Detection of breast cancer in digital breast tomosynthesis with vision transformers.
Journal
Scientific reports
ISSN: 2045-2322
Titre abrégé: Sci Rep
Pays: England
ID NLM: 101563288
Informations de publication
Date de publication:
27 Sep 2024
27 Sep 2024
Historique:
received:
14
01
2024
accepted:
10
09
2024
medline:
28
9
2024
pubmed:
28
9
2024
entrez:
27
9
2024
Statut:
epublish
Résumé
Digital Breast Tomosynthesis (DBT) has revolutionized more traditional breast imaging through its three-dimensional (3D) visualization capability that significantly enhances lesion discernibility, reduces tissue overlap, and improves diagnostic precision as compared to conventional two-dimensional (2D) mammography. In this study, we propose an advanced Computer-Aided Detection (CAD) system that harnesses the power of vision transformers to augment DBT's diagnostic efficiency. This scheme uses a neural network to glean attributes from the 2D slices of DBT followed by post-processing that considers features from neighboring slices to categorize the entire 3D scan. By leveraging a transfer learning technique, we trained and validated our CAD framework on a unique dataset consisting of 3,831 DBT scans and subsequently tested it on 685 scans. Of the architectures tested, the Swin Transformer outperformed the ResNet101 and vanilla Vision Transformer. It achieved an impressive AUC score of 0.934 ± 0.026 at a resolution of 384
Identifiants
pubmed: 39333178
doi: 10.1038/s41598-024-72707-2
pii: 10.1038/s41598-024-72707-2
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
22149Informations de copyright
© 2024. The Author(s).
Références
Bray, F. et al. Global cancer statistics 2022: Globocan estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J. Clin.74, 229–263. https://doi.org/10.3322/caac.21834 (2024).
Tan, X. J. et al. Breast cancer status, grading system, etiology, and challenges in Asia: An updated review. Oncologie25, 99–110. https://doi.org/10.1515/oncologie-2022-1011 (2023).
doi: 10.1515/oncologie-2022-1011
Bai, J., Posner, R., Wang, T., Yang, C. & Nabavi, S. Applying deep learning in digital breast tomosynthesis for automatic breast cancer detection: A review. Med. Image Anal.71, 102049. https://doi.org/10.1016/j.media.2021.102049 (2021).
doi: 10.1016/j.media.2021.102049
pubmed: 33901993
Oeffinger, K. C. et al. Breast cancer screening for women at average risk: 2015 guideline update from the American cancer society. JAMA314, 1599–1614. https://doi.org/10.1001/jama.2015.12783 (2015).
doi: 10.1001/jama.2015.12783
pubmed: 26501536
pmcid: 4831582
Lukasiewicz, S. et al. Breast cancer—epidemiology, risk factors, classification, prognostic markers, and current treatment strategies—an updated review. Cancers13, 4287. https://doi.org/10.3390/cancers13174287 (2021).
doi: 10.3390/cancers13174287
pubmed: 34503097
pmcid: 8428369
Saslow, D. et al. American cancer society guidelines for breast screening with MRI as an adjunct to mammography. CA Cancer J. Clin.57, 75–89. https://doi.org/10.3322/canjclin.57.2.75 (2007).
Sechopoulos, I. A review of breast tomosynthesis. Part I. The image acquisition process. Med. Phys.40, 014301. https://doi.org/10.1118/1.4770279 (2013).
