NHS England has been conducting pilot lung cancer checks supported by artificial intelligence technology.
According to Cancer Research UK, lung cancer is the most common cause of cancer deaths in the UK, but there are hopes around the impact of early detection, with one study finding that CT scan chest screening catches around 70 per cent of cases early, at a stage when patients can still be potentially treated and cured.
The NHS England Targeted Lung Health Check (TLHC) programme is a large-scale lung cancer screening initiative and, so far, from 19 pilot lung checks that have been rolled-out, 15 of those involved have used lung nodule management solutions developed by the health tech company Aidence, who also recently partnered with AstraZeneca to work on enabling and increasing early lung cancer diagnosis.
The TLHC programme offers low-dose CT scanning to eligible patients, aiming to to detect early signs of cancer and potentially save lives.
Within this process, Aidence’s technology – Veye Lung Nodules – is said to be able to reduce some of the additional workload pressures, by supporting radiologists with an AI solution that “automatically detects, measures, classifies and tracks the growth of pulmonary nodules on chest CTs” – tasks which can otherwise be time-consuming when performed manually.
Veye Reporting, a product add-on to Veye Lung Nodule, also prepopulates a report of AI findings, enabling radiologists to follow the reporting protocols required by the lung health checks. In addition, to further support the screening projects, the company has also been working with the British Society of Thoracic Imaging to provide lung nodule workshops.
Dr Oliver Byass, Clinical Director, Radiology at Hull University Teaching Hospitals NHS Trust, one of the first sites involved in the TLHC, said: “Early detection can make a real difference to patients’ lives. However, we have to be wary of the pressures on our radiology teams.”
The results of the TLHC pilot will determine the scope of nationwide lung cancer screening, with the potential that it could become part of long-term public health programmes.
Dr James Shambrook, Consultant Cardiothoracic Radiologist at University Hospital Southampton and Radiology Lead for the Southampton Targeted Lung Health Check, commented: “Healthcare is changing and we want to drive this change by exploring the opportunities of AI solutions and being a part of their development. Scale and precision are of utmost importance for making the Lung Health Check programme a success, and we see Aidence as a valuable partner for the years to come.”
Mark-Jan Harte, CEO of Aidence, added: “The UK is the first country in Europe to implement a screening initiative at scale, and we look forward to helping the NHS make it a resounding success.”
AI pilots have gained more prominence in recent years, including those being launched as part of NHSX’s Artificial Intelligence (AI) in Health and Care Award – part of the organisation’s AI Lab – which is making £140 million available over four years to accelerate testing and evaluation. While, UK Research and Innovation (UKRI) also recently named its first five Turing Artificial Intelligence World-Leading Researcher Fellows, who have been selected to tackle challenges including ‘decision-making in personalised medicine’ and ‘synthetic biology and drug design’.
Other recent news involving AI in cancer detection includes University College London scientists’ collaboration with the spinout company Odin Vision and clinicians from University College Hospitals London, on a project which uses AI to help detect early signs of oesophageal cancer, through a system called CADU that supports doctors to identify cancerous tissue.
Elsewhere, the University of Leeds has been working with Roche Diagnostics to develop a new AI algorithm that could help clinicians measure proteins and decide on the best treatments for patients with bowel cancer, while University of Cambridge scientists combined AI diagnostics with a new technique called Cytosponge to help diagnose Barrett’s oesophagus.
