickly switched gears to COVID-19 once the pandemic began.
- Although neighborhood-level data is made public, Data Driven Health has a premium product aimed at governments and health systems that uses more data types and offers predictions down to the block level.
The big picture: Others have used a mix of technology and data to map diseases. Google, for example, had a flu tracker that sought to predict flu trends through search queries. That effort closed in 2015 amid criticism. More recently, smart thermometer maker Kinsa has detected flu outbreaks.
What follows: In the long run, the company hopes to be able to update its models even more frequently, as well as address other types of health conditions.
- “We want people to be able to see what’s happening in their communities in real time or almost in real time,” CEO Jack Rush told Axios.