A novel scheme for performing detection and measurements in medical images is presented. The technique is based on active contours evolving in time according to intrinsic geometric measures of the image. The evolving contours naturally split and merge, allowing the simultaneous detection of several objects and both interior and exterior boundaries. The proposed approach is based on the relation between active contours and the computation of geodesics or minimal distance curves. The minimal distance curve lays in a Riemannian space whose metric is defined by the image content. Measurements are performed after the object is detected. Due to the high accuracy achieved by the proposed geodesic approach, it is natural to use it to compute area or length of the detected object, which are of extreme value for diagnosis. Open curves with fix boundaries are computed as well. This addition to the deformable model adds flexibility, allowing the user to choose guiding points in the image or to select regions for measurements. Experimental results of applying the scheme to real medical images demonstrate its potential. The results may be extended to 3D object segmentation as well.
[1]
S. Angenent.
Parabolic equations for curves on surfaces Part II. Intersections, blow-up and generalized solutions
,
1991
.
[2]
Laurent D. Cohen,et al.
On active contour models and balloons
,
1991,
CVGIP Image Underst..
[3]
Yun-Gang Chen,et al.
Uniqueness and existence of viscosity solutions of generalized mean curvature flow equations
,
1989
.
[4]
V. Caselles,et al.
A geometric model for active contours in image processing
,
1993
.
[5]
ISAAC COHEN,et al.
Using deformable surfaces to segment 3-D images and infer differential structures
,
1992,
CVGIP Image Underst..
[6]
Guillermo Sapiro,et al.
Area and Length Preserving Geometric Invariant Scale-Spaces
,
1995,
IEEE Trans. Pattern Anal. Mach. Intell..
[7]
V. Caselles,et al.
What is the best causal scale-space for 3D images
,
1994
.