Advancing deep learning-based segmentation for multiple lung cancer lesions in real-world multicenter CT scans
AI-driven segmentation comprehensively captures lesion burden, enhancing lung cancer assessment and disease monitoring KEY POINTS: Automatic multi-instance lung cancer lesion segmentation is underexplored yet crucial for disease assessment. Developed a deep learning-based segmentation pipeline trained on multi-center real-world data, which reached 85% sensitivity at external validation. Thoracic bounding box and false positive reduction techniques improved the pipeline's segmentation performance.
Authors
- Xavier Rafael-Palou (ORCID: https://orcid.org/0000-0002-4489-4806)
- Ángel Alberich‐Bayarri (ORCID: https://orcid.org/0000-0002-5932-2392)
- Carlos Francisco Muñoz-Núñez
- Ana Jiménez-Pastor (ORCID: https://orcid.org/0000-0002-0978-9429)
- Mario Laudazi (ORCID: https://orcid.org/0000-0001-6094-6330)
- Luis Martí‐Bonmatí (ORCID: https://orcid.org/0000-0002-8234-010X)
Institutions
- University of Rome Tor Vergata (IT)
- Xavier University (US)
- Hospital Universitari i Politècnic La Fe (ES)
- Valencian Infertility Institute (ES)
- Leitat Technological Center (ES)
- Instituto de Investigación Sanitaria La Fe (ES)
Publication Details
- Journal
- European Radiology Experimental
- Published
- 2025-08-18
- DOI
- https://doi.org/10.1186/s41747-025-00617-7
- Citations
- 4
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 5.44