AI in Determining Which COVID-19 Patients Require Priority
COVID-19 is making news across the world now, with millions of people falling prey to the virus globally. It might not come as a surprise that hospitals are not well-prepared to tackle such a drastic rise in demand for
intensive care beds. In fact, the NHS is already struggling to maintain a sufficient supply of intensive care beds that it has been compelled to postpone routine operations, like hip and knee replacements, by three months. There is also an incentive to get the patients that do not require intensive care to be discharged.
In which case, Artificial Intelligence might just be the saviour in terms of determining the severity of the patients suffering from the coronavirus.
With the number of COVID-19 cases rising rapidly, healthcare services are swiftly being depleted of essential resources required to treat the patients, making it critical to determine which patients are in most need of immediate healthcare assistance. To tackle this conundrum, computer scientists at the University of Copenhagen are developing an AI-enabled computer program with the ability to analyse the risk incurred by an individual to determine whether they are in immediate need of a ventilator or ICU.
The AI algorithm is designed to scrutinise a vast amount of data collected from previous Danish coronavirus patients who have been in the system so far. This enables the algorithm to identify any common traits among the patients who were severely affected by the virus. These traits could anything such as white blood cell count, prior consumption of any medication, etc. Consequently, upon identifying these traits, they will be compared with information from newly hospitalized patients. The data consists of X-rays, tests and measurements taken of patients at the time of their admittance to hospital, along with their electronic health records.
Although the model will not be used as a decision maker as to whether or not to treat patients, it will still be used as a planning tool that can help hospitals to restructure their administration in order to provide the most optimal level of healthcare assistance. Works on the development of the algorithm has been underway since the last three weeks and is expected to be able to be produce rough models by late April.
(Thumbnail credit: The Washington Post)