12 hours ago
What will Ireland's first in-hospital Artificial Intelligence centre be used for?
THE MATER MISERICORDIAE University Hospital has launched Ireland's first ever in-hospital AI centre.
The hub is developing and in some cases already using AI tools that will do everything from selecting patients for cancer clinical trials, to detecting fractures in scans, and even creating synthetic MRI scans from CT scans.
The work on synthetic MRI scans is aimed at speeding up the diagnosis of spinal injuries that require treatment within an urgent timeframe.
Paul Banahan, the Mater's Senior AI Research Fellow is leading the project.
'This is the first time something like this has been developed alongside medical professionals in a hospital setting, so I'm getting feedback in real time. It's already clear that this can be a real way to help with barriers to MRI scan access, especially in emergency out of hours settings,' he said.
AI tools are already being used for fracture detection in the Mater, by assisting clinicians when it comes to reading scans.
HSE funding needed
Banahan's work has been chiefly funded by Enterprise Ireland. At the end of a two year project funded by the agency, the hospital then decided to keep him on.
He believes that the HSE should be investing in AI research across clinical settings.
'Things are changing with the HSE and its interest in AI. There are a lot of people working in the health technology and AI sector, and funding is needed to help with that. I'm on my own doing a lot of this work in the hospitals, and the academics I work alongside want to know about a lot of specific points… it would be great to see a whole team of researchers in place,' he said.
Advertisement
Banahan said that the regulations around the use of AI in public and private healthcare settings needs to be developed, and fully understood by lawmakers and treatment providers.
'There's a lot of work that can be done with AI, but it needs to be done in the right way.
'There's research ethics and risk assessment, and it's still evolving. What we're launching here is the start of the process, and we want to be transparent in what we are doing,' he said.
AI can help detect blocked arteries
Prof Joe Galvin, Consultant Cardiologist at the Mater, is seeking to run a pilot project that will develop AI that can assist clinicians working with patients experiencing chest pain, coronary blockages, and arrhythmias.
'AI is most useful in medicine when it's used to help us interpret tests. In cardiology we have ECGs, they give us a lot of information about the heart.
'In emergency situations when someone comes into the hospital with chest pain we're trying to figure out: Is this their heart? Are they having a heart attack or not? AI can help us to do that,' he said.
Prof Galvin explained that clinicians are looking for tell-tale signs of patients presenting with chest pain having a fully blocked artery.
'AI can look at an ECG and see things that human eyes cannot see because it has looked at so many scans in high resolution,' he said.
The Mater doesn't have approval or funding yet to initiate the study, which will work as a pilot project for the entire country. Prof. Joe Galvin believes that the technology may potentially save lives.
Erin Daly, the Operations Manager of the Pillar Centre, says that Ireland and other countries are at a 'tipping point' when it comes to the integration of AI into health treatments
'I think we are going to get to the stage where patients really are going to be looking to go to hospitals that are AI enabled. They are going to want to know that that double-check is there, and that safety net is there. That's the future,' she said.
Daly added that it's essential that human expertise is still leading the way, however, at every part of the process.
Readers like you are keeping these stories free for everyone...
A mix of advertising and supporting contributions helps keep paywalls away from valuable information like this article.
Over 5,000 readers like you have already stepped up and support us with a monthly payment or a once-off donation.
Learn More
Support The Journal
