Automated detection and classification of osteolytic lesions in panoramic radiographs using CNNs and vision transformers
Promising deep learning models were developed for the detection of osteolytic lesions in PRs, particularly those with vision transformer components (Mask R-CNN with Swin-Tiny and Mask DINO). These results underline the potential of vision transformers for enhancing the automated detection of osteolytic lesions, offering a significant improvement over traditional deep learning models.
Authors
- Daniel G. E. Thiem (ORCID: https://orcid.org/0000-0002-4081-1487)
- Amir K. Bigdeli (ORCID: https://orcid.org/0000-0001-8852-5132)
- Stefaan Bergé
- Constantin von See (ORCID: https://orcid.org/0000-0001-6272-7326)
- Robert Gaudin (ORCID: https://orcid.org/0000-0001-9061-7220)
- Carsten Rendenbach (ORCID: https://orcid.org/0000-0002-9565-7640)
- Tong Xi (ORCID: https://orcid.org/0000-0002-9142-9253)
- Niels van Nistelrooij
- Shankeeth Vinayahalingam (ORCID: https://orcid.org/0000-0002-2679-3841)
- Max Heiland (ORCID: https://orcid.org/0000-0002-4987-2913)
- Iman Ghanad
- Ira Maistreli
Institutions
- Danube Private University (AT)
- Radboud University Nijmegen (NL)
- Johannes Gutenberg University Mainz (DE)
- Heidelberg University (DE)
- Ludwigshafen University of Business and Society (DE)
- Radboud University Medical Center (NL)
- Humboldt-Universität zu Berlin (DE)
- University Medical Center of the Johannes Gutenberg University Mainz (DE)
- Klinikum Ludwigshafen (DE)
Publication Details
- Journal
- BMC Oral Health
- Published
- 2025-06-21
- DOI
- https://doi.org/10.1186/s12903-025-06209-6
- Citations
- 5
- Primary Topic
- Dental Radiography and Imaging
- Type
- article
- Field-Weighted Citation Impact
- 8.95
Funders
- Charité – Universitätsmedizin Berlin
- Berlin Institute of Health