AI-driven detection and characterization of incidental lung nodules

Background

Lung cancer is often diagnosed in a late stage, and as a result, the 5-year survival rate for lung cancer is only 18%. If lung cancer is detected in an early stage, the prognosis is much better. Therefore, improving early detection is the most promising strategy to reduce lung cancer mortality.

Early stage lung cancer is typically diagnosed after the detection of an incidental lung nodule on CT imaging of the thorax that was ordered for other medical reasons. For this reason, accurate detection and characterization of incidentally detected lung nodules is of great importance.

Aim

In this project, we aim to develop artificial intelligence (AI) algorithms for fast and highly reliable detection and assessment of nodules in chest CT, and to validate the novel AI-assisted nodule workflow at two Dutch hospitals.

Funding

This project has received and continues to receive funding from Radboudumc and Jeroen Bosch Ziekenhuis in 's-Hertogenbosch.

People

Colin Jacobs

Colin Jacobs

Assistant Professor

Ward Hendrix

Ward Hendrix

PhD Candidate

Ernst Scholten

Ernst Scholten

Radiologist

Matthieu Rutten

Matthieu Rutten

Associate Professor

Mathias Prokop

Mathias Prokop

Professor

Radboudumc

Publications

  • K. Chung, O. Mets, P. Gerke, C. Jacobs, A. den Harder, E. Scholten, M. Prokop, P. de Jong, B. van Ginneken and C. Schaefer-Prokop, "Brock malignancy risk calculator for pulmonary nodules: validation outside a lung cancer screening population", Thorax, 2018;73:857-863. Abstract/PDF DOI PMID Cited by ~18