A deep learning algorithm accurately estimated pulmonary nodule malignancy risk using baseline data from three European trials, retrospectively reducing false-positive classifications for indeterminate nodules while preserving sensitivity for cancer detection.
External Test of a Deep Learning Algorithm for Pulmonary Nodule Malignancy Risk Stratification Using European Screening Data
N. Antonissen, K. Venkadesh, R. Dinnessen, E. Scholten, Z. Saghir, M. Silva, U. Pastorino, G. Sidorenkov, M. Heuvelmans, G. de Bock, F. Mohamed Hoesein, P. de Jong, H. Groen, R. Vliegenthart, H. Gietema, M. Prokop, C. Schaefer-Prokop, C. Jacobs, F. the consortium, J. Aerts, R. Cornelissen, R. Stadhouders, J. van Rooij, L. Trap, K. Nackaerts, W. de Wever, D. Zhong, N. van der Velden, S. Bunk, G. Downward and R. Vermeulen
Radiology 2025;316.