Is AI the key to effective and sustainable lung cancer screening? (Part 1)

The benefits and impact of lung cancer screening initiatives

Last year, I felt like I’d let people down after deciding to leave my clinical radiology training post. Recently, however, I found myself proud and excited to have the opportunity, with my current role, to support radiologists involved in what I believe could greatly improve people’s lives: lung cancer screening programmes.

End of January, I stood in front of an audience of healthcare practitioners at a conference hosted by the British Institute of Radiology, talking about the work we are doing to support NHSEs targeted lung health checks. Beyond the proven benefits of screening initiatives, I feel we should also focus on improving the work-life of radiologists, and I highlighted ways to do that in my talk.

In this blog series, I have gathered ideas and information about the main developments in lung cancer screening and built a case on how AI can make it successful. I understand that screening is likely to be delayed due to the COVID-19 pandemic, but it remains important to plan these programmes for when the time is right.

An increasingly relevant topic of discussion

Lung cancer is the leading cause of cancer-related deaths worldwide. This is largely due to symptoms developing late, thus delaying the diagnosis until the disease is already advanced. Treatment options are more effective for early-stage lung cancer, so, logically, research has shown that screening asymptomatic at-risk individuals can improve survival rates by increasing early detection and diagnosis.

Until recently, screening could only be done using chest X-rays (2D imaging), which unfortunately aren’t sensitive enough for detecting early, subtle disease. But advances in CT technology have enabled reduced CT radiation exposure whilst achieving good quality (3D) imaging of the lungs. Low-dose CT scanning has become a particularly hot topic in recent years, contributing to the discussion around the benefits of lung cancer screening.

Lung cancer screening advocacy and pilots

Targeted lung cancer screening pilots have been and are being implemented in Europe and the US. In addition, an increasing number of publications are bringing forward evidence for the efficacy of screening and advocating for screening initiatives in Europe. Here is an overview of the main recent developments:

  • In 2014, the US initiated one of the first national lung cancer screening programmes using low dose CT scanning. However, they have reported a low uptake of the programme by eligible individuals (<20%).
  • Smaller, local trials, such as in Manchester (UK), adopted new methods of mobile scanning units to make the programme more accessible for the harder-to-reach population. By setting up the scanners in supermarket car parks and next to football grounds, they yielded a good uptake (>50%).
  • In February 2019, the NHS announced a multicentre, four-year pilot programme across 10 regions in England who have some of the highest rates of mortality from lung cancer. The pilot is a stepping stone to initiating a national programme.
  • At the beginning of 2020, the well-known NELSON study was published in The New England Journal of Medicine. The Dutch-Belgian trial provides evidence that supports the efficacy of targeted, low-dose CT lung cancer screening in at-risk populations.
  • Shortly after, the European Society of Radiology (ESR) and the European Respiratory Society (ERS) released a new joint statement recommending the introduction of targeted lung cancer screening programmes across Europe.
  • Also recently, Croatia has introduced a nationwide lung cancer early detection programme for all at-risk individuals.
Promotional material for the NHSE Lung Health Checks, retrieved via Twitter.

Why some countries remain hesitant

Other countries are hesitant to launch lung cancer screening initiatives. For governments to introduce nationwide screening, the potential benefits to the target population must outweigh the potential risks. In the case of lung cancer, this means being able to accurately diagnose and treat early-stage lung cancer and therefore improve survival rates. But there are possible risks or disadvantages to consider:

  • Most pulmonary nodules are benign and would go unnoticed if not scanned for; detecting them can cause unnecessary concern for otherwise-well individuals.
  • There are risks associated with biopsy procedures, therefore minimising the number of false positive nodules reported, and thus reducing benign biopsies, is essential.
  • The risks associated with radiation exposure, albeit low doses, must be taken into account.
  • The cost of a screening programme needs to be justified for it to be endorsed by governments. Complex cost-benefit analyses to demonstrate the potential long-term savings (e.g. reduced need for expensive advanced-stage treatments) play a part in agreeing to fund the set-up and running of screening programmes.
  • Finally, a screening programme requires a significant, long-term commitment in order to achieve the intended population health benefits. The Italian MILD trial demonstrated that prolonged screening beyond 5 years can enhance the benefit of early detection and achieve a greater lung cancer mortality reduction.

The impact on healthcare

The increased volume of work that screening programmes generate demands a significant amount of planning and availability of resources to ensure the programme runs effectively.

Lung cancer screening requires input from a number of healthcare professionals involved in the end-to-end lung cancer pathway, including GPs, Radiologists, Radiographers, Nurses, Physicians, Oncologists, Surgeons, IT, Administrative staff. And, of course, to achieve low-dose imaging, the country must have access to specialist equipment, such as CT scanners – the availability of which will depend on access to funding.

Specific concerns will vary depending on the workforce capacity of each healthcare system. In the UK, for example, the NHS workforce is already under strain – with a particular shortage of radiologists (and reporting radiographers) – and the UK pilot programme is predicted to generate 200,000 additional CT scans over four years. Detailed reporting protocols add up to the workload, making practitioners wary of the pressure on the radiology teams.

AI can support radiologists

Although the benefits of lung cancer screening are proven, the concerns around the efficacy and sustainability of these programmes are also understandable. How can AI-driven solutions support radiologists and contribute to successful lung cancer screening?

Find out in part two of this blog post.

About Lizzie

Lizzie Barclay

Lizzie Barclay is Medical Director at AidenceLizzie is originally from Manchester, UK. After graduating from the University of Leeds Medical School (MBChB), and Barts and the London School of Medicine (BSc sports & exercise medicine), she spent four years working as a doctor in Manchester and Liverpool NHS Trusts, including two years in Clinical Radiology. Lizzie’s areas of interest are thoracic radiology and medicine, innovation, and improving patient outcomes and healthcare professionals’ wellbeing. She has presented her work on lung cancer imaging at national/international conferences, and recently contributed to Lung Cancer Europe’s “Early Diagnosis and Screening” event at the EU Parliament in Brussels.

Connect with Lizzie on

Follow Aidence on Medium

Book a demo


    * Required fields.
    For more information on how to unsubscribe, our privacy practices, and how we are committed to protecting and respecting your privacy, please review our Privacy Policy.