AI doctors ‘significantly’ improve hospital patient outcomes

A US study suggests machine learning interventions mean better care and less chance of death, with one hospital rolling the system out as standard. 

medical

Published this week in the journal Critical Care Medicine, medical researchers looked at 2,740 adult patients who were admitted to four medical-surgical units at The Mount Sinai Hospital in New York.

Divided into two groups, one received real-time alerts triggered when there was a likelihood of clinical deterioration. Notifications were sent directly to nurses, physicians, and members of a rapid response team of intensive care professionals. The other were treated in a standard approach, with deterioration criteria triggering urgent responses from doctors, but without automation. 

The results showed a 43% increase in the likelihood of having care escalated as a result of AI notifications. There was also a ‘significant’ reduction in the chance of death within a 30 day period after treatment when machine learning was involved. Additionally, the ‘intervention group’ had a higher chance of getting medications to support heart and circulation, a sign doctors were taking early action to treat emerging conditions. 

‘We wanted to see if quick alerts made by AI and machine learning, trained on many different types of patient data, could help reduce both how often patients need intensive care and their chances of dying in the hospital,’ said Matthew A. Levin, M.D. Professor of Anesthesiology, Perioperative and Pain Medicine, and Genetics and Genomic Sciences, at Icahn Mount Sinai, and Director of Clinical Data Science at The Mount Sinai Hospital.

‘Traditionally, we have relied on older manual methods such as the Modified Early Warning Score (MEWS) to predict clinical deterioration,’ he continued. ‘However, our study shows automated machine learning algorithm scores that trigger evaluation by the provider can outperform these earlier methods in accurately predicting this decline. Importantly, it allows for earlier intervention, which could save more lives.’

Although the study was terminated due to the outbreak of the COVID-19 pandemic, the algorithm involved in the intervention group has now been deployed at all stepdown units in The Mount Sinai Hospital, where patients receive care for stable conditions but still require close monitoring.

These wards are utilised after people have left intensive care. Specially trained doctors visit the 15 patients with the highest prediction scores for deterioration on a daily basis, making treatment recommendations to the team in charge of general care. As this happens, the machine learning programme is retrained, improving accuracy over time. 

In related news, a study by the Reuters Institute and Oxford University recently found that the ‘hype’ around AI far outweighs current usage. According to the investigation, just 2% of UK adults currently use artificial intelligence tools on a daily basis, although this increases with younger  demographics. Those aged 18-24 were the most eager adopters. 

Image: The Mount Sinai Hospital