Radiation dose reduction in digital breast tomosynthesis (DBT) by means of deep-learning-based supervised image processing

Liu, Junchi; Zarshenas, Amin; Qadir, Ammar; Zheng, Wei; Yang, Limin; Fajardo, Laurie L.; Suzuki, Kenji

doi:10.1117/12.2293125

Cited by 28 publications

(30 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Under this terminology, we can construct a variety of early and recent DLIP models including fully convolutional networks. We applied NNC for radiation dose reduction in CT and breast imaging …”

Section: Methodsmentioning

confidence: 99%

“…We applied NNC for radiation dose reduction in CT and breast imaging. 32,34 Given a set of input and desired output (or "teaching") images, one can extract regions R and the corresponding output pixels, from the input and desired output images, respectively. Having this set, the goal of the machine learning is to find the best parameter vector h of an NNC model so that the predictions are closest to the desired values.…”

Section: B Neural Network Convolutionmentioning

confidence: 99%

See 1 more Smart Citation

Separation of bones from soft tissue in chest radiographs: Anatomy‐specific orientation‐frequency‐specific deep neural network convolution

et al. 2019

Self Cite

View full text Add to dashboard Cite

Purpose Lung nodules that are missed by radiologists as well as by computer‐aided detection (CAD) systems mostly overlap with ribs and clavicles. Removing the bony structures would result in better visualization of undetectable lesions. Our purpose in this study was to develop a virtual dual‐energy imaging system to separate ribs and clavicles from soft tissue in chest radiographs. Methods We developed a mixture of anatomy‐specific, orientation‐frequency‐specific (ASOFS) deep neural network convolution (NNC) experts. Anatomy‐specific (AS) NNC was designed to separate the bony structures from soft tissue in different lung segments. While an AS design was proposed previously under our massive‐training artificial neural networks (MTANN) framework, in this work, we newly mathematically defined an AS experts model as well as its learning and inference strategies in a probabilistic deep‐learning framework. In addition, in combination with our AS experts design, we newly proposed the orientation‐frequency‐specific (OFS) NNC models to decompose bone and soft‐tissue structures into specific orientation‐frequency components of different scales using a multi‐resolution decomposition technique. We trained multiple NNC models, each of which is an expert for a specific orientation‐frequency component in a particular anatomic segment. Perfect reconstruction discrete wavelet transform was used for OFS decomposition/reconstruction, while we introduced a soft‐gating layer to merge the predictions of AS NNC experts. To train our model, we used the bone images obtained from a dual‐energy system as the target (or teaching) images while the standard chest radiographs were used as the input to our model. The training, validation, and test were performed in a nested two‐fold cross‐validation manner. Results We used a database of 118 chest radiographs with pulmonary nodules to evaluate our NNC scheme. In order to evaluate our scheme, we performed quantitative and qualitative evaluation of the predicted bone and soft‐tissue images from our model as well as the ones of a state‐of‐the‐art technique where the “gold‐standard” dual‐energy bone and soft‐tissue images were used as the reference images. Both quantitative and qualitative evaluations demonstrated that our ASOFS NNC was superior to the state‐of‐the‐art bone‐suppression technique. Particularly, our scheme was better able to maintain the conspicuity of nodules and lung vessels, comparing to the reference technique, while it separated ribs and clavicles from soft tissue. Comparing to a state‐of‐the‐art bone suppression technique, our bone images had substantially higher (t−test; P < 0.01) similarity, in terms of structural similarity index (SSIM) and peak signal‐to‐noise ratio (PSNR), to the “gold−standard” dual−energy bone images. Conclusions Our deep ASOFS NNC scheme can decompose chest radiographs into their bone and soft‐tissue images accurately, offering the improved conspicuity of lung nodules and vessels, and therefore would be useful for radiologists as well as CAD systems in d...

show abstract

Section: Methodsmentioning

confidence: 99%

Section: B Neural Network Convolutionmentioning

confidence: 99%

Separation of bones from soft tissue in chest radiographs: Anatomy‐specific orientation‐frequency‐specific deep neural network convolution

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…[11][12][13][14] In medical imaging, DL methods are increasingly studied for noise reduction in low-dose imaging data such as computed tomography (CT), PET and SPECT. [15][16][17][18][19] These methods were found to achieve superior performance in terms of image resolution and noise level [15][16][17][18][19] compared to traditional filtering methods such as Gaussian filter, blockmatching 3D (BM3D), 20 and k-means singular value decomposition. 21 In this study, we investigate a DL approach for denoising in cardiac SPECT-MPI images.…”

