Lung cancer is the common cause of cancer deaths worldwide and the leading reason is the absence of early detection. In most cases, lung cancer is often detected at a later stage and then the survival rates go down. According to the findings published in PLOS medicine, machine learning models were able to identify the simplest way to screen for lung cancer. This model has been developed by researchers from UCL and the University of Cambridge. They have developed models to predict the risk of a person getting lung cancer in the next five years.
Machine learning is the application of artificial intelligence to detect lung cancer disease more quickly. The team of researchers used the datasets to experiment with over 60 different machine learning pipelines to see the most effective at predicting lung cancer risk. They selected four model pipelines and combined them into an ensemble to predict lung cancer risk with same or improved accuracy. They were also able to achieve this accuracy using only a third of variables, simplifying the process of gathering the required data.
The models were externally validated in the US Prostate, Lung, Ovarian and Colorectal cancer screening trial. They benchmarked against models that are in use or have performed strongly in previous analyses. The models were fed using three variables: a person’s age, years they smoked and the average number of cigarettes smoked per day.
The authors of the study hope that the findings will be used to make any national lung cancer screening programme in a quicker and cheaper way to implement. This also helps to achieve the primary aim of reducing lung cancer mortality. This work was supported by Wellcome, the National Science Foundation, the Medical Research Council and Cancer Research UK.
The information shared in this blog is for educational purposes only.