We are investigating computerized techniques for sorting mammograms ac
cording to whether the breast tissue is fatty or dense. The hypothesis
is that areas of dense tissue are a major factor in making certain ma
mmograms harder for both radiologists and computers to interpret. Bein
g able to identify dense mammograms automatically could permit better
use of the time and skills of expert radiologists by allowing the diff
icult mammograms to be examined by the most experienced readers. In ad
dition, the scope for computer-aided detection of abnormalities might
be increased by concentrating on the easier, fatty mammograms. The mam
mograms used in the experiment were classified independently by two ra
diologists, who agreed in almost all cases. A number of local statisti
cal and texture measures were then computed for patches from digitizat
ions of these mammograms. One of the measures (local skewness in tiles
) gives a good separation between fatty and dense patches. This measur
e has been incorporated into an automated procedure that separates off
approximately two thirds of the fatty mammograms. This finding has be
en replicated on mammograms taken from a UK screening programme. The r
elationship between the fatty/dense distinction and the classification
proposed by Wolfe is discussed.