Studies reported in the literature indicate that breast cancer risk is asso
ciated with mammographic densities. An objective, repeatable, and a quantit
ative measure of risk derived from mammographic densities will be of consid
erable use in recommending alternative screening paradigms and/or preventiv
e measures. However, image processing efforts toward this goal seem to be s
parse in the literature, and automatic and efficient methods do not seem to
exist. In this paper, we describe and validate an automatic and reproducib
le method to segment dense tissue regions from fat within breasts from digi
tized mammograms using scale-based fuzzy connectivity methods. Different me
asures for characterizing mammographic density are computed from the segmen
ted regions and their robustness in terms of their linear correlation acros
s two different projections-cranio-caudal and medio-lateral-oblique-are stu
died. The accuracy of the method is studied by computing the area of mismat
ch of segmented dense regions using the proposed method and using manual ou
tlining. A comparison between the mammographic density parameter taking int
o account the original intensities and that just considering the segmented
area indicates that the former may have some advantages over the latter.