Section: Introductionmentioning

confidence: 99%

Deep learning with noise‐to‐noise training for denoising in SPECT myocardial perfusion imaging

et al. 2020

Self Cite

View full text Add to dashboard Cite

Purpose Post‐reconstruction filtering is often applied for noise suppression due to limited data counts in myocardial perfusion imaging (MPI) with single‐photon emission computed tomography (SPECT). We study a deep learning (DL) approach for denoising in conventional SPECT‐MPI acquisitions, and investigate whether it can be more effective for improving the detectability of perfusion defects compared to traditional postfiltering. Methods Owing to the lack of ground truth in clinical studies, we adopt a noise‐to‐noise (N2N) training approach for denoising in SPECT‐MPI images. We consider a coupled U‐Net (CU‐Net) structure which is designed to improve learning efficiency through feature map reuse. For network training we employ a bootstrap procedure to generate multiple noise realizations from list‐mode clinical acquisitions. In the experiments we demonstrated the proposed approach on a set of 895 clinical studies, where the iterative OSEM algorithm with three‐dimensional (3D) Gaussian postfiltering was used to reconstruct the images. We investigated the detection performance of perfusion defects in the reconstructed images using the non‐prewhitening matched filter (NPWMF), evaluated the uniformity of left ventricular (LV) wall in terms of image intensity, and quantified the effect of smoothing on the spatial resolution of the reconstructed LV wall by using its full‐width at half‐maximum (FWHM). Results Compared to OSEM with Gaussian postfiltering, the DL denoised images with CU‐Net significantly improved the detection performance of perfusion defects at all contrast levels (65%, 50%, 35%, and 20%). The signal‐to‐noise ratio (SNRD) in the NPWMF output was increased on average by 8% over optimal Gaussian smoothing (P < 10−4, paired t‐test), while the inter‐subject variability was greatly reduced. The CU‐Net also outperformed a 3D nonlocal means (NLM) filter and a convolutional autoencoder (CAE) denoising network in terms of SNRD. In addition, the FWHM of the LV wall in the reconstructed images was varied by less than 1%. Furthermore, CU‐Net also improved the detection performance when the images were processed with less post‐reconstruction smoothing (a trade‐off of increased noise for better LV resolution), with SNRD improved on average by 23%. Conclusions The proposed DL with N2N training approach can yield additional noise suppression in SPECT‐MPI images over conventional postfiltering. For perfusion defect detection, DL with CU‐Net could outperform conventional 3D Gaussian filtering with optimal setting as well as NLM and CAE.

show abstract

“…Interestingly, it has just recently been shown that classical denoising using a Wiener filter improves automatic neural-based detection of calcifications [5]. In [6], a convolutional neural network (CNN) was proposed for the denoising of low-dose digital breast tomosynthesis (DBT) images. The network was trained and tested using two breast cadaver phantoms.…”

Section: Introductionmentioning

confidence: 99%

Neural Denoising of Ultra-low Dose Mammography

Green

Sklair‐Levy

Kiryati

et al. 2019

Machine Learning for Medical Image Reconstruction

View full text Add to dashboard Cite

X-ray mammography is commonly used for breast cancer screening. Radiation exposure during mammography restricts the screening frequency and minimal age. Reduction of radiation dose decreases image quality. Image denoising has been recently considered as a way to facilitate dose reduction in mammography without impacting its diagnostic value. We propose a convolutional locally-consistent non-local means (CLC-NLM) algorithm for ultra-low dose mammography denoising. The proposed method achieves powerful denoising while preserving fine details in high resolution mammography. Validation is performed using a dataset of 16 digital mammography cases (4-views each). Since obtaining true low-dose and high-dose mammogram pairs raises regulatory concerns, we applied the X-ray specific and validated method of Veldkamp et al. to simulate 90% dose reduction. The proposed algorithm is shown to compete favorably, both quantitatively and qualitatively, against stateof-the-art neural denoising algorithms. In particular, tiny micro-calcifications are better preserved using the proposed algorithm.

show abstract

Radiation dose reduction in digital breast tomosynthesis (DBT) by means of deep-learning-based supervised image processing

Cited by 28 publications

References 29 publications

Separation of bones from soft tissue in chest radiographs: Anatomy‐specific orientation‐frequency‐specific deep neural network convolution

Separation of bones from soft tissue in chest radiographs: Anatomy‐specific orientation‐frequency‐specific deep neural network convolution

Deep learning with noise‐to‐noise training for denoising in SPECT myocardial perfusion imaging

Neural Denoising of Ultra-low Dose Mammography

Contact Info

Product

Resources

About