Connect with us

Insight

The growing importance of machine learning and AI in healthcare

Published

on

Cloud Technology Solutions’ Alison King on the uses of nascent technologies in healthcare and how the NHS can learn from the UK’s European neighbours.

The NHS is constantly looking for ways to improve effectiveness across the organisation. Whether that’s through innovations in technology or changing policies to streamline ways of working, maximising productivity has always been key.

But that uphill battle to drive effectiveness has become steeper in the wake of the coronavirus pandemic. With millions of procedures put on hold, carrying out the backlog of operations and treatments will require staff productivity levels to be at an all high time high, and further investment will be needed in areas like recruitment and capacity.

But for industries like healthcare, where staff are some of the most highly skilled and trained professionals of any sector, recruitment drives take years, not weeks. NHS leadership teams will therefore need to look at other areas to increase productivity – such as technology.

The NHS has always been an avid adopter of new technologies. The inception of NHSX in July 2019 was a great step forward in the NHS’s digital transformation journey and it has since invested considerable sums of money into technology. For instance, in summer last year, NHSX pledged £250m for a National Artificial Intelligence Lab to improve diagnostics and screening across the healthcare system.

Nascent technologies like artificial intelligence (AI) and machine learning (ML) are increasingly being used across all industries. Banks are using ML to help identify and block fraudulent activity on customer accounts in real-time, while retailers are using the technology to help identify trends and manage stock levels. The NHS is following suit with its approach to digital adoption too.

Last year, Google worked with Cancer Research UK Imperial Centre, Northwestern University, Royal Surrey County Hospital and DeepMind to see how AI could help radiologists spot the signs of breast cancer more accurately.

The AI technology analysed mammograms of more than 76,000 women in the UK and 15,000 in the US to learn how to spot the signs of breast cancers in scans. The system was able to reduce the number of false positives by 5.7% in the US and 1.2%, as well as a 9.4% reduction in false negatives in the US and 2.7% reduction in the UK.

Learning from Europe

The possibilities of these technologies are endless. And right across Europe, healthcare providers and institutions are using nascent technologies to help with administrative tasks right through to detecting diseases.

Accurate and timely note-taking is one of the most important administrative tasks for any healthcare professional. But at Leiden University Medical Centre (LUMC) in the Netherlands, leadership teams noticed that their clinicians were spending as much time typing up patient notes as they were on patient care.

To stop administrative tasks hampering productivity and getting in the way of delivering the best care, the organisation turned to ML. By recording a patient consultation, LUMC is able to transcribe it using ML technology. The organisation taught the technology to recognise medical terms, allowing it to provide instant analysis on the patient’s illness. Parts of the conversation are also automatically sorted into answers to the anamnesis questions, further streamlining the consultation process.

Following the consultation, the clinician can upload the notes onto the patient’s records instantly.

Meanwhile also in the Netherlands, ML is being used to help slow down the effects of a stroke. One in six people suffer from a stroke at some point during their lifetime, and the longer a person is left without treatment, the more damage can be done.

To help speed up the treatment process, algorithms were developed that could quickly analyse CT scans and detect deviations. To ensure the scans could be uploaded quickly, Google Cloud Platform was used so that the images were quickly accessible for the doctors and the ML technology to analyse.

By utilising these systems, doctors are able to save crucial time that can significantly limit the amount of damage caused by the stroke.

These two examples, while vastly different in how they help, have provided an intelligent and efficient way of sorting two very separate problems within their own organisations. And they highlight just how versatile ML and AI can be when used in different scenarios.

Looking ahead for the NHS, utilising technologies available to them will be a core facet of their recovery from the coronavirus, as Trusts work through the backlog of procedures while also continuing to battle through the pandemic itself.

Technology won’t be the silver bullet for getting procedures back on track and enhancing effectiveness, but it can certainly be an important weapon in the NHS’s arsenal.

Alison King represents Cloud Technology Solutions as account manager for NHS and Government.

 

Continue Reading
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *