External Test of a Deep Learning Algorithm for Pulmonary Nodule Malignancy Risk Stratification Using European Screening Data

Внешняя проверка алгоритма глубокого обучения для стратификации риска злокачественности легочных узлов с использованием европейских данных скрининга
Roel Vermeulen, Cornelia Schaefer‐Prokop, Mario Silva, Ralph Stadhouders, Geertruida H. de Bock, Pim A. de Jong, Ernst T. Scholten, Kristiaan Nackaerts, Marjolein A. Heuvelmans, Mathias Prokop, Firdaus A. A. Mohamed Hoesein, Joachim G.J.V. Aerts, Rozemarijn Vliegenthart, George S. Downward, Walter De Wever, Ugo Pastorino, Jeroen van Rooij, Hester A. Gietema, Colin Jacobs, Robin Cornelissen, Zaigham Saghir, Noa J.C. Antonissen, Kiran Vaidhya Venkadesh, Renate Dinnessen, Grigory Sidorenkov, Harry J. M. Groen, Lianne Trap, Danrong Zhong, Harry J.M. Groen, Nils van der Velden, Pim A. de Jong, Stijn Bunk
2025-09-01

European screening datasetsPanCan modeldeep learning algorithmpulmonary nodule malignancyrisk stratification
< .01), respectively. Conclusion The DL algorithm outperformed the PanCan model across multiple European screening datasets, demonstrating superior malignancy prediction while substantially reducing false-positive classifications for indeterminate nodules. © RSNA, 2025
1
Compared with PanCan, the deep learning approach substantially reduced false-positive classifications for indeterminate pulmonary nodules.
2
The abstract reports statistically significant comparative results, although the provided excerpt omits the specific performance metrics.
3
The algorithm outperformed the PanCan model in predicting pulmonary nodule malignancy across European screening populations.
4
The deep learning algorithm was externally tested using multiple European lung cancer screening datasets.

Deep learning algorithm for pulmonary nodule malignancy risk stratification applied to European lung cancer screening data

Malignancy prediction performance and false-positive classification reduction compared with the PanCan model across European screening datasets

Publication Details
Publication Date
2025-09-01
Journal
Publisher
ISSN
Cited by
13
Access Type
Author Information
Authors
Roel Vermeulen
Cornelia Schaefer‐Prokop
Mario Silva
Ralph Stadhouders
Geertruida H. de Bock
Pim A. de Jong
Ernst T. Scholten
Kristiaan Nackaerts
Marjolein A. Heuvelmans
Mathias Prokop
Firdaus A. A. Mohamed Hoesein
Joachim G.J.V. Aerts
Rozemarijn Vliegenthart
George S. Downward
Walter De Wever
Ugo Pastorino
Jeroen van Rooij
Hester A. Gietema
Colin Jacobs
Robin Cornelissen
Zaigham Saghir
Noa J.C. Antonissen
Kiran Vaidhya Venkadesh
Renate Dinnessen
Grigory Sidorenkov
Harry J. M. Groen
Lianne Trap
Danrong Zhong
Harry J.M. Groen
Nils van der Velden
Pim A. de Jong
Stijn Bunk
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat
Make a presentation
100%