Lung cancer risk could be estimated accurately and cheaply by using artificial intelligence systems, British experts say.
Dr Tom Callender of University College London, UK, and colleagues looked into the best way of screening for lung cancer in those at high risk. This could reduce the risk of death by up to 24%, they say.
“Existing approaches are resource intensive,” they point out in yesterday’s PLoS Medicine so the team investigated a new approach using the UK Biobank and US National Lung Screening Trial datasets.
They created over 60 AI models to simplify the prediction of lung cancer risk in the subsequent five years, all including age, duration of smoking, and average number of cigarettes per day.
Next, four promising looking models were combined into one which could predict lung cancer risk to at least the same accuracy as current methods but requiring less information.
This could inform the UK’s planned national screening programme for lung cancer for people aged 55 to 74 years, the team believes.
Dr Callender said: “Screening for cancer and other diseases saves lives and we are increasingly able to personalise this process. But such personalised screening and disease prevention programmes present important logistical challenges at scale.
“Our study shows that artificial intelligence can be used to accurately predict lung cancer risk using just three pieces of information that would be easy to gather during routine GP appointments, online or via apps.
“This approach has the potential to greatly simplify population level screening for lung cancer and help to make it a reality.”
Callender, T. et al. Assessing eligibility for lung cancer screening using parsimonious ensemble machine learning models: A development and validation study. PLoS Medicine 3 October 2023 doi: 10.1371/journal.pmed.1004287
Leave a Reply