Remarkable for a Machine: In-Home Care Chatbots Among Artificial Intelligence Solutions Being Embraced by Australia's Health System

Peta Rolls came to anticipate getting Aida's regular check-in at 10am.

A routine morning call by an automated voice assistant wasn't initially included in the care package the participant expected when she signed up for St Vincent’s in-home support however when she was invited to be part of the trial several months back, the 79-year-old said yes because she wished to contribute. Even though, to be honest, her expectations were low.

Even so, when the call came through, she says: “I was so overtaken by how interactive she was. It was remarkable for a machine.”

“The system would inquire ‘how are you feeling today?’ and that provides a chance if you feel unwell to mention your symptoms, or I just say ‘I’m fine, thank you’.”

“She would go on to ask follow-up questions – ‘have you had a chance to step outside today?’”

The virtual assistant would also inquire about what the user had planned for the day and “it would reply appropriately.”

“When I mentioned I’m going shopping, she’d say are you shopping for clothes or groceries? It was quite engaging.”

AI Reducing the Administrative Burden on Medical Staff

This pilot, which has recently concluded its first phase, is an example in which progress in AI technology are being integrated in the medical field.

Digital health company Healthily approached the care organization about the program to utilize its generative AI technology to offer companionship, as well as an opportunity for elderly recipients to log any health issues or concerns for a staff member to follow up.

Dean Jones, head of St Vincent’s At Home, says the service under evaluation is not a substitute for any face to face interactions.

“Recipients continue to get a weekly personal visit, but in between visits … the automated system allows a daily check-in, which can then flag any possible issues to either our team or a client’s family,” the director notes.

Dr Tina Campbell, the managing director of Healthily, says there haven’t been any adverse incidents reported from the St Vincent’s trial.

Healthily uses advanced AI “with very clear guardrails and prompts” to guarantee the interaction is secure and procedures are established to respond to serious health issues promptly, the director says. As an instance, if a patient is reporting chest pains, it would be flagged to the medical staff and the conversation ended so the person could dial triple zero.

She thinks artificial intelligence has an important role given staffing shortages throughout the medical industry.

“The benefit very safely, using such systems, is reduce the admin burden on the staff so qualified health professionals can focus on performing the duties that they’re trained to do,” she comments.

Artificial Intelligence Long Established as Often Believed

Prof Enrico Coiera, the co-founder of the national AI health alliance, says established types of AI have been a standard part of medicine for a long time, often in “back office services” such as analyzing scans, ECGs and lab reports.

“Software that performs a function that requires judgment in certain aspects is artificial intelligence, regardless of how it achieves that,” states the professor, who is also the head of the Centre for Health Informatics at a leading university.

“When visiting the imaging department, radiology department or pathology lab, you will find software in machines doing just that.”

In recent years, advanced versions of artificial intelligence called “deep learning” – an algorithmic approach that enables algorithms to analyze very large sets of data – have been employed to interpret medical imaging and improve diagnosis, the expert notes.

In November, BreastScreen NSW became Australia’s pioneering population-based screening program to introduce machine reading technology to support specialists in interpreting a select range of mammography images.

These represent advanced systems that still require a qualified physician to evaluate the diagnosis they could indicate, and the accountability for a medical decision rests with the medical practitioner, the professor says.

The Function of AI in Identifying Illness Early

The Murdoch Children’s Research Institute in Melbourne has been collaborating with researchers from UCL London who first developed artificial intelligence techniques to detect epilepsy brain abnormalities called focal cortical dysplasias from brain scans.

These lesions cause epileptic episodes that often cannot be controlled with medication, meaning surgery to excise the tissue becomes the only treatment available. However, the procedure can only be performed if the doctors can pinpoint the abnormal tissue.

A study published this week in the journal Epilepsia, a group from the institute, headed by specialist the lead researcher, showed their “neural network tool” could detect the abnormalities in nearly all of cases from MRI and PET scans in a subtype of the lesions that have traditionally been overlooked in more than half of cases (60%).

The AI was developed using the images of 54 patients and then tested on 17 children and 12 adults. Among the youngsters, twelve underwent operations and eleven became free of seizures.

The tool uses AI algorithms comparable with the breast cancer screening – flagging regions of abnormality, which are still checked by specialists “but it makes it a lot quicker to get to the answers,” the researcher says.

She stresses the team are still in the “early phases” of the work, with a additional research required to get the technology toward clinical implementation.

Prof Mark Cook, a neurologist who was independent from the study, says modern imaging now generate such vast quantities of high-resolution data that it is challenging for a person to review it thoroughly. Thus for clinicians the difficulty of finding these lesions was like “identifying the needle in the haystack.”

“It’s a great demonstration of how AI can support clinicians in making earlier, precise identifications, and has the potential to improve operation opportunities and outcomes for kids with treatment-resistant seizures,” the professor says.

Disease Detection in the Future

A public health expert, the vice-president of the international body's AI health division, says advanced AI systems are additionally used to track and forecast disease outbreaks.

Buttigieg, who spoke last month at the Public Health of Australia’s conference in Wollongong, gave as an example Blue Dot, a company established by infectious disease specialists and which was an early detector to detect the Covid-19 outbreak.

Content-creating AI is a additional branch of machine learning, in which the technology can generate new content using training data. Such applications in healthcare include tools such as the virtual assistant as well as the automated note-takers doctors and allied health professionals are adopting more.

Dr Michael Wright, the president of the national GP body, says GPs have been embracing AI scribes, which records the appointment and converts it to a medical summary that can be included in the health file.

Wright says the primary advantage of the scribes is that it enhances the quality of the interaction between the doctor and patient.

Dr Danielle McMullen, the president of the national doctors' group, agrees that AI note-takers are helping physicians manage schedules and adds artificial intelligence also has the potential to prevent repeated examinations and imaging for their clients, if the {promised digitisation|planned digitalization