NUH employs AI tool for faster, more effective treatment of stroke patients in Singapore

Artificial intelligence platform RapidAI quickly identifies suitable patients by rapidly processing data from CT scans

Revolutionising stroke care: Singapore's National University Hospital employs AI triage for faster and more precise treatment.
Revolutionising stroke care: Singapore's National University Hospital employs AI triage for faster and more precise treatment. (PHOTO: NUH)

SINGAPORE — Stroke patients at Singapore's National University Hospital (NUH) can benefit from a faster and more efficient care delivery system, thanks to artificial intelligence (AI).

Using an AI triage tool called RapidAI, the hospital can quickly identify stroke patients who would benefit from endovascular treatment - a less invasive procedure for treating stroke.

NUH said on Thursday (18 May) that this approach saves crucial minutes, enhances patient outcomes, and potentially establishes a new standard of care for stroke patients in the country.

Since February, all suspected stroke patients at NUH have been receiving care with the assistance of RapidAI, with 400 patients benefitting from the tool so far.

Transforming stroke care

Stroke, particularly acute ischemic stroke which results from blockages in major brain blood vessels, is a leading cause of disability around the world. According to the Ministry of Health, it is the fourth leading cause of death in Singapore, affecting four per cent of adults aged 50 and above

Restoring blood flow promptly is crucial for effective treatment, and patients with blockages in their brain's large arteries undergo mechanical thrombectomy, which physically removes clots and unblocks the artery.

According to NUH, the success of this treatment depends heavily on time, making advanced imaging crucial for selecting appropriate candidates for treatment.

The RapidAI platform automatically processes imaging data from CT scans, generating colour-coded images with relevant information, and instantly notifies stroke neurologists and interventional radiologists. It also informs doctors about the extent of brain damage caused by the stroke and the volume at risk that can potentially be saved through mechanical thrombectomy.

Previously, radiographers manually processed CT scans, and it could take up to 20 minutes before relevant information was available for review by doctors.

Caring for over 1,000 ischaemic stroke patients annually

The stroke team at NUH takes care of over 1,000 patients with ischaemic stroke every year. About one out of every six cases receives endovascular treatment.

Adjunct Associate Professor Anil Gopinathan, head and senior consultant of the NUH's Interventional Radiology division, expressed his optimism about the AI platform,

"AI has the potential to completely change the speed at which therapeutic decisions are made, thereby reducing the amount of brain cells lost and improving the likelihood of a stroke patient walking out of the hospital," he said.

Dr Leonard Yeo, NUH senior consultant in the Division of Neurology, said the hospital sets a target of providing appropriate treatment to at least half of acute ischemic stroke patients within an hour of their arrivals.

He believes that with the AI-powered platform, NUH can significantly improve on this target and reduce the door-to-puncture time for patients.

"With each minute of delay in opening a blocked artery, the patient could lose 1.9 million brain cells. Hence, 'time is brain'," he said.

Do you have a story tip? Email:

You can also follow us on Facebook, Instagram, TikTok and Twitter. Also check out our Southeast Asia, Food, and Gaming channels on YouTube.

Yahoo Singapore Telegram
Yahoo Singapore Telegram