Multiclass Segmentation of Breast Tissue and Suspicious Findings: A Simulation-Based Study for the Development of Self-Steering Tomosynthesis
Tomography · 2023
DOI: 10.3390/tomography9030092 ↗Kurz gefasst
Fachartikel in der Zeitschrift Tomography, 2023. Verfasst von Bruno Barufaldi und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, University of Pennsylvania, Philadelphia, PA 19104, USA. Hauptthemen laut MEDLINE: Mammography, Breast Neoplasms. Die Arbeit nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NCI NIH HHS, NIH HHS, Terri Brodeur Breast Cancer Foundation.
Abstract (Originalsprache)
In breast tomosynthesis, multiple low-dose projections are acquired in a single scanning direction over a limited angular range to produce cross-sectional planes through the breast for three-dimensional imaging interpretation. We built a next-generation tomosynthesis system capable of multidirectional source motion with the intent to customize scanning motions around “suspicious findings”. Customized acquisitions can improve the image quality in areas that require increased scrutiny, such as breast cancers, architectural distortions, and dense clusters. In this paper, virtual clinical trial techniques were used to analyze whether a finding or area at high risk of masking cancers can be detected in a single low-dose projection and thus be used for motion planning. This represents a step towards customizing the subsequent low-dose projection acquisitions autonomously, guided by the first low-dose projection; we call this technique “self-steering tomosynthesis.” A U-Net was used to classify the low-dose projections into “risk classes” in simulated breasts with soft-tissue lesions; class probabilities were modified using post hoc Dirichlet calibration (DC). DC improved the multiclass segmentation (Dice = 0.43 vs. 0.28 before DC) and significantly reduced false positives (FPs) from the class of the highest risk of masking (sensitivity = 81.3% at 2 FPs per image vs. 76.0%). This simulation-based study demonstrated the feasibility of identifying suspicious areas using a single low-dose projection for self-steering tomosynthesis.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Radiology, University of Pennsylvania, Philadelphia, PA 19104, USA.
- Center of Informatics, Federal University of Paraiba, Joao Pessoa 58051-900, PB, Brazil.
- Department of Engineering Mathematics, University of Bristol, Bristol BS8 1QU, UK.
ORCID-Profile: Bruno Barufaldi, Giulia Carvalhal, Telmo M Silva Filho, Thais G do Rego, Yuri Malheiros, Raymond J Acciavatti
Förderung
- NCI NIH HHS
- NIH HHS
- Terri Brodeur Breast Cancer Foundation
- American Association of Physicists in Medicine, 2020 Research Seed Funding
- Breast Cancer Alliance, Inc.
- 2022 Young Investigator grant
- Department of Defense Breast Cancer Research Program
- Burroughs Wellcome Fund
Zitieren
APA
Barufaldi, B., da Nobrega, Y. N. G., Carvalhal, G., Teixeira, J. P. V., Silva Filho, T. M., do Rego, T. G., Malheiros, Y., Acciavatti, R. J., & Maidment, A. D. A. (2023). Multiclass Segmentation of Breast Tissue and Suspicious Findings: A Simulation-Based Study for the Development of Self-Steering Tomosynthesis. Tomography. https://doi.org/10.3390/tomography9030092
Vancouver
Barufaldi B, da Nobrega YNG, Carvalhal G, Teixeira JPV, Silva Filho TM, do Rego TG, et al. Multiclass Segmentation of Breast Tissue and Suspicious Findings: A Simulation-Based Study for the Development of Self-Steering Tomosynthesis. Tomography. 2023. doi:10.3390/tomography9030092
BibTeX
@article{Barufaldi2023,
author = {Barufaldi, Bruno and da Nobrega, Yann N G and Carvalhal, Giulia and Teixeira, Joao P V and Silva Filho, Telmo M and do Rego, Thais G and Malheiros, Yuri and Acciavatti, Raymond J and Maidment, Andrew D A},
title = {Multiclass Segmentation of Breast Tissue and Suspicious Findings: A Simulation-Based Study for the Development of Self-Steering Tomosynthesis},
journal = {Tomography},
year = {2023},
doi = {10.3390/tomography9030092}
}
EUVIMED