Artificial Intelligence (AI) has been a hot topic in business transformation over the past few years, but in the healthcare sector, it’s been quietly chipping away at manual processes since the 1970s.

Back then, AI was used to process X-ray images to look for fractures and very early versions of AI was used in medical notes, albeit without the computational power of today.

The healthcare sector now features robotic surgery, voice activated medical notes, AI-powered analysis of datasets to develop new medicines and vaccines, and machine learning (ML) predicts the treatment best suited to a given patient.

The healthcare sector needs these types of innovations as an ageing population, expanding patient expectations and rising costs place extra demand on the system.

David Hansen is CEO of the Australian e-Health Research Centre at the CSIRO, Australia’s national science agency. He says two factors are making it easier to support AI in digital health.  First, more data is being collected electronically – one of the issues in the early days of AI in health was its inability to gather enough data. The second development is compute power that can work across big data sets.

Data is now gathered in cloud-based systems and is enabling data scientists to do some very interesting things – in imaging, for instance. AI is now able to “read” vast numbers of images, and make predictions and diagnoses.

“We can use this, for example, to understand the progression of Alzheimer's disease in people. We can measure the difference between a longitudinal set of brain images from a cohort of patients,” says Hansen. “And that provides us a fabulous data set for supporting clinical trials into Alzheimer's disease.”

Radiotherapy is another example of how AI is providing more timely, and therefore more accurate, data which in turn improves patient outcomes. Radiotherapy treatments often extend over a 7 or 8-week period. For prostate cancer in particular, the treatment planning prior to treatment is conducted using the imaging gained from a CT scan, which does not clearly show the boundaries of the prostate and surrounding soft tissue. This plan is then used for the duration of the treatment period. Hansen says machine learning allows clinicians to plan using MRI imaging – it can be done weekly and provides higher quality and timely images to plan the radiotherapy.

AI transforms how health data is collected and used

Unlike image data collection, which uses similar technologies, other data uses a variety of formats and vocabularies. “And so it's difficult to extract information out of these free text records,” says Hansen.

Nevertheless, it can be done and Hansen cites an example of using artificial challenges and machine learning to extract clinical data from pathology reports. “We're doing that to inform the cancer registry. Rather than cancer coders reading reports and extracting information, we’ve developed AI that can do this automatically as the reports are captured. This means information can be up-to-date … as opposed to it being potentially years behind.”

Another great opportunity is data collection from sensors. “We're seeing more people using wearable devices and we’re also seeing more sensors in the home, such as motion sensors, or temperature sensors, among others. We can use data from these sensors, along with machine learning and artificial intelligence, to support someone living alone,” says Hansen.

The CSIRO recently finished a trial on how an AI-powered system could support someone living alone and determine if there’s a decline in functional independence. “We work with service providers to place sensors in people’s homes. We use AI to map the sensor data to provide an ‘activities of daily living’ measure. This measure helps carers and family determine if there are changes to a person’s activities that may indicate decline or a need for assistance.

Considering that 70% of people want to die at home yet only 14% get to do so, home sensors that relay data to a treating physician can meet patient needs while reducing the need for out-of-home care.1

AI in primary healthcare settings

AI is used in primary care settings such as general practitioner clinics to streamline administration tasks, automate payment and claims systems, and enhance telehealth consultations.

Albert Naffah, CEO of CommBank Health says the increasing uptake of telehealth during the pandemic coupled with the government’s decision to make it a permanent feature of Medicare, will accelerate the uptake of AI solutions, especially as consultations move from audio to video.

He says telehealth brings with it great benefits for patients for rural health and sustainability.

“We know that health outcomes in remote areas fall below capital cities. I think the government is getting more comfortable with investing in virtual hospitals. The other real benefit is an environmental one. Virtual care does not require single use PPE. There’s a huge opportunity from a waste and environmental protection aspect.”

The digitisation of medical records alone will have a significant impact on patient care, their journey through the system and on costs.

Some 14% of pathology tests are ordered due to a lack of access to a patient’s history.2

Barriers to AI uptake

Hansen explains that the full benefit of AI and machine learning can’t be realised without healthcare operators having implemented digital health.

“What we're seeing at the moment is health systems at various stages of implementing electronic health records."

Naffah says investing in these technologies is a big capital expenditure for providers.

“I’m very interested in the intersection of financial services and how we can support the adoption of these technologies. Obviously there’s a huge capital expense required to implement this technology, particularly in primary care,” he says. “But if we can keep people out of hospitals, that’s a much better outcome for the patient and for the economy.”

Our experts

Dr David Hansen is CEO of the Australian e-Health Research Centre at the CSIRO. He holds several leadership positions in many national research initiatives including the Australian Genomics Health Alliance, the NHMRC Centre for Research Excellence in Digital Health and the Australian Alliance for Artificial Intelligence in Healthcare. He’s also a workstream leader for the Global Alliance for Genomics and Health. David is the previous Board Chair of the Health Informatics Society of Australia and member of the Clinical and Technical Board Advisory committee for the Australian Digital Health Agency.
Before joining CSIRO, David held senior positions leading technology research and development for SRS with LION Bioscience Ltd and before that the European Bioinformatics Institute. He is passionate about the role of information and communication technologies in health care and the role of digital health professionals in developing a safe, high quality efficient and sustainable healthcare system in Australia.

About the AEHRC The AEHRC is a digital health research program and is a joint venture between the CSIRO and Queensland Health. With more than 100 scientists and engineers, it is Australia’s largest digital health research centre. With research across health informatics, biomedical informatics and health services research, the AEHRC develops technology and designs digitally enabled services to improve the safety, quality and efficiency of healthcare.

Albert Naffah was appointed CEO of CommBank Health in 2021 after several years as the General Manager of Payments and the Data Economy at CommBank where he was responsible for shaping the bank’s payments strategy, managing the retail and business payments product suite and industry and regulatory engagement. Albert’s remit has included the commercialisation of emerging technology such as the New Payments Platform, APIs, open banking and mobile payments. In his current role, he leads a health industry vertical focussed on serving the financial services needs for healthcare professionals and businesses of all sizes, as well as working with insurers, governments and patients to deliver innovative solutions.

Albert has 25 years’ experience in financial services across general management, strategy, business development, corporate and regulatory affairs, and operations. He’s had extensive experience in Australia, New Zealand and the US with CommBank, Mastercard and Westpac. He has a Bachelor of Commerce and Bachelor of Law.

To learn more from leading experts about what’s important to business and the economy visit CommBank Foresight™ – insights for future-facing businesses.