PMID- 29254355
OWN - NLM
STAT- MEDLINE
DCOM- 20180731
LR  - 20181113
IS  - 1600-0455 (Electronic)
IS  - 0284-1851 (Print)
IS  - 0284-1851 (Linking)
VI  - 59
IP  - 9
DP  - 2018 Sep
TI  - New reconstruction algorithm for digital breast tomosynthesis: better image 
      quality for humans and computers.
PG  - 1051-1059
LID - 10.1177/0284185117748487 [doi]
AB  - Background The image quality of digital breast tomosynthesis (DBT) volumes 
      depends greatly on the reconstruction algorithm. Purpose To compare two DBT 
      reconstruction algorithms used by the Siemens Mammomat Inspiration system, 
      filtered back projection (FBP), and FBP with iterative optimizations (EMPIRE), 
      using qualitative analysis by human readers and detection performance of machine 
      learning algorithms. Material and Methods Visual grading analysis was performed 
      by four readers specialized in breast imaging who scored 100 cases reconstructed 
      with both algorithms (70 lesions). Scoring (5-point scale: 1 = poor to 
      5 = excellent quality) was performed on presence of noise and artifacts, 
      visualization of skin-line and Cooper's ligaments, contrast, and image quality, 
      and, when present, lesion visibility. In parallel, a three-dimensional 
      deep-learning convolutional neural network (3D-CNN) was trained (n = 259 
      patients, 51 positives with BI-RADS 3, 4, or 5 calcifications) and tested (n = 46 
      patients, nine positives), separately with FBP and EMPIRE volumes, to 
      discriminate between samples with and without calcifications. The partial area 
      under the receiver operating characteristic curve (pAUC) of each 3D-CNN was used 
      for comparison. Results EMPIRE reconstructions showed better contrast (3.23 vs. 
      3.10, P = 0.010), image quality (3.22 vs. 3.03, P < 0.001), visibility of 
      calcifications (3.53 vs. 3.37, P = 0.053, significant for one reader), and fewer 
      artifacts (3.26 vs. 2.97, P < 0.001). The 3D-CNN-EMPIRE had better performance 
      than 3D-CNN-FBP (pAUC-EMPIRE = 0.880 vs. pAUC-FBP = 0.857; P < 0.001). Conclusion 
      The new algorithm provides DBT volumes with better contrast and image quality, 
      fewer artifacts, and improved visibility of calcifications for human observers, 
      as well as improved detection performance with deep-learning algorithms.
FAU - Rodriguez-Ruiz, Alejandro
AU  - Rodriguez-Ruiz A
AUID- ORCID: 0000-0002-7554-5561
AD  - 1 Department of Radiology and Nuclear Medicine, Radboud University Medical 
      Center, Nijmegen, the Netherlands.
FAU - Teuwen, Jonas
AU  - Teuwen J
AD  - 1 Department of Radiology and Nuclear Medicine, Radboud University Medical 
      Center, Nijmegen, the Netherlands.
FAU - Vreemann, Suzan
AU  - Vreemann S
AD  - 1 Department of Radiology and Nuclear Medicine, Radboud University Medical 
      Center, Nijmegen, the Netherlands.
FAU - Bouwman, Ramona W
AU  - Bouwman RW
AD  - 2 Dutch Expert Centre for Screening (LRCB), Nijmegen, the Netherlands.
FAU - van Engen, Ruben E
AU  - van Engen RE
AD  - 2 Dutch Expert Centre for Screening (LRCB), Nijmegen, the Netherlands.
FAU - Karssemeijer, Nico
AU  - Karssemeijer N
AD  - 1 Department of Radiology and Nuclear Medicine, Radboud University Medical 
      Center, Nijmegen, the Netherlands.
FAU - Mann, Ritse M
AU  - Mann RM
AD  - 1 Department of Radiology and Nuclear Medicine, Radboud University Medical 
      Center, Nijmegen, the Netherlands.
FAU - Gubern-Merida, Albert
AU  - Gubern-Merida A
AD  - 1 Department of Radiology and Nuclear Medicine, Radboud University Medical 
      Center, Nijmegen, the Netherlands.
FAU - Sechopoulos, Ioannis
AU  - Sechopoulos I
AD  - 1 Department of Radiology and Nuclear Medicine, Radboud University Medical 
      Center, Nijmegen, the Netherlands.
AD  - 2 Dutch Expert Centre for Screening (LRCB), Nijmegen, the Netherlands.
LA  - eng
PT  - Journal Article
DEP - 20171218
PL  - England
TA  - Acta Radiol
JT  - Acta radiologica (Stockholm, Sweden : 1987)
JID - 8706123
SB  - IM
MH  - *Algorithms
MH  - Artifacts
MH  - Breast Neoplasms/*diagnostic imaging
MH  - Female
MH  - Humans
MH  - Machine Learning
MH  - Mammography/*methods
MH  - Radiographic Image Enhancement/*methods
MH  - Radiographic Image Interpretation, Computer-Assisted/*methods
PMC - PMC6088454
OTO - NOTNLM
OT  - Digital breast tomosynthesis
OT  - deep learning
OT  - reconstruction algorithms
OT  - visual grading analysis
EDAT- 2017/12/20 06:00
MHDA- 2018/08/01 06:00
PMCR- 2018/08/13
CRDT- 2017/12/20 06:00
PHST- 2017/12/20 06:00 [pubmed]
PHST- 2018/08/01 06:00 [medline]
PHST- 2017/12/20 06:00 [entrez]
PHST- 2018/08/13 00:00 [pmc-release]
AID - 10.1177_0284185117748487 [pii]
AID - 10.1177/0284185117748487 [doi]
PST - ppublish
SO  - Acta Radiol. 2018 Sep;59(9):1051-1059. doi: 10.1177/0284185117748487. Epub 2017 
      Dec 18.