Tampere University of Technology

TUTCRIS Research Portal

Pipeline for effective denoising of digital mammography and digital breast tomosynthesis

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Standard

Pipeline for effective denoising of digital mammography and digital breast tomosynthesis. / Borges, Lucas R.; Bakic, Predrag R.; Foi, Alessandro; Maidment, Andrew D.A.; Vieira, Marcelo A.C.

Medical Imaging 2017: Physics of Medical Imaging. SPIE, 2017. 1013206 (Progress in biomedical optics and imaging).

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Harvard

Borges, LR, Bakic, PR, Foi, A, Maidment, ADA & Vieira, MAC 2017, Pipeline for effective denoising of digital mammography and digital breast tomosynthesis. in Medical Imaging 2017: Physics of Medical Imaging., 1013206, Progress in biomedical optics and imaging, SPIE, Medical Imaging, 1/01/00. https://doi.org/10.1117/12.2255058

APA

Borges, L. R., Bakic, P. R., Foi, A., Maidment, A. D. A., & Vieira, M. A. C. (2017). Pipeline for effective denoising of digital mammography and digital breast tomosynthesis. In Medical Imaging 2017: Physics of Medical Imaging [1013206] (Progress in biomedical optics and imaging). SPIE. https://doi.org/10.1117/12.2255058

Vancouver

Borges LR, Bakic PR, Foi A, Maidment ADA, Vieira MAC. Pipeline for effective denoising of digital mammography and digital breast tomosynthesis. In Medical Imaging 2017: Physics of Medical Imaging. SPIE. 2017. 1013206. (Progress in biomedical optics and imaging). https://doi.org/10.1117/12.2255058

Author

Borges, Lucas R. ; Bakic, Predrag R. ; Foi, Alessandro ; Maidment, Andrew D.A. ; Vieira, Marcelo A.C. / Pipeline for effective denoising of digital mammography and digital breast tomosynthesis. Medical Imaging 2017: Physics of Medical Imaging. SPIE, 2017. (Progress in biomedical optics and imaging).

Bibtex - Download

@inproceedings{7c1bba5c1d89446aba0a72b952838edb,
title = "Pipeline for effective denoising of digital mammography and digital breast tomosynthesis",
abstract = "Denoising can be used as a tool to enhance image quality and enforce low radiation doses in X-ray medical imaging. The effectiveness of denoising techniques relies on the validity of the underlying noise model. In full-field digital mammography (FFDM) and digital breast tomosynthesis (DBT), calibration steps like the detector offset and flat-fielding can affect some assumptions made by most denoising techniques. Furthermore, quantum noise found in X-ray images is signal-dependent and can only be treated by specific filters. In this work we propose a pipeline for FFDM and DBT image denoising that considers the calibration steps and simplifies the modeling of the noise statistics through variance-stabilizing transformations (VST). The performance of a state-of-the-art denoising method was tested with and without the proposed pipeline. To evaluate the method, objective metrics such as the normalized root mean square error (N-RMSE), noise power spectrum, modulation transfer function (MTF) and the frequency signal-to-noise ratio (SNR) were analyzed. Preliminary tests show that the pipeline improves denoising. When the pipeline is not used, bright pixels of the denoised image are under-filtered and dark pixels are over-smoothed due to the assumption of a signal-independent Gaussian model. The pipeline improved denoising up to 20{\%} in terms of spatial N-RMSE and up to 15{\%} in terms of frequency SNR. Besides improving the denoising, the pipeline does not increase signal smoothing significantly, as shown by the MTF. Thus, the proposed pipeline can be used with state-of-the-art denoising techniques to improve the quality of DBT and FFDM images.",
keywords = "Denoising, Digital breast tomosynthesis, Full field digital mammography, Variance stabilization",
author = "Borges, {Lucas R.} and Bakic, {Predrag R.} and Alessandro Foi and Maidment, {Andrew D.A.} and Vieira, {Marcelo A.C.}",
note = "jufoid=65546",
year = "2017",
doi = "10.1117/12.2255058",
language = "English",
series = "Progress in biomedical optics and imaging",
publisher = "SPIE",
booktitle = "Medical Imaging 2017",
address = "United States",

}

RIS (suitable for import to EndNote) - Download

TY - GEN

T1 - Pipeline for effective denoising of digital mammography and digital breast tomosynthesis

AU - Borges, Lucas R.

