Find the PLOS Medicine paper here.
Targeted lung cancer screening for high-risk individuals could potentially reduce lung cancer-specific mortality by 20-24%. A highly significant potential given that lung cancer stands as the leading global cause of cancer-related deaths, claiming an estimated 1.8 million lives in 2020 alone.
To investigate the potential of AI and machine learning to improve early detection, our study harnessed data from the UK Biobank and the US National Lung Screening Trial to create novel models for simplifying the prediction of an individual’s lung cancer risk over the next five years.
The team explored over 250 machine learning pipelines using these datasets to determine the most effective approach for predicting lung cancer risk, ultimately relying on just three variables: age, smoking duration, and daily cigarette consumption.
They selected four model pipelines and combined them into an ensemble that achieved the same or improved accuracy compared to the best existing models, all while using only a third of the variables, streamlining data collection. A promising result that could illustrate the potential for this method for other diseases.
“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.”
Dr Thomas Callender
These findings will streamline the implementation of a national lung cancer screening program, making it faster, more cost-effective, and more accessible while effectively reducing lung cancer mortality.
“This research is a prime example of how machine learning tools such as
Prof Mihaela van der Schaar
AutoPrognosis, combined with innovative clinical researchers, can make a real impact in
healthcare at a population level. While AutoPrognosis has already been applied for risk prediction and prognosis in numerous diseases, this is the first time it has been used to determine the minimal information needed to screen patients. I think this is the future of preventive medicine and I’m optimistic that the same approach could be applied to screening for other diseases.”
The research received support from Wellcome, the National Science Foundation, the Medical Research Council, and Cancer Research UK.
Find out more about AutoPrognosis 2.0 here.
