A residual dense network assisted sparse view reconstruction for breast computed tomography

Fu, Zhiyang; Tseng, Hsin Wu; Vedantham, Srinivasan; Karellas, Andrew; Bilgin, Ali

doi:10.1038/s41598-020-77923-0

Cited by 14 publications

(13 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This also eliminates the generalization problem of supervised networks. For instance, a fully-supervised network trained using sparse-view breast CT data showed reduced performance on calcifications due to being a minority class in the training data 26 . In contrast, we showed unimpaired calcification resolution in our AFN-assisted reconstruction.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

An attenuation field network for dedicated cone beam breast CT with short scan and offset detector geometry

Fu,

Tseng,

Vedantham

2024

Sci Rep

Self Cite

View full text Add to dashboard Cite

The feasibility of full-scan, offset-detector geometry cone-beam CT has been demonstrated for several clinical applications. For full-scan acquisition with offset-detector geometry, data redundancy from complementary views can be exploited during image reconstruction. Envisioning an upright breast CT system, we propose to acquire short-scan data in conjunction with offset-detector geometry. To tackle the resulting incomplete data, we have developed a self-supervised attenuation field network (AFN). AFN leverages the inherent redundancy of cone-beam CT data through coordinate-based representation and known imaging physics. A trained AFN can query attenuation coefficients using their respective coordinates or synthesize projection data including the missing projections. The AFN was evaluated using clinical cone-beam breast CT datasets (n = 50). While conventional analytical and iterative reconstruction methods failed to reconstruct the incomplete data, AFN reconstruction was not statistically different from the reference reconstruction obtained using full-scan, full-detector data in terms of image noise, image contrast, and the full width at half maximum of calcifications. This study indicates the feasibility of a simultaneous short-scan and offset-detector geometry for dedicated breast CT imaging. The proposed AFN technique can potentially be expanded to other cone-beam CT applications.

show abstract

Section: Discussionmentioning

confidence: 99%

“…We also trained two fully supervised networks independently to tackle the incomplete data problem, where the network inputs were obtained using either FDK with Parker weight or FDK with the offset-detector weight. We adopted a multi-slice residual dense network (MS-RDN) 26 as the architecture, which was designed for breast CT reconstruction.…”

Section: Methodsmentioning

confidence: 99%

An attenuation field network for dedicated cone beam breast CT with short scan and offset detector geometry

Fu,

Tseng,

Vedantham

2024

Sci Rep

Self Cite

View full text Add to dashboard Cite

show abstract

“…A total of 75 CBBCT datasets were used in the study. The projection images were reconstructed to (0.273 mm) 3 isotropic voxels by a deep learning‐based, multi‐slice residual dense network (MS‐RDN) 36 assisted algorithm, as this reconstruction provides images with lower noise. Segmentation of the breast volume into air, skin, fibroglandular tissue, and adipose tissue followed the methodology described in prior literature 13,37 .…”

Section: Methodsmentioning

confidence: 99%

Dedicated cone‐beam breast CT: Data acquisition strategies based on projection angle‐dependent normalized glandular dose coefficients

