AI Flags High-Risk COVID-19 Patients
In Israel, an AI-based system has been found to be an effective way in helping to find people who are most at risk in developing severe COVID-19 complications. To help in making its predictions, the system draws on a range of medical data such as a person’s age, health conditions such as heart disease or diabetes, weight and body mass index, respiratory disease history, and previous history of hospital admissions. The AI is able to trawl through a vast number of records and spot at-risk individuals who might have been missed otherwise.
Macabbi Health Services, a leading health maintenance organisation (HMO) in the country, said the system enable it to identify who among the 2.4 million people were within the high-risk group. It has been known to have already flagged 2% of its members, which is the almost equivalent to 40,000 members. Once these individuals were identified, they were then prioritised for testing.
Medical team members, wearing protective gear, handle a coronavirus test from patients in Jerusalem. Image Credit: The Times of Israel.
Macabbi also uses the AI to help assess the level of treatment that these high-risk members might require should they get sick: 1) home-based care; 2) confinement in a quarantine hotel; or 3) admission to hospital. The organisation says that it is currently talking to major health providers in the United States who are interested in using the AI system to help flag their own high-risk patients in order to reduce the spread of Coronavirus.
“Using AI to identify vulnerable people could save lives which develops software to analyse unstructured medical data, such as doctors’ notes.”
Schulte believes that the Macabbi tool could also be used to isolate potential high-risk members of the population when lockdown measures are relaxed, possibly by moving those diagnosed into special housing situated away from family members who may be undiagnosed carriers of the virus.
However, bringing such a tool to the US and other countries is not always that easy. For instance, in the US medical records are kept in ‘data silos’ of hospitals and many different health-care systems.
“Our ability to develop algorithms to identify individuals as high risk is limited by the lack of data sets. Even in New York City, I suspect it’s a challenge to craft a single data set that brings together patient information across the large hospitals.”
The US Office of the National Coordinator for Health Information Technology, a division in the US government responsible for health-care IT, has recently introduced regulations supporting secure data transfer between different hospitals.
“We just need providers to make patient data possible”