Rationale and Objectives. In some full-field digital mammography systems, m
ultiple detectors are abutted :together, and the physical gaps between adja
cent detectors produce seams between the resultant subimages. In this study
, a variety of interpolation algorithms for estimating the missing informat
ion in the seams were compared, and their effect on image quality was evalu
ated.
Materials and Methods. Eight representative interpolation algorithms were s
elected, including nearest neighbor, one-dimensional and two-dimensional we
ighting, mean value, one-dimensional and two-dimensional polynomial, and on
e-dimensional and two-dimensional cubic spline interpolation methods. These
methods were applied to digital mammograms and phantom images. The effecti
veness of each algorithm was evaluated for accuracy and geometric distortio
n.
Results, These interpolation algorithms offered similar accuracy in estimat
ing missing image information. The weighting. polynomial, and cubic spline
interpolation algorithms introduced less geometric distortion than the near
est neighbor and mean value interpolation algorithms. All algorithms were m
ore effective in estimating larger, lower-contrast features (such as breast
masses) than in estimating smaller, higher-contrast features (such as brea
st microcalcifications). Small microcalcifications within the seams cannot
be recovered with interpolation. The probability of a microcalcification in
a seam is small, however, and the failure to image a few microcalcificatio
ns of a cluster generally does not substantially alter diagnostic performan
ce.
Conclusion. In the development of full-field digital breast imaging systems
, appropriate interpolation algorithms can satisfactorily fill in narrow pa
r between adjacent detectors. The one-dimensional weighting interpolation m
ethod seems an effective and efficient choice.