New study from the George Institute for Global Health at the University of Oxford has found that machine learning can help healthcare workers predict whether patients may require emergency hospital admission.
Image: Machine learning could predict the needs for emergency admission in hospitals. Photo: Courtesy of The George Institute for Global Health.
Machine learning is a field of artificial intelligence that uses statistical techniques to enable computer systems to learn from data.
The research, which was published in the journal PLOS Medicine, suggests that the use of these techniques could help health practitioners to accurately monitor the risks faced by patients and can put in place measures to avoid unplanned admissions, which could be a major source of healthcare spending, the report stated.
The George Institute former data scientist and study lead, Fatemeh Rahimian said: “There were over 5.9 million recorded emergency hospital admissions in the UK in 2017, and a large proportion of them were avoidable. We wanted to provide a tool that would enable healthcare workers to accurately monitor the risks faced by their patients, and as a result make better decisions around patient screening and proactive care that could help reduce the burden of emergency admissions.”
The study involved the records of 4.6 million patients between 1985 and 2015. It was conducted using linked electronic health records from the UK’s Clinical Practice Research Datalink.
Wide range of factors were also taken into consideration such as age, sex, ethnicity, socioeconomic status, family history, lifestyle factors, comorbidities, medication and marital status, as well as the time since first diagnosis, last use of the health system and latest laboratory tests.
The study also used more variables combined with the information about their timing and the machine learning models could provide more robust prediction of the risk of emergency hospital admission than any models used previously.
Rahimian said: “Our findings show that with large datasets which contain rich information about individuals, machine learning models outperform one of the best conventional statistical models. “We think this is because machine learning models automatically capture and ‘learn’ from interactions between the data that we were not previously aware of.”
The team also stated that further research is needed to know whether machine learning models can lead to similarly strong improvements in risk prediction in other areas of medicine.