UNSW expert Dr Beena Ahmed says the way we collect and analyse medical and health information in the future could improve life expectancy.
AI and machine learning systems could be used in the future to make predictions on specific health outcomes for individuals based on medical data collected from large populations. Image from Shutterstock
Using artificial intelligence and machine learning on large-scale medical data could be the key to helping us all live longer in the future.
That is the opinion of Dr Beena Ahmed, an Associate Professor at UNSW Sydney, who is an expert in applying machine learning and remote monitoring in healthcare and therapeutic applications.
Dr Ahmed will be taking part in a UNSW Engineering the Future event on March 23 entitled ‘Technology and longevity – how far can we go?’ and believes collecting and analysing large amounts of data using AI could be the next big thing in medicine.
The process could be similar to ChatGPT, which makes predictions in terms of text content based on its huge dataset of words and natural human dialogues.
Read more: Smartphone app to screen for early signs of dementia
In the same way, it may be possible in the future for AI systems to make predictions on a person’s potential health outcomes if a similarly large amount of data had been collected and could be analysed quickly and efficiently.
“My research is related to monitoring health in a non-invasive way, which includes monitoring from sensors on the body, activities being done, even from a person’s speech,” Dr Ahmed says.
“And then after that we need to extract information from the data collected to better inform clinicians and patients, and even healthy people, to help the population stay healthy.
“I think society in general has been quite reactive in terms of healthcare, so we only think about taking action when we realise something is wrong. But medical research is showing that what we do in our 20s and 30s and 40s has a big influence on your outcomes in later life, so it’s becoming ever more important that we monitor our health at every stage.
“We are getting better at collecting data like that, but we are not analysing that data at speed. We may go to our GP who measures our vitals, such as pulse rate, respiration rate and blood pressure, and stores that information electronically.
“The same thing happens if you go into hospital and have some sort of imaging done, as in a CT scan or MRI. The medical team will make decisions on your current health based on those images, but how is that incorporated once you have been discharged?"
Analysing the data
Dr Ahmed highlights the fact that at present there is no way of analysing massive amounts of medical data over a longer period of time, and says the immense power of Artificial Intelligence is so far not being put to use to help health professionals comprehensively evaluate and interpret such information.
“There are some medical researchers building individual AI systems to analyse certain images for certain diseases, such as types of cancer. But that’s just one medical issue in one group of patients in a small number of hospitals," she adds.
“What would be transformative would be for all the medical and health data that is currently being collected to be brought together and for an AI system to be built that can analyse everything as a whole.
“We look at ChatGPT and see what it is capable of producing by using tremendous amounts of data to predict and generate text. Nothing like that is happening in the medical field, even though we’ve probably collected the equivalent amount of data.
Dr Beena Ahmed says a vast amount of medical information that is currently collected, such as an individual's blood pressure, is not being fully analysed across time and across populations. Image from Shutterstock
“If you applied the same sort of models used in that sort of artificial intelligence and machine learning to medical information then you may be able to predict some major health issues for an individual person based on their readings at any given time.
“So, for example, you might be able to assess that someone has a much greater chance of having a heart attack in the near future – and then be able to treat them and prevent the heart attack from happening.”
Technology and longevity
Dr Ahmed will be joined on the ‘Technology and longevity’ discussion panel by Dr. Sze-Yuan Ooi director of the coronary care unit at the Prince of Wales Hospital and clinical lead of Connected Health at UNSW's Tyree Foundation Institute for Health Engineering.
Also on the panel will be Professor Jackie Leach Scully from the UNSW Disability Innovation Institute, and Eugene Salole who is principal at a Sydney consultancy specialising in value-based healthcare and also Adjunct Professor in UNSW’s Faculty of Medicine & Health.
They will debate the impact and challenges of biomedical technology advancements which are aimed at helping people to live longer.
And Dr Ahmed acknowledges that one of the big challenges when it comes to data collection and analysis in the field of medicine is the issue of privacy.
“The single biggest issue for me is how do we ensure data protection of all that medical information we might collect?” she says.
“How do we create a system that makes sharing that data safe, but also easy? There has been some conversation about this, but I don’t think it has been taken seriously enough.
“I think this is something the government or another autonomous body needs to take control of and implement guidelines that everyone must follow. At the moment the data belongs to whoever is collecting it, and that is very often a tech company who can then just sell that information to the highest bidder.
“If governments are really serious about ensuring their people live healthier lives longer, we need to fast-track that data protection issue. Without that, there’s no way you can develop systems that can actually be properly implemented over the long-term to ensure that people stay healthy.”
Dr Beena Ahmed's work focuses on applying machine learning and remote monitoring in healthcare and therapeutic applications.
Dr Ahmed’s own work has pioneered the use of machine learning to detect errors in disordered speech and predict the risk of dementia from speech.
She has also worked on developing novel algorithms to quantify the complexity of the sleep electroencephalogram and detect the presence of sleep disorders such as insomnia to assist clinicians in the diagnosis process.
Her research has also included the use of machine learning to predict mental stress levels using wearable signals from wearable sensors to help users self-regulate their stress levels.
And she believes in the future there will be even more ways for medical experts to collect data from the population.
“I think we will develop more intelligent ways of monitoring people in general – both those with conditions, but also healthy people living their normal lives," she says.
“Wearables, such a health watches, are becoming more popular, but what about monitoring people’s sleep? At the moment, to properly monitor that you need to put electrodes on someone’s head and it’s not very practical.
“But in the future there may be something built into your pillow or in your bed, so you would not even realise there was any monitoring going on.
“If we look way into the future, more than 20 years ahead, we might think about so called lab-on-a-chip technology that could be a tiny device implanted inside your own body that is continually testing and monitoring your health.
“It could be measuring all the vitals that are taken when you visit your GP, but also more detailed things such as the actual quality of your blood, or your cholesterol and sugar levels.
“We need to work out what such a device could be made of so that your body would accept it, and also how it would be powered over a long period of time to be able to collect data and communicate it back outside the body, but research is being done on all those things.”