Lehman, C. D. et al. National performance benchmarks for modern screening digital mammography: Update from the breast cancer surveillance consortium. Radiology283, 49–58. https://doi.org/10.1148/radiol.2016161174 (2017).
doi: 10.1148/radiol.2016161174
pubmed: 27918707
Niklason, L. T. et al. Digital tomosynthesis in breast imaging. Radiology205, 399–406. https://doi.org/10.1148/radiology.205.2.9356620 (1997).
doi: 10.1148/radiology.205.2.9356620
pubmed: 9356620
Vedantham, S., Karellas, A., Vijayaraghavan, G. R. & Kopans, D. B. Digital breast tomosynthesis: State of the art. Radiology277, 663–684. https://doi.org/10.1148/radiol.2015141303 (2015).
doi: 10.1148/radiol.2015141303
pubmed: 26599926
Zhang, X. et al. Classification of whole mammogram and tomosynthesis images using deep convolutional neural networks. IEEE Trans. NanoBioscience17, 237–242. https://doi.org/10.1109/TNB.2018.2845103 (2018).
Kleinknecht, J. H., Ciurea, A. I. & Ciortea, C. A. Pros and cons for breast cancer screening with tomosynthesis—a review of the literature. Med. Pharm. Rep.93, 335–341. https://doi.org/10.15386/mpr-1698 (2020).
Azavedo, E., Zackrisson, S., Mejàre, I. & Heibert Arnlind, M. Is single reading with computer-aided detection (CAD) as good as double reading in mammography screening? A systematic review. BMC Med. Imaging12, 22. https://doi.org/10.1186/1471-2342-12-22 (2012).
doi: 10.1186/1471-2342-12-22
pubmed: 22827803
pmcid: 3464719
Jiao, Z., Gao, X., Wang, Y. & Li, J. A deep feature based framework for breast masses classification. Neurocomputing197, 221–231. https://doi.org/10.1016/j.neucom.2016.02.060 (2016).
doi: 10.1016/j.neucom.2016.02.060
Jadoon, M. M., Zhang, Q., Haq, I. U., Butt, S. & Jadoon, A. Three-class mammogram classification based on descriptive CNN features. BioMed Res. Int.2017, e3640901. https://doi.org/10.1155/2017/3640901 (2017).
Henry, E. U., Emebob, O. & Omonhinmin, C. A. Vision transformers in medical imaging: A review. https://doi.org/10.48550/arXiv.2211.10043 (2022). arXiv:2211.10043 [cs].
Mendes, J., Matela, N. & Garcia, N. Avoiding tissue overlap in 2D images: Single-slice DBT classification using convolutional neural networks. Tomography9, 398–412 https://doi.org/10.3390/tomography9010032 (2023).
Samala, R. K. et al. Evolutionary pruning of transfer learned deep convolutional neural network for breast cancer diagnosis in digital breast tomosynthesis. Phys. Med. Biol.63, 095005. https://doi.org/10.1088/1361-6560/aabb5b (2018).
Li, X. et al. Digital breast tomosynthesis versus digital mammography: integration of image modalities enhances deep learning-based breast mass classification. Eur. Radiol.30, 778–788. https://doi.org/10.1007/s00330-019-06457-5 (2020).
doi: 10.1007/s00330-019-06457-5
pubmed: 31691121
Ayana, G. et al. Vision-transformer-based transfer learning for mammogram classification. Diagnostics13, 178. https://doi.org/10.3390/diagnostics13020178 (2023).
Raghu, M., Unterthiner, T., Kornblith, S., Zhang, C. & Dosovitskiy, A. Do vision transformers see like convolutional neural networks? CoRR (2021). arXiv:2108.08810 .
Cantone, M., Marrocco, C., Tortorella, F. & Bria, A. Convolutional networks and transformers for mammography classification: An experimental study. Sensors23, 1229. https://doi.org/10.3390/s23031229 (2023).
Lee, W. et al. Transformer-based deep neural network for breast cancer classification on digital breast Tomosynthesis images. Radiol. Artif. Intell. e220159. https://doi.org/10.1148/ryai.220159 (2023).
Liu, F. et al. Self-supervised mean teacher for semi-supervised chest X-ray classification. https://doi.org/10.48550/arXiv.2103.03629 (2021). arXiv:2103.03629 [cs].
Liu, Z. et al. Swin transformer: Hierarchical vision transformer using shifted windows. CoRR (2021). arXiv:2103.14030 [abs].
