Most Recent Blog Post:
April 2021: How artificial intelligence helped the NHS make better decisions during COVID-19
Author: Marco Antonio Vieira Marques - Marketing officer
Hi I’m Marco, a first year medical student at Warwick and one of the friendly committee members of Warwick SMT. I do the media side of things which is quite fun, and try to help spread the word about the events, internships and opportunities we are eager to bring to you.
Like many of you I've only just started dipping my toe into the world of med tech. It's a rapidly evolving and dynamic world of medicine that I'm wanting to explore, but at the same time, it's complex, vast and that feels quite intimidating at times. I want to share a little bit about what I've learnt so far about how machine learning and AI is currently developing and growing in the UK and in the NHS, and my hope is to make the subject accessible to anyone interested in this kind of exciting science, wanting to learn more! So here's a story about how COVID-19 has given a boost to using artificial intelligence more, in everyday medicine.
Imagine you're a doctor, working in intensive care, and you need to manage the equipment you have available in the hospital, balancing that with the needs of your patients in that ward. You might not have to worry too much usually if there's plenty of equipment to go around, but then a coronavirus pandemic ensues, and you need to be more strategic to manage the ward, and ultimately, give patients the best chance of surviving.
You might ask questions such as:
- Which of these two patients will get the most benefit from going on a ventilator today?
- Which patients are most likely to need ventilators within a week?
- How many free ICU beds is this hospital likely to have in a week?
Answering these questions normally means having to make an informed decision, based on the best available evidence, guidance on a particular disease, and statistical models in place to determine good decisions. What machine learning and artificial intelligence can bring about, is an answer to those questions, on the go, and more accurately than a statistical model could. By developing programs which work with data on the patients and the hospital equipment, artificial intelligence can be asked the questions, and provide the answers. This idea, was made a reality in 2018 at the Cambridge Centre for AI in medicine. Researchers there created a programme called AutoPrognosis which specifically addressed the need to manage patients and hospital resources, to help medical teams make better decisions for their patients.
It helped answer questions such as:
- When should a patient be discharded from that ward?
- What patients should go on ventilators, and for how long?
- Which patients should be sent to the general ward instead of the ICU?
Then in 2020, the NHS started working with Cambridge University and Public Health England (PHE) on this technology, and produced the Coronavirus (COVID-19) Capacity Planning and Analysis System (CPAS). CPAS is one of the first machine learning-based systems to be deployed in hospitals on a national scale to address the COVID-19 pandemic. I think it's really exciting that this is happening, as programmes like this can translates to so many areas of health care. The hope is that adapting the CPAS programme to other areas in hospitals and even in doctors surgeries, could mean health professionals will be able to improve the outcome of their patients by making better decisions. This would also help make cost savings in the long run, as services and equipment is used more efficiently, and patients get better quicker, meaning they spend less time in hospital.
Another thing I like about using programmes to support decision making in hospital, is that it has the potential to reduce health inequalities, or uneven treatment, by reducing bias in healthcare decisions. I really hope that the NHS and government continue to invest in this kind of technology. There hasn't been much news or talk of how CPAS is doing. A search on the NHS Digital website only shows a handful of results dating back a year ago. It would be a shame for that technology not to be invested in. The pressures of the pandemic on healthcare gave the NHS a push to develop these technologies, now the question is; will the pressures of the health needs of the people, be enough for the health service to continue developing these technologies.
If you are interested in collaborating, becoming part of the committee, or getting involved with our events, please contact warwickSMT@gmail.com
You can find out more about Warwick SMT here: https://warwick.ac.uk/fac/sci/med/study/ugr/mbchb/societies/smt/
If you're interested in reading any of the previous blog posts, find them on the links below: