Blog

To achieve this, the research team used data from the UK Biobank, a long-running health project tracking around 500,000 volunteers. The database contains anonymized health records and biological samples that help scientists study disease development over time. Using plasma samples from nearly 48,000 participants, researchers trained a machine-learning model to analyze protein patterns alongside factors such as age, sex, smoking status, and lung cancer diagnosis history. The model identified 14 proteins strongly associated with future lung cancer risk.