Researchers have been exploring ways to improve outcomes along the screening continuum, such as utilizing clinical prediction modeling and harnessing the power of the electronic medical record and radiomics. The recently developed Sybil artificial intelligence and deep learning model is a vital first step toward a precise approach to lung cancer screening.
It can predict an individual’s future risk of developing lung cancer after a single baseline computed tomography chest scan. This model may improve adherence to follow-up, reduce unnecessary workup and invasive tests, and extend the period between screenings or stop screening altogether in those at low risk of cancer. Additionally, the model could identify at-risk populations that do not meet US Preventative Services Task Force requirements for inclusion in a screening program with the help of future prospective evaluations. This data could aid shared decision-making visits for patients with a high comorbidity load and a short life expectancy.