2022

Self Cite

View full text Add to dashboard Cite

Background: Dedicated cone-beam breast computed tomography (CBBCT) using short-scan acquisition is being actively investigated to potentially reduce the radiation dose to the breast.This would require determining the optimal x-ray source trajectory for such short-scan acquisition. Purpose: To quantify the projection angle-dependent normalized glandular dose coefficient (DgN CT ) in CBBCT, referred to as angular DgN CT , so that the x-ray ray source trajectory that minimizes the radiation dose to the breast for short-scan acquisition can be determined. Materials and Methods: A cohort of 75 CBBCT clinical datasets was segmented and used to generate three breast models -(I) patient-specific breast with heterogeneous fibroglandular tissue distribution and real breast shape, (II) patient-specific breast shape with homogeneous tissue distribution and matched fibroglandular weight fraction, and (III) homogeneous semi-ellipsoidal breast with patient-specific breast dimensions and matched fibroglandular weight fraction, which corresponds to the breast model used in current radiation dosimetry protocols. For each clinical dataset, the angular DgN CT was obtained at 10 discrete angles, spaced 36 • apart, for full-scan, circular, x-ray source trajectory from Monte Carlo simulations. Model III is used for validating the Monte Carlo simulation results. Models II and III are used to determine if breast shape contributes to the observed trends in angular DgN CT . A geometry-based theory in conjunction with center-of -mass (COM) based distribution analysis is used to explain the projection angle-dependent variation in angular DgN CT . Results: The theoretical model predicted that the angular DgN CT will follow a sinusoidal pattern and the amplitude of the sinusoid increases when the centerof -mass of fibroglandular tissue (COM f ) is farther from the center-of -mass of the breast (COM b ). It also predicted that the angular DgN CT will be minimized at x-ray source positions complementary to the COM f . The COM f was superior to the COM b in 80% (60/75) of the breasts. From Monte Carlo simulations and for homogeneous breasts (models II and III), the deviation in breast shape from a semi-ellipsoid had minimal effect on angular DgN CT and showed less than 4% variation. From Monte Carlo simulations and for model I, as predicted by our theory, the angular DgN CT followed a sinusoidal pattern with maxima and minima at x-ray source positions superior and inferior to the breast, respectively. For model I, the projection angle-dependent variation in angular DgN CT was 16.4%. Conclusion:The heterogeneous tissue distribution affected the angular DgN CT more than the breast shape. For model I, the angular DgN CT was lowest when the x-ray source was inferior to the breast. Hence, for short-scan CBBCT 1406

show abstract

“…The breast images were all first reconstructed by our developed deep learning-based algorithm, multi-slice residual dense network (MS-RDN) [13], that reduces image noise. Then all MS-RDN reconstructed images were segmented into air, skin, adipose, and fibroglandular tissues (Fig.…”

Section: Angular 𝑫𝒈𝑵 𝑪𝑻 Computationmentioning

confidence: 99%

“…To our best knowledge, studies in literature only considered 𝐷𝑔𝑁 𝐶𝑇 , but none have considered the variation of the 𝐷𝑔𝑁 𝐶𝑇 with the projection angles, i.e., angular 𝐷𝑔𝑁 𝐶𝑇 . We are particularly interested in 𝐷𝑔𝑁 𝐶𝑇 because we have developed feasible image reconstruction algorithms for short-scan and sparse-view acquisitions [12], [13], and we would like to understand which angular range should be used for short-scan acquisition to reduce the radiation to the breast either for prone or upright patient-position CBBCT systems. A cohort of 75 CBBCT datasets from a research database of subjects who had participated in a prior IRB-approved clinical trial was used in this study.…”

Section: Introductionmentioning

confidence: 99%

Angular normalized glandular dose coefficient in breast CT: clinical data study

Tseng

Karellas²,

Vedantham³

2022

7th International Conference on Image Formation in X-Ray Computed Tomography

View full text Add to dashboard Cite

The goal of this study is to understand how the normalized glandular dose coefficient (𝐷𝑔𝑁 𝐶𝑇 ) varies with projection angle in dedicated cone-beam breast computed tomography (CBBCT). Seventy-five CBBCT clinical datasets from a research database were used for this study. All samples were segmented into skin, adipose and fibroglandular tissues. The segmented volumes were used in a Monte Carlo simulation package (GATE 8.0) to estimate the radiation dose at 10 angles in a full scan. An analytical model is proposed, and this model predicted that the angular 𝐷𝑔𝑁 𝐶𝑇 follows a sine wave and the maximum is related to the center of geometry of the fibroglandular tissue (COG f ). The angular 𝐷𝑔𝑁 𝐶𝑇 from Monte Carlo simulations was consistent with our model and follows a sine wave with amplitude of 0.0376. The maximum of the wave occurs when the x-ray source is approximately at head position, which is consistent with our model. Our results indicate that the higher angular 𝐷𝑔𝑁 𝐶𝑇 occurs when the x-ray source is superior to the breast. This suggests using a x-ray source trajectory inferior to the breast for short-scan CBBCT design.

show abstract

A residual dense network assisted sparse view reconstruction for breast computed tomography

Cited by 14 publications

References 63 publications

An attenuation field network for dedicated cone beam breast CT with short scan and offset detector geometry

An attenuation field network for dedicated cone beam breast CT with short scan and offset detector geometry

Dedicated cone‐beam breast CT: Data acquisition strategies based on projection angle‐dependent normalized glandular dose coefficients

Angular normalized glandular dose coefficient in breast CT: clinical data study

Contact Info

Product

Resources

About