A novel image processing scheme for extracting the intracranial contou
rs in axial magnetic resonance data sets is proposed. The scheme incor
porates the method of active contour models, a recently introduced par
adigm for contour extraction. Its performance is nearly ideal for T2-w
eighted images. Although the performances for proton-density-weighted
images and T1-weighted images drop slightly, qualitatively satisfactor
y extraction still can be obtained for T1-weighted images. Due to high
degree of automation, the scheme should help speed up some image proc
essing applications that require the presegmentation of the intracrani
al cavity.