Saved in:
| Main Authors: | , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2511.05967 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915606318546944 |
|---|---|
| author | Nguyen, Tri-Thien Kapsner, Lorenz A. Hepp, Tobias Heidarikahkesh, Shirin Schreiter, Hannes Brock, Luise Skwierawska, Dominika Hadler, Dominique Hossbach, Julian Wenkel, Evelyn Ohlmeyer, Sabine Laun, Frederik B. Liebert, Andrzej Maier, Andreas Uder, Michael Bickelhaupt, Sebastian |
| author_facet | Nguyen, Tri-Thien Kapsner, Lorenz A. Hepp, Tobias Heidarikahkesh, Shirin Schreiter, Hannes Brock, Luise Skwierawska, Dominika Hadler, Dominique Hossbach, Julian Wenkel, Evelyn Ohlmeyer, Sabine Laun, Frederik B. Liebert, Andrzej Maier, Andreas Uder, Michael Bickelhaupt, Sebastian |
| contents | Background: Magnetic resonance imaging (MRI) has high sensitivity for breast cancer detection, but interpretation is time-consuming. Artificial intelligence may aid in pre-screening. Purpose: To evaluate the DINOv2-based Medical Slice Transformer (MST) for ruling out significant findings (Breast Imaging Reporting and Data System [BI-RADS] >=4) in contrast-enhanced and non-contrast-enhanced abbreviated breast MRI. Materials and Methods: This institutional review board approved retrospective study included 1,847 single-breast MRI examinations (377 BI-RADS >=4) from an in-house dataset and 924 from an external validation dataset (Duke). Four abbreviated protocols were tested: T1-weighted early subtraction (T1sub), diffusion-weighted imaging with b=1500 s/mm2 (DWI1500), DWI1500+T2-weighted (T2w), and T1sub+T2w. Performance was assessed at 90%, 95%, and 97.5% sensitivity using five-fold cross-validation and area under the receiver operating characteristic curve (AUC) analysis. AUC differences were compared with the DeLong test. False negatives were characterized, and attention maps of true positives were rated in the external dataset. Results: A total of 1,448 female patients (mean age, 49 +/- 12 years) were included. T1sub+T2w achieved an AUC of 0.77 +/- 0.04; DWI1500+T2w, 0.74 +/- 0.04 (p=0.15). At 97.5% sensitivity, T1sub+T2w had the highest specificity (19% +/- 7%), followed by DWI1500+T2w (17% +/- 11%). Missed lesions had a mean diameter <10 mm at 95% and 97.5% thresholds for both T1sub and DWI1500, predominantly non-mass enhancements. External validation yielded an AUC of 0.77, with 88% of attention maps rated good or moderate. Conclusion: At 97.5% sensitivity, the MST framework correctly triaged cases without BI-RADS >=4, achieving 19% specificity for contrast-enhanced and 17% for non-contrast-enhanced MRI. Further research is warranted before clinical implementation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_05967 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Adapted Foundation Models for Breast MRI Triaging in Contrast-Enhanced and Non-Contrast Enhanced Protocols Nguyen, Tri-Thien Kapsner, Lorenz A. Hepp, Tobias Heidarikahkesh, Shirin Schreiter, Hannes Brock, Luise Skwierawska, Dominika Hadler, Dominique Hossbach, Julian Wenkel, Evelyn Ohlmeyer, Sabine Laun, Frederik B. Liebert, Andrzej Maier, Andreas Uder, Michael Bickelhaupt, Sebastian Computer Vision and Pattern Recognition Artificial Intelligence Background: Magnetic resonance imaging (MRI) has high sensitivity for breast cancer detection, but interpretation is time-consuming. Artificial intelligence may aid in pre-screening. Purpose: To evaluate the DINOv2-based Medical Slice Transformer (MST) for ruling out significant findings (Breast Imaging Reporting and Data System [BI-RADS] >=4) in contrast-enhanced and non-contrast-enhanced abbreviated breast MRI. Materials and Methods: This institutional review board approved retrospective study included 1,847 single-breast MRI examinations (377 BI-RADS >=4) from an in-house dataset and 924 from an external validation dataset (Duke). Four abbreviated protocols were tested: T1-weighted early subtraction (T1sub), diffusion-weighted imaging with b=1500 s/mm2 (DWI1500), DWI1500+T2-weighted (T2w), and T1sub+T2w. Performance was assessed at 90%, 95%, and 97.5% sensitivity using five-fold cross-validation and area under the receiver operating characteristic curve (AUC) analysis. AUC differences were compared with the DeLong test. False negatives were characterized, and attention maps of true positives were rated in the external dataset. Results: A total of 1,448 female patients (mean age, 49 +/- 12 years) were included. T1sub+T2w achieved an AUC of 0.77 +/- 0.04; DWI1500+T2w, 0.74 +/- 0.04 (p=0.15). At 97.5% sensitivity, T1sub+T2w had the highest specificity (19% +/- 7%), followed by DWI1500+T2w (17% +/- 11%). Missed lesions had a mean diameter <10 mm at 95% and 97.5% thresholds for both T1sub and DWI1500, predominantly non-mass enhancements. External validation yielded an AUC of 0.77, with 88% of attention maps rated good or moderate. Conclusion: At 97.5% sensitivity, the MST framework correctly triaged cases without BI-RADS >=4, achieving 19% specificity for contrast-enhanced and 17% for non-contrast-enhanced MRI. Further research is warranted before clinical implementation. |
| title | Adapted Foundation Models for Breast MRI Triaging in Contrast-Enhanced and Non-Contrast Enhanced Protocols |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2511.05967 |