ASEAN

Singapore to use AI in hospital ERs to assess patients

SINGAPORE: Hospitals in Singapore will start using artificial intelligence (AI) to assess patients in the emergency room (ER) to identify those who need immediate treatment and those who can wait.

Singapore's The Straits Times reported that a predictive model that can flag patients at risk of death, as well as those who are safe for discharge, has been developed.

It says that AI will analyse patients' records and current diagnoses in the ER before determining whether they need immediate care based on the severity of their condition at that time or otherwise.

Currently, assessing and prioritising patients is done by triage nurses, who inquire about patients' conditions and assess their vital signs before assigning them a Patient Acuity Category (PAC) score.

PAC 1 is the most severe and requires life-saving intervention, while PAC 4 is the least severe and usually non-emergent cases.

The lead researcher of the project and a third-year medical student at the Duke-National University of Singapore's (NUS) Medical School, Yvonne Wong, said that it is quite challenging to accurately measure the risk of death based on the initial assessment by triage nurses.

"This is because it varies depending on the nurse's experience, unlike AI which takes into account all the patient's historical records and current condition before making an assessment," she said.

"Therefore, to address this dilemma, the Duke-NUS team developed an AI model known as PAC+ which utilises data from over 300,000 ED patients from 2016 to 2018.

"The data was used to predict whether patients are at high risk of mortality based on their vital signs and medical history.

"They may generally appear well, but if their risk of death is high according to AI, then doctors can reassess; whether they need priority and faster treatment," she said.

PAC+ is one of nearly 220 projects submitted for the Singapore General Hospital's (SGH) Annual Scientific Meeting awards 2024, held over two days at SGH Campus, which concludes today.

The PAC+ team hopes their model can also help identify patients suitable for Mobile Inpatient Care at Home (MIC@Home), a programme providing hospital-level care to patients in their homes.

Senior Health Minister Dr Janil Puthucheary, who was the guest of honour at the annual event, said that the emergence of AI provides new opportunities to utilise digital technology to enhance disease treatment efficiently and timely.

"With initiatives to ensure the safe implementation of AI in the healthcare industry, such as guideline development in 2021, Dr Janil hopes AI can be safely adopted to complement the healthcare sector.

"For example, it can reduce administrative workload, allowing frontline healthcare workers to focus on patient relationships and create a 'stronger therapeutic alliance' with them," he said.

Most Popular
Related Article
Says Stories