UK to embrace machine learning and AI in healthcare
The power of machine learning and AI in healthcare is set to make a difference in the UK through data interoperability and cloud computing
The power of machine learning and AI in healthcare is set to make a difference in the UK through data interoperability and cloud computing
In two separate updates, the UK is exploring the application of machine learning and AI in healthcare.
£240,000 has been made available through an Innovate UK Small Business Research Initiative (SBRI) competition, with funding from Opportunity North East and NHS Scotland, to investigate the use of AI and machine learning in the NHS. The competition will explore the usage of machine learning and AI in healthcare, particularly to support limb radiographs in the diagnosis of fractures. Possible improvements include diagnosis accuracy and treatment and increased productivity in radiology departments.
Similarly, Dr Kenji Takeda, Director of Academic Health and AI Partnerships at Microsoft Research in Cambridge, told the All Party Parliamentary Group (APPG) on Heart and Circulatory Diseases that helping the medical sector is one of the most important uses of AI in healthcare. A survey by APPG found that people in the UK support using more technology in the healthcare sector, with 85% of respondents backing the use of AI in diagnostics and treatment, and 86% saying they were happy for their anonymised health data to be shared to better diagnose medical conditions. The event was attended by the Secretary of State for Health and Social Care Matt Hancock and Simon Gillespie, Chief Executive of the British Heart Foundation.
An NHS hospital in Scotland is improving care for people with Chronic Obstructive Pulmonary Disease (COPD) by leveraging AI and machine learning to bring an easier way to manage the condition from the comfort of patients’ own homes.
X-rays of arms and legs are among the most frequent diagnosis processes used by NHS Scotland, with around 5,000 procedures annually.
Although injuries in these areas are often categorised as minor, misdiagnosis and mismanagement can hamper recovery and lead to financial cost. However, the use of artificial intelligence (AI) and machine learning could help create systems that prevent misdiagnosis.
Projects applying for the funds available through SBRI competition funded by Opportunity North East and NHS Scotland must use a dataset of peripheral limb X-rays and reports from the University of Aberdeen to develop AI algorithms that will interpret the current text-based report to correctly categorise fractures and use radiograph images to identify the presence of fractures to ensure the AI product can function at real-world accuracy.
The competition will comprise 2 phases. In phase 1, applicants will conduct technical feasibility studies on their proposed solution. Up to £100,000 including VAT is available in phase 1, and as many as 5 projects are expected to be funded. Phase 2 will include prototype development and evaluation. Up to £140,000 including VAT is available at this stage.
Successful applicants will be able to receive input from NHS Grampian, NHS Greater Glasgow and Clyde, the University of Aberdeen and Canon Medical Research Europe.
A report released by APPG at an event at the Houses of Parliament revealed that there are seven million people living with heart and circulatory diseases, such as coronary heart disease and vascular dementia, in the UK and they cause a quarter of all deaths.
It found that there was huge potential for AI to transform the lives of those people and a greater need for them to be included in discussions about the development and adoption of new technology.
The report suggests that the focus of the inquiry is on heart and circulatory diseases however the recommendations and discussion are applicable more broadly.
Hancock said: “There is no doubt the technology that has been developed, including AI, has huge potential for saving and improving people’s lives.
“Data improves the technology, it improves prediction and prevention and knowing which treatments work better. Using patient data improves treatment for everybody because every time there is a new data point, we can learn from that and improve.
“But to get the most out of this opportunity, you need people’s consent, and that requires trust. In order to have trust, we need to have strong rules around privacy and cyber security, so people know the data will be safe. We also need to explain why it’s important for people to allow that information to be used for research purposes.”