Model-based deep CNN-regularized reconstruction for digital breast tomosynthesis with a task-based CNN image assessment approach
Creative Commons Attribution license..
Objective. Digital breast tomosynthesis (DBT) is a quasi-three-dimensional breast imaging modality that improves breast cancer screening and diagnosis because it reduces fibroglandular tissue overlap compared with 2D mammography. However, DBT suffers from noise and blur problems that can lower the detectability of subtle signs of cancers such as microcalcifications (MCs). Our goal is to improve the image quality of DBT in terms of image noise and MC conspicuity.Approach. We proposed a model-based deep convolutional neural network (deep CNN or DCNN) regularized reconstruction (MDR) for DBT. It combined a model-based iterative reconstruction (MBIR) method that models the detector blur and correlated noise of the DBT system and the learning-based DCNN denoiser using the regularization-by-denoising framework. To facilitate the task-based image quality assessment, we also proposed two DCNN tools for image evaluation: a noise estimator (CNN-NE) trained to estimate the root-mean-square (RMS) noise of the images, and an MC classifier (CNN-MC) as a DCNN model observer to evaluate the detectability of clustered MCs in human subject DBTs.Main results. We demonstrated the efficacies of CNN-NE and CNN-MC on a set of physical phantom DBTs. The MDR method achieved low RMS noise and the highest detection area under the receiver operating characteristic curve (AUC) rankings evaluated by CNN-NE and CNN-MC among the reconstruction methods studied on an independent test set of human subject DBTs.Significance. The CNN-NE and CNN-MC may serve as a cost-effective surrogate for human observers to provide task-specific metrics for image quality comparisons. The proposed reconstruction method shows the promise of combining physics-based MBIR and learning-based DCNNs for DBT image reconstruction, which may potentially lead to lower dose and higher sensitivity and specificity for MC detection in breast cancer screening and diagnosis.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2023 |
|---|---|
| Erschienen: |
2023 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:68 |
|---|---|
| Enthalten in: |
Physics in medicine and biology - 68(2023), 24 vom: 13. Dez. |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Gao, Mingjie [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
Deep convolutional neural network |
|---|
| Anmerkungen: |
Date Completed 16.12.2023 Date Revised 16.12.2023 published: Electronic Citation Status MEDLINE |
|---|
| doi: |
10.1088/1361-6560/ad0eb4 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM364800437 |
|---|
| LEADER | 01000caa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM364800437 | ||
| 003 | DE-627 | ||
| 005 | 20231227133744.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231226s2023 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1088/1361-6560/ad0eb4 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1231.xml |
| 035 | |a (DE-627)NLM364800437 | ||
| 035 | |a (NLM)37988758 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Gao, Mingjie |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Model-based deep CNN-regularized reconstruction for digital breast tomosynthesis with a task-based CNN image assessment approach |
| 264 | 1 | |c 2023 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 16.12.2023 | ||
| 500 | |a Date Revised 16.12.2023 | ||
| 500 | |a published: Electronic | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a Creative Commons Attribution license. | ||
| 520 | |a Objective. Digital breast tomosynthesis (DBT) is a quasi-three-dimensional breast imaging modality that improves breast cancer screening and diagnosis because it reduces fibroglandular tissue overlap compared with 2D mammography. However, DBT suffers from noise and blur problems that can lower the detectability of subtle signs of cancers such as microcalcifications (MCs). Our goal is to improve the image quality of DBT in terms of image noise and MC conspicuity.Approach. We proposed a model-based deep convolutional neural network (deep CNN or DCNN) regularized reconstruction (MDR) for DBT. It combined a model-based iterative reconstruction (MBIR) method that models the detector blur and correlated noise of the DBT system and the learning-based DCNN denoiser using the regularization-by-denoising framework. To facilitate the task-based image quality assessment, we also proposed two DCNN tools for image evaluation: a noise estimator (CNN-NE) trained to estimate the root-mean-square (RMS) noise of the images, and an MC classifier (CNN-MC) as a DCNN model observer to evaluate the detectability of clustered MCs in human subject DBTs.Main results. We demonstrated the efficacies of CNN-NE and CNN-MC on a set of physical phantom DBTs. The MDR method achieved low RMS noise and the highest detection area under the receiver operating characteristic curve (AUC) rankings evaluated by CNN-NE and CNN-MC among the reconstruction methods studied on an independent test set of human subject DBTs.Significance. The CNN-NE and CNN-MC may serve as a cost-effective surrogate for human observers to provide task-specific metrics for image quality comparisons. The proposed reconstruction method shows the promise of combining physics-based MBIR and learning-based DCNNs for DBT image reconstruction, which may potentially lead to lower dose and higher sensitivity and specificity for MC detection in breast cancer screening and diagnosis | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a deep convolutional neural network | |
| 650 | 4 | |a digital breast tomosynthesis | |
| 650 | 4 | |a image denoising | |
| 650 | 4 | |a image reconstruction | |
| 650 | 4 | |a microcalcification | |
| 650 | 4 | |a model observer | |
| 650 | 4 | |a task-based image quality assessment | |
| 700 | 1 | |a Fessler, Jeffrey A |e verfasserin |4 aut | |
| 700 | 1 | |a Chan, Heang-Ping |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Physics in medicine and biology |d 1956 |g 68(2023), 24 vom: 13. Dez. |w (DE-627)NLM000336505 |x 1361-6560 |7 nnns |
| 773 | 1 | 8 | |g volume:68 |g year:2023 |g number:24 |g day:13 |g month:12 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1088/1361-6560/ad0eb4 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 68 |j 2023 |e 24 |b 13 |c 12 | ||