Padraig Belton
Technological journalist
Mane
Dublin Mater Hospital – houses the most popular emergency service in Ireland
For a famous country as a European discourse of Big Tech, Irish hospitals are often late in technology.
They lack the files of shared computerized patients or unique identifiers to follow people when they move between clinics.
In July 2024, a failure of the computer system made the Dublin Mater Hospital to repel surgeries and beg people not to come to its A&E.
Three years earlier, the Russian ransomware attackers closed the entire IT network of the Irish health system and published 520 online medical records.
But Ireland now has ambitious objectives to modernize its health care.
This includes a program called Sláintecare. Announced in 2017, the plan is to use part of its budget surplus of 22.9 billion euros (20 billion pounds sterling; $ 24 billion) to create a healthy health service, such as the United Kingdom or Canada.
To improve health care, pinching points and diagnoses will have to be improved.
This is a problem addressed at Mater Hospital in Dublin, 164 years old and on the most popular emergency location in Ireland.
This is particularly true in winter, while one day in January in January, the Irish services had 444 people on trolleys waiting to be seen.
“In Ireland, the big problem we have is waiting lists, and in particular awaiting diagnostics, MRI analyzes (magnetic resonance imaging) or CT (Tomodensitometry)”, explains Professor Peter McMahon, consultant radiologist at The Mater.
Due to Professor Macmahon, who, as a student in medicine, has been the subject of an amateur programmer, the mater is now among the first hospitals in Ireland to use artificial intelligence (AI) through its radiology service – the part of a hospital providing medical imaging to diagnose diseases and guide treatments.
To ensure that patients with the most urgent needs are observed first, Professor Macmahon says: “We use AI to immediately analyze all head scanners for bleeding, all thoracic scans for blood clots and all bone radiographs for fractures.”
AI is particularly useful for helping young doctors, when they do not have experienced consultants to turn to.
“Now, a nurse or a junior doctor at 2 am is not the only one, they have a wing man,” he said.
Matering Hospital
Professor Peter McMahon introduced the AI to scan at the Mater Hospital in Dublin
Rural hospitals are faced with different types of challenges.
The Letterkenny University Hospital in Donegal is without MRI installations in the evening and on weekends.
Currently, a patient needs an urgency MRI at night can face an ambulance ride in Dublin.
But now Professor MacMahon and the IA research researcher of Mater Paul Banahan have formed a model of test AI to create a “synthetic MRI” from computed tomography, immediately to patients with patients with suspected vertebral injuries.
This was done by nourishing an “generative AI” model around 9,500 pairs of CT and MRI images of the same area on the same person.
Now, AI can predict what MRI would look like from computed tomography, something available in all emergency services.
And as radiology analyzes are also delivered with the text reports of doctors, it also explores using major language models to identify the models and trends of important diseases.
Peter MacMahon
Ireland maintains digital scans in a central digital library
Applying AI to medical images in Ireland is easier because the country has stored scans in a central digital classification system since 2008.
But many other important information, such as medical notes or electrocardiograms (ECG), remain largely in paper format in most Irish hospitals, or in smaller databases that are not centralized.
This will seriously delay the application of AI to identify potential diseases and improve clinical care, underlines Professor Macmahon.
Aging computer systems in Irish health care is more broadly a challenge.
“Associate, many hospitals are dealing with inherited computer systems where they simply try to keep the show on the road,” said Dr. Robert Ross, computer lecture in Dublin technological university.
“Doing something else like integrating AI is not easy to do,” he says.
The use of AI in health care is not without problems.
An example here is the tools for recognizing the word of AI. Using them could allow doctors to spend less time on taking notes and reporting the writing.
But some have been found to invent things, especially to invent nonexistent drugs.
To prevent such a hallucination AI, “you must make sure that it is penalized in its training, if it gives you something that does not exist”, explains Professor MacMahon.
AIS can have biases, but “humans also have biases,” he said.
A tired doctor, expecting a young patient to be healthy, can ignore his blood clot.
“For any reason, we are much more open to accept human error”, than in the new health technology where “the acceptable risk is zero,” said Professor Seán Kennelly, consultant at the Tallaght University Hospital and Professor at Trinity College Dublin.
This means that we “continue with the illusion of 100% precision in humans”, and ignore the areas where the technology supported by AI can make better clinical decisions, he says.
Tallaght University Hospital
Prof Seán Kennelly (right) and Dr Aidan Boran
Health care regulators, which already have a “sufficiently low” understanding of the software as a medical device, have not at all caught up in AI rules, explains Dr. Aidan Boran, founder of an Irish medical technology start-up called Digital Gait Labs and researcher at the Dublin City University.
For example, obtaining an CE brand, which shows that a medical device responds to EU security regulations, includes the supply of details on the factory where the product is manufactured.
But in the case of software that is not relevant, says Dr. Boran. “For us, manufacturing literally means copying software,” he said.
The AI can have a black box problem: we can see what is going on and what comes out, but the in -depth learning systems that feed these models are so complex that even their creators do not understand exactly what is happening inside.
This can create difficulties for a doctor trying to explain the treatment decisions that involve AI, explains Dr. Paul Gilligan, head of St Patrick’s Mental Health Services, one of the biggest mental health providers in Ireland who heads the St Patrick hospital in Dublin.
When AI influences their decisions, doctors must “be able to articulate reasoning behind these decisions in a way accessible and understandable to affected people,” he said.
More business technology
Source link
