Every morning at around 10am, 79-year-old Peta Rolls waits for a call from “Aida”. A daily check-in call from an AI voice assistant is not what Rolls expected when she signed up for home care with St Vincent’s. But when they asked her to join the trial four months ago, she agreed because she wanted to help. Although she honestly didn’t expect much.
But when she got the call, she said: “I was surprised by how responsive she was. It’s impressive for a robot. She always asks: How are you feeling today? And that gives you the opportunity if you’re sick to say you’re feeling sick, or you just say: I’m fine, thank you.”
Bots reduce administrative burden for home care
The trial – now in its early stages – is one of the ways AI advances are being applied in healthcare.
Health tech company Healthily has asked St Vincent’s to trial generative AI for social interaction, allowing care home residents to record health issues or concerns for staff to follow up on. Dean Jones, National Director of St Vincent’s At Home, said the service does not replace face-to-face interactions. “Clients still have face-to-face meetings each week, but in between, the AI system allows for daily check-ins, where concerns can be escalated to the team or family,” he said.
The company uses open AI technology with clear safeguards and prompts to ensure safe conversations, and has a mechanism to respond quickly to serious health issues. For example, if a customer experiences chest pain, this is reported to the care team and the call is ended so the person can call 000. Tina Campbell, CEO of Healthily, believes AI has an important role to play in the context of the healthcare industry facing many workforce challenges. “What we can do very safely with technology like this is reduce the administrative burden on the workforce, allowing qualified health professionals to focus on the work they are trained to do,” she said.
AI is not as new to medicine as you think.
Enrico Coiera – founder of the Australian Alliance for Artificial Intelligence in Healthcare and director of the Centre for Health Informatics at Macquarie University – says older forms of AI have long been a standard part of healthcare – often in “behind-the-scenes services” such as interpreting medical images, ECGs and test results.
“Any computer program that performs a task that involves decision-making in some way is AI, regardless of how it does it,” he said. Over the past decade, newer forms of AI called “deep learning” – neural network methods that allow algorithms to learn from large amounts of data – have been used to interpret medical images and improve diagnoses.
The role of AI in early disease detection
Murdoch Children's Research Institute in Melbourne, in collaboration with researchers from University College London, has developed AI methods to detect brain lesions caused by epilepsy (focal cortical dysplasias) from MRI images.
These lesions cause severe seizures that are often uncontrollable with medication, leaving surgery as the only option - but surgery is only possible if doctors can find the abnormal tissue.
In a study published this week in the journal Epilepsia, a team led by neurologist Emma Macdonald-Laurs showed that their “epilepsy AI detective” can detect lesions in up to 94 percent of MRI and PET scans, a type of lesion that is typically missed more than half the time. The system was trained on images of 54 patients and then tested on 17 children and 12 adults. Of the 17 children, 12 had surgery and 11 are now seizure-free.
Future disease detection
Stefan Buttigieg, vice president of the European Public Health Association’s Digital Health and AI division, said deep neural networks are also being used to monitor and predict disease outbreaks. He cited the example of Blue Dot, a company founded by an infectious disease expert and one of the first to detect the Covid-19 outbreak.
Generative AI is a branch of deep learning in which technology can generate new content based on training data. These applications in healthcare include voice chatbots like Healthily’s, as well as AI “scribes” that doctors and medical professionals are increasingly using.
Michael Wright, president of the Royal Australian College of GPs, says GPs have embraced AI scribes – which record visits and turn them into notes that can be saved in a patient’s file. He says the main benefit of “scribe” is improving the quality of doctor-patient interactions.
Danielle McMullen, president of the Australian Medical Association, said one of the biggest potentials of AI is to deliver more personalised healthcare. “For many years, healthcare has been delivered in a ‘one size fits all’ way – for example, with medicines – but now we are moving towards a future where there are more sophisticated solutions, and AI must follow suit.”
(Source: theguardian.com)
Source: https://vietnamnet.vn/ai-trong-benh-vien-australia-buoc-vao-ky-nguyen-y-te-thong-minh-2467008.html
Comment (0)