Will artificial intelligence improve health outcomes?

How artificial intelligence is revolutionising healthcare

Technology – it never ceases to amaze. And as artificial intelligence (AI) applications become more and more advanced, they may soon improve access to healthcare. So, this month we looked at what’s happening in this space, to keep you in the know.

Animated digital people helping real ones

Three New Zealand-led international research projects have just been awarded a total of $5.1 million by the MBIE. All three projects will use autonomously animated digital people, created by AI company Soul Machines, to explore how AI technology can provide more personalised healthcare.

More specifically, one project will develop a digital person to provide easier and bespoke access to mental health support. The second project will create a platform of interconnected digital tools to help patients with Type 2 diabetes and cardiovascular disease understand their conditions. And lastly, the third project will focus on people with autism spectrum disorder, using digital avatars to facilitate the training in facial expression recognition.

As Soul Machines’ CEO Greg Cross explained, the future of healthcare lies in creating “empathetic, personalised experiences across cultures, languages, and geographic regions that are available 24/7.” And as we’ll see shortly, there’s a lot happening in this space all over the world.

Lower costs for developing medicines

In 2015, pharmaceutical research company Atomwise used a super-computer to analyse data and identify what medicine would be most effective against the Ebola virus outbreak in Western Africa. Normally, this process would have taken months or years – but the AI analysis took less than one day. And in saving time, it has also reduced costs significantly.

Faster and more accurate medical imaging

Medical imaging is another key area where AI is transforming healthcare. A recent example is ‘Patient Synopsis’, a solution developed by IBM that provides medical personnel with faster access to patient insights.

Thanks to it, clinicians no longer need to identify patients’ relevant information on their own, potentially spending a long time tracing back old diagnoses and medical procedures. The ‘Patient Synopsis’ AI tool does it for them, quickly running through the patient’s medical history and delivering a friendly report relevant to the procedure.

From early detection to prediction

Certain conditions are more difficult to detect than others, and can potentially lead to serious issues. That’s where AI is proving to be of great help.

In 2019, the US Department of Veterans Affairs announced the development of an AI tool that, thanks to machine learning, can predict acute kidney injury up to 48 hours sooner than with usual care. Elsewhere, a 2018 study used AI and machine learning to combine genetic and physical data, and found that this technology can help estimate the risk of breast cancer accurately.

The potential of AI in early detection is enormous, and in the future, it may enable specialists to predict and track the progression rate of more and more chronic and serious diseases.

Support for emergency medical staff

When emergency services are called to respond, time is of the essence. And one interesting application in this space is Corti, an AI tool that’s able to analyse both verbal and non-verbal cues during an emergency call, detect crucial factors like heart attacks, and alert emergency staff to it. What’s more, Corti trains itself – so the more calls it ‘listens’ to, the more accurate it becomes.

It’s yet another powerful example of the role that AI can play in saving people’s lives. By making diagnoses faster, more accurate and personalised, this breakthrough technology may one day become the quintessential public health tool.