Validation of AI Model to Predict Cancer Risk
In this validation study, we assessed if a deep learning model that uses chest radiographs and other electronic medical record data can identify high-risk individuals for lung cancer screening, complementary to the lung cancer screening computed tomography eligibility criteria.
We analyzed electronic medical record data of 14,737 patients in the Mass General Brigham health system, a deep learning model that identified persons at high risk for lung cancer, including those missed by Medicare lung cancer screening eligibility criteria.
These results suggest that a deep learning model can help identify high-risk individuals who may benefit from lung cancer screening with chest computed tomography.
AIM Investigators
