Accessibility Links
Cookies on our website
By continuing to use this website we will assume you are happy to receive cookies as outlined in our cookie policy
Accept Policy

Applying Artificial Intelligence within the NHS


Computers are taking on ever more complex roles in healthcare. Through the concept of Artificial Intelligence (AI) - intelligence displayed by machines in contrast with natural intelligence displayed by humans – computers are already managing patient appointments and supporting research and analysis, having been programmed to make informed decisions.




With some £120m of NHS cash earmarked to speed up the implementation of AI, that could eventually see machines diagnosing illnesses, reading x-rays, and analysing tissue samples and other information collected from patients about their symptoms.

AI could play a role in the NHS

Senior NHS figures suggest they could soon take on some routine roles currently conducted by doctors and nurses within a few years. NHS medical director Sir Bruce Keogh said recently that studies had shown that “in certain circumstances, AI is better than doctors at diagnosing certain conditions” and indicated that AI could soon play a role within the NHS. In the United States, scientists are using AI to identify some conditions.

AI predicts breast lesions

Research conducted by scientists at Harvard Medical School, the Massachusetts Computer Science and Artificial Intelligence Lab, and Massachusetts General Hospital, showed how AI could predict whether breast lesions identified from a biopsy will turn out to cancerous.

The machine learning system has been tested on 335 high-risk lesions, and correctly diagnosed 97% as malignant and reduced the number of unnecessary surgeries by more than 30%.

The system was trained on information about such lesions and looked for patterns among a range of data points, such as demographics, family history, biopsies and pathology reports.

Shifting analysis from doctors to machines

Research also suggests greater use of AI could lead to a shift in radiology and pathology from doctors to machines.

A Lancet article, for example, noted that over the course of a career a radiologist may read 10 million images, a dermatologist might analyse 200,000 skin lesions and a pathologist review 100,000 specimens and that eventually AI may be able to automate “huge swathes” of that work.

AI removes healthcare drudgery

One UK expert, Dr Ameet Bakhai, believes AI will revolutionise the delivery of healthcare as it removes the “drudgery” of routine tasks, speeds up the process of identifying clusters of patients by digging deep into electronic health records, and even replaces staff roles for some tasks.

A consultant cardiologist at the Royal Free London NHS Foundation Trust, Dr Bakhai said AI is already offering solutions to clinicians in areas such as diabetes, self-management of epilepsy and rare diseases. Elsewhere, AI is scanning records and interrogating databases to identify patients that meet certain criteria for clinical trials.

However, while AI has potential within an NHS setting, concerns remain over security and safeguarding confidential patient data with so much of it being processed by machines.

What do you think about the idea of using artificial intelligence within healthcare settings? Let us know using the comments box below. 


Email a friend

Meet the Events Manager

Add new comment
 
ISO Logo   REC Member Logo Investors in People Logo CHAS Logo