Source:
iStock
Decentralisation, data sharing and federated learning have the potential to reduce the cost and improve the efficacy of delivering consultations and treatments
Healthcare’s digital evolution has accelerated in recent months, as the coronavirus pandemic made alternatives to face-to-face consultations a matter of necessity. Sebastien Ourselin, head of the School of Biomedical Engineering and Imaging Science at King’s College London, believes that telemedicine – physicians offering clinical care via online video platforms such as Zoom, Skype and Microsoft Teams – is “one of the biggest, most profound changes we will see” in healthcare.
He predicts that if and when a Covid-19 vaccine is developed, telemedicine will remain the standard format of many consultations. “I think what Covid-19 has done is make people really start to appreciate that this can be done and it works,” says Dr Ourselin. “The technology has been around for a very long time, but how do you make change at a community level, at a countrywide level? It is usually very slow.”
Dr Ourselin describes this as a low-tech revolution. As such, it is paradoxical to speak of telemedicine, which exploits technology that is a decade old or more, as a healthcare frontier, but the cultural change required for people to buy into consultation via a screen is revolutionary. Like many physicians, Dr Ourselin would prefer face-to-face consultations, where subtle cues in body language can aid diagnosis, but he sees the widespread adoption of video platforms as an opportunity to improve video quality and digital platforms’ features. This would increase the efficacy of telemedicine consultations, which would in turn enable other healthcare systems to benefit.
Right now, telemedicine is changing how we see our GPs. Eventually, Dr Ourselin predicts that hospital consultations will follow suit.
“You won’t have to wait,” he says. “You want to build a healthcare system which is financially sustainable but is also friendly to patients, and often the best thing you can do for a patient is avoid the hospital. Hopefully, this is going to reduce the cost of delivering healthcare.
“We have to be honest,” he continues, “we have been in a financial crisis with our healthcare system since the mid-1980s, when we spend more money as a percentage of GDP into healthcare. It’s just going up and up and up.”
The development of new systems to revolutionise diagnoses and patient treatment will be critical in securing the future of the NHS, says Dr Ourselin. He foresees technology as a driver of decentralisation, placing high-tech diagnostic imaging equipment into high-street settings.
“A lot of places have OCT [optical coherence tomography] that images the back of your eye,” he says. “Until a few years ago, if you wanted access to such technology, you had to go to a specialised hospital like Moorfields [Eye Hospital, London]. This technology is now at your optician. The same thing will happen with large imaging equipment like MRI scanners. In 15 years’ time, perhaps, at the back of a Boots pharmacy you will have an MRI scanner which will not require any expertise to run.”
The importance of improved diagnostics should not be underestimated. New diagnostic tools that take advantage of big data and artificial intelligence can relieve pressure on healthcare systems. The NHS is a case in point. As Dr Ourselin explains, introducing systemic change within the NHS is difficult because it is a large, complex institution. “You go to a hospital and you realise that you have got over 300 different databases and some of them are not talking to each other. You need to have 25 different passwords to access different systems in your day-to-day work looking after a patient. Transferring the data is one of the biggest issues.”
To rectify this, Dr Ourselin advocates a system of federated learning, where hospitals store data securely on their own servers, but healthcare researchers and providers are given access to it, running diagnostic tools to make effective medical interventions. As with telemedicine, a cultural change needs to occur before such a model can be adopted. “The barrier to progress is not technology,” says Dr Ourselin. “It is infrastructure, and changing the modes of working. Federated learning is what is going to make the concept of data in healthcare successful. We should build the infrastructure and link all these hospitals together. But it is a very different philosophy, a very different model.”
Sebastien Ourselin was a participant in the UK Academic Salon 2020.
Learn more about Huawei and higher education.
