Optimization of tomosynthesis acquisition parameters: Angular range and number of projections

Thomas Mertelmeier, Jasmina Ludwig, Bo Zhao, Wei Zhao

Research output: Chapter in Book/Report/Conference proceedingConference contribution

35 Scopus citations


In breast tomosynthesis several projection views of the object are acquired over a limited angular range from which slice images are reconstructed. The image quality of the reconstructed slices, however, depends on the geometric acquisitions parameters and on the reconstruction algorithm. We investigated the impact of the angular range of data acquisition and number of projection views on image quality by performing 3-D cascaded linear system modeling and phantom measurements. The results show that a larger angular range not only increases resolution perpendicular to the slices but also improves the visibility of low-frequency objects since the sampling area in Fourier space is mainly limited at low-frequencies. The effect on contrast resolution is demonstrated also with clinical data sets.

Original languageEnglish (US)
Title of host publicationDigital Mammography - 9th International Workshop, IWDM 2008, Proceedings
Number of pages8
StatePublished - 2008
Event9th International Workshop on Digital Mammography, IWDM 2008 - Tucson, AZ, United States
Duration: Jul 20 2008Jul 23 2008

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume5116 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Other9th International Workshop on Digital Mammography, IWDM 2008
Country/TerritoryUnited States
CityTucson, AZ


  • Artifacts
  • Breast tomosynthesis
  • Data acquisition
  • Filtered backprojection
  • Linear systems theory
  • Reconstruction

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'Optimization of tomosynthesis acquisition parameters: Angular range and number of projections'. Together they form a unique fingerprint.

Cite this