Artificial Intelligence: How it Works in Healthcare? – SPICe Spotlight

Machine Learning


Reading Time: Four Minutes

The Scottish Artificial Intelligence Strategy, published in 2020, noted that Scotland has the potential to become the “go-to place” for AI and the healthcare market. There are currently multiple AI pilots running across NHS Scotland, with a small number of AI tools already being used in clinical settings.

SPICe recently published an overview on Artificial Intelligence and Healthcare in Scotland. This blog summarises some of the key themes from that overview and focuses on the current and near future use of AI in NHS Scotland. This is the second blog in a series of publications by SPICe on AI. You can read the first part here.

What is AI?

Scotland’s AI Strategy defines AI as:

Technology used to enable computers to perform tasks that require human intelligence, such as visual recognition, speech recognition, and language translation.

AI is based on algorithms. An algorithm is a set of instructions used to perform a task, such as analysis or calculation, usually on a computer. Recently, AI has made great strides through machine learning, in which an AI system is first given a large amount of training data (such as photos of x-ray scans with and without tumors). The AI ​​uses statistical methods to infer rules from the training data and then uses these rules to perform a task (such as classifying a scan as benign or worrisome). This method enables AI systems to successfully perform tasks in areas such as image recognition and speech recognition, where it is very difficult to achieve the same results with traditional programming.

For a detailed explanation of how AI works, check out SPICe’s recent blog.

How is AI being used in NHS Scotland?

At the time of writing, a small number of AI tools are already being used in clinical practice at NHS Scotland, this section outlines most of them, although it may not be a comprehensive list.

BoneXpert is a tool that uses AI to automatically determine a child's bone age from hand x-rays, helping doctors spot potential growth disorders and other medical conditions. The tool is used widely in the UK, including at NHS Greater Glasgow and Clyde and the Royal Children's Hospital in Edinburgh.

Ethos is a system that uses AI to automatically target and adapt radiation therapy delivery in cancer treatment. When a patient has cancer, it's normal for the tumor and surrounding tissue to change as treatment and the disease progresses. Traditionally, doctors have spent a lot of time continually adjusting treatment plans. The Ethos system uses AI to help doctors make these changes more quickly. Ethos is currently being used at Beatson West of Scotland Cancer Centre.

Some of the latest medical imaging equipment is already using AI as a built-in feature – for example, the computed tomography (CT) scanners at Clydebank's Golden Jubilee Hospital use AI to deliver clearer images in less time than before.

Voice recognition software has also improved significantly thanks to recent advances in machine learning, and healthcare professionals in Scotland and other parts of the UK are now using such software to speed up repetitive tasks relating to clinical notes and letters.

What other uses for AI might we see in the future?

In addition to the few AI tools already in use, a number of AI applications are currently being tested within NHS Scotland, including applications in medical imaging, disease prediction and prevention, and automation of administrative tasks. In this section, we'll highlight one current example from each of these areas.

Medical Imaging

NHS Grampian recently tested an AI tool called MIA, developed by Chiron Medical Technologies, for reading mammograms. The hope is that using AI in breast cancer screening will improve the accuracy of tests, reduce waiting times and ease workload. Currently, standard practice is for each scan to be read by at least two experts. In the future, AI could take over one of the two readers.

During trials, MIA spotted signs of cancer in 11 women that were missed by human doctors, according to early findings reported by the BBC. Trial data is currently being analysed by researchers at the University of Aberdeen, with final results due to be published later this year. The research was funded by NHS England's AI in Health and Care Award.

Disease prediction and prevention

NHS Greater Glasgow and Clyde is currently running a project called Dynamic-AI that aims to explore whether an AI tool can improve care for patients with COPD (chronic obstructive pulmonary disease), which affects around 120,000 people in Scotland and is the second most common reason for emergency hospital admissions. The idea is that AI could predict worsening symptoms and prevent hospital admission through early intervention.

The Dynamic-AI research is being carried out in collaboration with the University of Glasgow and industry partners Lenus Health Ltd and Storm ID, and is funded by the AI ​​in Health and Care Award. It will run until January 2025.

Administrative tasks

The Scottish Government announced in May 2024 that a new type of operating theatre scheduling software will be introduced nationwide in a bid to reduce waiting times. This comes after successful pilots in three NHS boards and evaluation by the Scottish Health Technology Group.

Developed by Infix, the software uses machine learning to optimize operating room schedules and, during pilot runs, increased OR efficiency by up to 25% without requiring additional medical staff or rooms.

Who is driving AI development in Scotland?

The development of AI tools for healthcare often involves partners from academia, industry and NHS Scotland. To facilitate such collaboration, NHS Scotland has three Regional Innovation HubIt covers all 14 local NHS boards. The hub is: Office of the Chief Scientist.

Many national NHS boards are also playing a role in AI innovation. For example, the NHS Golden Jubilee included: Sustainable Delivery Center The Computation for Sustainable Development (CfSD) is involved in several AI pilots. CfSD is also Accelerated National Innovation Adoption (ANIA) Pathwayaims to speed up the adoption of new and proven technologies across the Scottish National Health Service (NHS).

Health Improvement Scotland (HIS) Scottish Medical Technology Group The SHTG is an organization that evaluates and recommends new technologies, including AI. SHTG's evaluations are as follows: Multidisciplinary Health Technology Assessment Methodology (HTA)The ANIA Pathway may refer to technologies that SHTG assesses.

assignment

Despite promising developments, the use of AI in healthcare also raises several challenges, including issues around safety, bias, explainability, data protection and legal liability. Many of the promising results with AI have yet to be widely replicated in real-world clinical settings, presenting challenges for local implementation.

These issues are discussed in further detail in a recent SPICe briefing on AI and healthcare in Scotland and we will also explore these issues in more detail in part three of this blog series, to be published next week.



Source link

Leave a Reply

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