Introduction
In May this year, it was reported in many UK news publications, including The Guardian, that scientists had used artificial intelligence (AI) to discover a new antibiotic capable of killing a deadly species of superbug.
https://www.theguardian.com/technology/2023/may/25/artificial-intelligence-antibiotic-deadly-superbug-hospital
Using AI, the team of scientists narrowed down thousands of potential chemicals to a handful that could be tested in laboratory conditions. The result was a potent, experimental antibiotic called abaucin. Researchers in Canada and the US say AI has the power to massively accelerate the discovery of new drugs. It is the latest example of how the tools of AI can be a revolutionary force in advances in science and medicine.
To find a new antibiotic, the researchers first had to train the AI. They took thousands of drugs where the precise chemical structure was known, and manually tested them on one of the most problematic species of bacteria, called Acinetobacter Baumann to see which of them could slow it down or kill it.
This information was fed into the AI so it could learn the chemical features of drugs that could attack the problematic bacterium. The AI was then unleashed on a list of 6,680 compounds whose effectiveness was unknown. The results showed it took the AI an hour and a half to produce a shortlist. The researchers tested 240 compounds in the laboratory and found nine potential antibiotics. One of them was the incredibly potent antibiotic abaucin.
It is well known that antibiotics kill bacteria but there has been a lack of new drugs for decades and bacteria are becoming harder to treat, as they evolve resistance to the ones we have. This development offers much hope that AI can significantly accelerate and expand research for novel antibiotics More than a million people a year are estimated to have died from infections that resist treatment with antibiotics according to University of Oxford:
An estimated 1.2 million people died in 2019 from antibiotic-resistant bacterial infections | University of Oxford
How can AI be used responsibly in healthcare?
On 29th March this year, UK Secretary of State for Science, Innovation & Technology, Michelle Donelan, MP announced the launch of the government’s white paper to guide the use of AI in the UK to drive responsible innovation and maintain public trust in this revolutionary technology.
Five principles, including safety, transparency and fairness will, according to the government, ‘guide the use of artificial intelligence in the UK, as part of a new national blueprint for our world class regulators to drive responsible innovation and maintain public trust in this revolutionary technology.’
AI is already delivering real, social and economic benefits for people, including helping doctors to identify diseases more quickly, so it is clear that it has the potential to be a real gamechanger in public health. The UK government has said it wants AI and data to “transform the prevention, early diagnosis and treatment of chronic diseases by 2030”.
The National Health Service (NHS) AI Lab – transformative healthcare in action
The NHS AI lab was set up by the UK government in 2019 with a £250m fund. Its goals are to reduce clinicians’ workloads, give patients tools to access services directly, ensure clinical information can be accessed safely where needed, and enhance patient safety. It’s doing this by bringing together government, health and care providers, academics and technology companies.
https://transform.england.nhs.uk/ai-lab/
The NHS Lab’s AI roadmap seeks to ensure the development and adoption of safe, effective and ethical AI in health and care and includes The AI in Health and Care Award, making more than £123m available over 4 years to accelerate the most promising AI technologies for health and social care. Some notable winners of the award are as follows:
• Deep learning for effective triaging of skin disease in the NHS (University of Dundee). This project is developing an AI (deep learning) system to distinguish common benign skin lesions from common skin cancers with state-of-the-art accuracy. This research will develop the system with representative image data from NHS clinics.
• Development of AI techniques to predict eye cancer using longitudinal data(University of Liverpool). This project will further develop a novel, fully automatic AI-powered diagnostic tool to support the accurate diagnosis and monitoring of choroidal naevi (patches of pigment at the back of the eye) and to predict the risk of ocular melanoma, the major form of eye cancer. The aim is to help to streamline the management of patients and reduce cost for the NHS by assessing and monitoring in the community for low-risk lesions and follow up conditions with high risk factors in secondary care.
• R-CANCER (Imperial College London). R-CANCER will improve the quality of decisions made by doctors when deciding how best to detect and diagnose cancer, by intelligently collating, analysing and interpreting new data on cancer from academic and open data sources.
The development of AI in healthcare – some other examples
Since 2020, Addenbrooke’s hospital in Cambridge has been using Microsoft’s InnerEye system to automatically process scans for patients with prostate cancer. The system takes a scan image, outlines the prostate on the image, marks up the presence of tumours and reports back. This is giving clinicians a very good understanding of where tumours are present and speeding up prostate cancer treatment.
Whilst machine learning models developed using the tool need to be tested and validated in each individual healthcare setting, doctors at Cambridge University Hospitals (CUH) have demonstrated how the technology can be applied in clinical settings.
https://innereye.ai/
HeartFlow’s AI technology is also being used in the NHS. This system analyses CT scans of patients who are suspected of having coronary heart disease and then creates a personalised 3D model of the heart that shows how blood is flowing around it. This helps doctors spot where blood flow is disrupted by blockages. This non-invasive personalised cardiac test provides unprecedented visualisation of each patient’s coronary arteries, enabling physicians to create more effective treatment plans for their patients. HeartFlow analysis is less costly and intensive than the standard angiogram procedure.
https://www.heartflow.com/
The “C the Signs” app is being piloted by GPs and practise nurses in the UK. The tool uses AI to help GPs and nurses check combinations of signs, symptoms and risk factors during patient consultations so that they can identify patients at risk of cancer earlier. The tool also helps to identify which referrals and investigations the patient might need. C the Signs is fully customised to cancer pathways, and diagnostics across each hospital servicing primary care, ensuring that at point of referral, GPs are made aware of the correct pathway to use.
https://cthesigns.co.uk/
Summary and conclusions
Following the government’s white paper in March, UK Secretary of State Michelle Donelan, MP further announced in August that there would be 22 projects to explore how to develop and use AI in health. This comes on the same day that the government appointed two leading experts to spearhead preparations for the UK to host the first major international summit on the safe use of AI. The projects will involve universities stretching from Edinburgh to Surrey. They will be supported by £13 million from UK Research and Innovation’s (UKRI) Technology Missions Fund, previously announced in the Science and Technology Framework, to support AI innovation to accelerate health research, showing that AI in healthcare will continue to grow and enhance healthcare. Source: Med-Tech Innovation News:
Government announces 22 projects worth £13m exploring AI in health (ampproject.org)
There are many advantages and disadvantages to using AI in healthcare. Among the advantages are improved diagnostics and treatment though analysis of vast amounts of data, increased efficiency by automating routine work, the ability to offer personalised medicine and reduced cost through automation of routine tasks.
Those advantages must be weighed up against some of the disadvantages such as bias and discrimination (algorithms can only rely on the data they are provided), a lack of human involvement which is essential to build trust with patients and legal & ethical issues, e.g; a mistake that leads to a bad patient outcome, and concerns over control of the data.
The benefits for patients are clear to see, with earlier diagnosis and management of complex medical problems, leaving them better informed and able to access the right treatment pathway for their condition. The life insurance industry can benefit greatly from AI, as patients buying protection products are likely to be much better informed about the nature of their condition and can access the right treatment more efficiently. This should mean that applicants for life insurance can make even more accurate disclosures accelerating the underwriting decision and at claims stage when diagnoses and treatments can be important in determining the claims decision.
David Ferguson
Underwriting and Claims Development Manager
For further information about this article, please contact David Ferguson.