AU - Bakic, Predrag R.

AU - Foi, Alessandro

AU - Maidment, Andrew D.A.

AU - Vieira, Marcelo A.C.

N1 - jufoid=65546

PY - 2017

Y1 - 2017

N2 - Denoising can be used as a tool to enhance image quality and enforce low radiation doses in X-ray medical imaging. The effectiveness of denoising techniques relies on the validity of the underlying noise model. In full-field digital mammography (FFDM) and digital breast tomosynthesis (DBT), calibration steps like the detector offset and flat-fielding can affect some assumptions made by most denoising techniques. Furthermore, quantum noise found in X-ray images is signal-dependent and can only be treated by specific filters. In this work we propose a pipeline for FFDM and DBT image denoising that considers the calibration steps and simplifies the modeling of the noise statistics through variance-stabilizing transformations (VST). The performance of a state-of-the-art denoising method was tested with and without the proposed pipeline. To evaluate the method, objective metrics such as the normalized root mean square error (N-RMSE), noise power spectrum, modulation transfer function (MTF) and the frequency signal-to-noise ratio (SNR) were analyzed. Preliminary tests show that the pipeline improves denoising. When the pipeline is not used, bright pixels of the denoised image are under-filtered and dark pixels are over-smoothed due to the assumption of a signal-independent Gaussian model. The pipeline improved denoising up to 20% in terms of spatial N-RMSE and up to 15% in terms of frequency SNR. Besides improving the denoising, the pipeline does not increase signal smoothing significantly, as shown by the MTF. Thus, the proposed pipeline can be used with state-of-the-art denoising techniques to improve the quality of DBT and FFDM images.

AB - Denoising can be used as a tool to enhance image quality and enforce low radiation doses in X-ray medical imaging. The effectiveness of denoising techniques relies on the validity of the underlying noise model. In full-field digital mammography (FFDM) and digital breast tomosynthesis (DBT), calibration steps like the detector offset and flat-fielding can affect some assumptions made by most denoising techniques. Furthermore, quantum noise found in X-ray images is signal-dependent and can only be treated by specific filters. In this work we propose a pipeline for FFDM and DBT image denoising that considers the calibration steps and simplifies the modeling of the noise statistics through variance-stabilizing transformations (VST). The performance of a state-of-the-art denoising method was tested with and without the proposed pipeline. To evaluate the method, objective metrics such as the normalized root mean square error (N-RMSE), noise power spectrum, modulation transfer function (MTF) and the frequency signal-to-noise ratio (SNR) were analyzed. Preliminary tests show that the pipeline improves denoising. When the pipeline is not used, bright pixels of the denoised image are under-filtered and dark pixels are over-smoothed due to the assumption of a signal-independent Gaussian model. The pipeline improved denoising up to 20% in terms of spatial N-RMSE and up to 15% in terms of frequency SNR. Besides improving the denoising, the pipeline does not increase signal smoothing significantly, as shown by the MTF. Thus, the proposed pipeline can be used with state-of-the-art denoising techniques to improve the quality of DBT and FFDM images.

KW - Denoising

KW - Digital breast tomosynthesis

KW - Full field digital mammography

KW - Variance stabilization

U2 - 10.1117/12.2255058

DO - 10.1117/12.2255058

M3 - Conference contribution

T3 - Progress in biomedical optics and imaging

BT - Medical Imaging 2017

PB - SPIE

ER -