He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition (2015). arXiv:1512.03385 [cs].
Dosovitskiy, A. et al. an image is worth 16x16 words: Transformers for image recognition at scale (2021). arXiv:2010.11929 [cs].
Cserni, G., Chmielik, E., Cserni, B. & Tot, T. The new tnm-based staging of breast cancer. Virchows Arch.472, 697–703. https://doi.org/10.1007/s00428-018-2301-9 (2018).
doi: 10.1007/s00428-018-2301-9
pubmed: 29380126
Koh, J. & Kim, M. J. Introduction of a new staging system of breast cancer for radiologists: An emphasis on the prognostic stage. Korean J. Radiol.20, 69–82. https://doi.org/10.3348/kjr.2018.0231 (2019).
doi: 10.3348/kjr.2018.0231
pubmed: 30627023
Checka, C. M., Chun, J. E., Schnabel, F. R., Lee, J. & Toth, H. The relationship of mammographic density and age: Implications for breast cancer screening. Am. J. Roentgenol.198, W292–W295. https://doi.org/10.2214/AJR.10.6049 (2012).
Balleyguier, C. et al. BIRADS™ classification in mammography. Eur. J. Radiol.61, 192–194. https://doi.org/10.1016/j.ejrad.2006.08.033 (2007).
doi: 10.1016/j.ejrad.2006.08.033
pubmed: 17164080
Selvaraju, R. R. et al. Grad-cam: Why did you say that? visual explanations from deep networks via gradient-based localization. CoRR (2016). arXiv:abs/1610.02391 .
Spak, D., Plaxco, J., Santiago, L., Dryden, M. & Dogan, B. Bi-rads[Formula: see text] fifth edition: A summary of changes. Diagn. Interv. Imaging98, 179–190. https://doi.org/10.1016/j.diii.2017.01.001 (2017).
doi: 10.1016/j.diii.2017.01.001
pubmed: 28131457
Franchini, M. et al. The p.i.n.k. study approach for supporting personalized risk assessment and early diagnosis of breast cancer. Int. J. Environ. Res. Public Health18. https://doi.org/10.3390/ijerph18052456 (2021).
Destounis, S. V., Morgan, R. & Arieno, A. Screening for dense breasts: Digital breast tomosynthesis. Am. J. Roentgenol.204, 261–264 (2015). https://doi.org/10.2214/AJR.14.13554 .
Lång, K. et al. Artificial intelligence-supported screen reading versus standard double reading in the Mammography Screening with Artificial Intelligence trial (MASAI): a clinical safety analysis of a randomised, controlled, non-inferiority, single-blinded, screening accuracy study. Lancet Oncol.24, 936–944. https://doi.org/10.1016/S1470-2045(23)00298-X (2023).
doi: 10.1016/S1470-2045(23)00298-X
pubmed: 37541274
Ridnik, T., Ben-Baruch, E., Noy, A. & Zelnik-Manor, L. ImageNet-21K Pretraining for the Masses. https://doi.org/10.48550/arXiv.2104.10972 (2021). arXiv:2104.10972 [cs].
Weiss, K., Khoshgoftaar, T. M. & Wang, D. A survey of transfer learning. J. Big Data3, 9. https://doi.org/10.1186/s40537-016-0043-6 (2016).
doi: 10.1186/s40537-016-0043-6
Loshchilov, I. & Hutter, F. Decoupled weight decay regularization (2019). arXiv:1711.05101 [cs, math].
Tarvainen, A. & Valpola, H. Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results (2018). arXiv:1703.01780 [cs, stat].
DeLong, E. R., DeLong, D. M. & Clarke-Pearson, D. L. Comparing the areas under two or more correlated receiver operating characteristic curves: A nonparametric approach. Biometrics44, 837–845 (1988).
doi: 10.2307/2531595
pubmed: 3203132