Automatic segmentation of whole breast using atlas approach and deformable image registration

Abstract

Purpose: To compare interobserver variations in delineating the whole breast for treatment planning using two contouring methods.

Methods and materials: Autosegmented contours were generated by a deformable image registration-based breast segmentation method (DEF-SEG) by mapping the whole breast clinical target volume (CTVwb) from a template case to a new patient case. Eight breast radiation oncologists modified the autosegmented contours as necessary to achieve a clinically appropriate CTVwb and then recontoured the same case from scratch for comparison. The times to complete each approach, as well as the interobserver variations, were analyzed. The template case was also mapped to 10 breast cancer patients with a body mass index of 19.1-35.9 kg/m(2). The three-dimensional surface-to-surface distances and volume overlapping analyses were computed to quantify contour variations.

Results: The median time to edit the DEF-SEG-generated CTVwb was 12.9 min (range, 3.4-35.9) compared with 18.6 min (range, 8.9-45.2) to contour the CTVwb from scratch (30% faster, p = 0.028). The mean surface-to-surface distance was noticeably reduced from 1.6 mm among the contours generated from scratch to 1.0 mm using the DEF-SEG method (p = 0.047). The deformed contours in 10 patients achieved 94% volume overlap before correction and required editing of 5% (range, 1-10%) of the contoured volume.

Conclusion: Significant interobserver variations suggested a lack of consensus regarding the CTVwb, even among breast cancer specialists. Using the DEF-SEG method produced more consistent results and required less time. The DEF-SEG method can be successfully applied to patients with different body mass indexes.

Fig. 1

(a) Auto-segmentation of the whole breast using a model template patient. The contoured breast (red), sternum (blue), and spinal cord (purple) in a model patient are shown in the top row. The unmodified, deformed contours in the test patient after a deformable image registration are shown as solid lines in the bottom row. (b) Additional axial images in the test patient from the superior border (top-left) to the inferior border (bottom-right).

Fig. 1

(a) Auto-segmentation of the whole breast using a model template patient. The contoured breast (red), sternum (blue), and spinal cord (purple) in a model patient are shown in the top row. The unmodified, deformed contours in the test patient after a deformable image registration are shown as solid lines in the bottom row. (b) Additional axial images in the test patient from the superior border (top-left) to the inferior border (bottom-right).

Fig. 2

An axial (left), coronal (middle), and sagittal (right) CT images showing the whole breast contours of the model patient (top panel) and the mapped (and deformed) whole breast contours of a test patient (bottom panel). The red contours are the whole breast definition as specified by our in-house guidelines. The green contours are based on the National Surgical Adjuvant Breast and Bowel Project whole breast definition. The deformed contours show a good consistency in contour definition with the model case. The arrows illustrate a few anatomical landmarks, at which the deformed contours maintained the same relationship as their original definition in the model patient.

Fig. 3

Inter-observer variations in contouring the whole breast. The standard deviations (left), maximum deviations (middle), and mean differences (right) among eight physicians are plotted in color-coded 3D surface maps. The color scale is shown at the bottom in centimeter units. To demonstrate the maximum differences, the surface plots were generated from an angle from the inside of the patient looking out, as indicated by the yellow arrow in the bottom-left diagram. The plots in row (A) show the variations for contouring from scratch, and the plots in row (B) show the variations when the contours were modified from the computer-generated contours.

Fig. 4

Inter-observer variations in contouring of the posterior-lateral border. The solid green contour is the “average” contour, overlayed with all eight physician contours in red. The relationship of the posterior-lateral border with the maximum 3D surface-to-surface distance plot is also shown. The 3D surface distance plots are in the same orientation as in Figure 3.

Fig. 5

Three-dimensional spatial deviations between the unedited DEF-SEG-generated CTVwb and the physician-edited CTVwb in the 10 test patients. Orientation of the 3D CTVwb and the color scale of the plot are the same as in Figure 3.

Fig. 6

The volume overlap index, as indicated by the Dice Similarity Coefficient (DSC), reduced with the increase of the difference between the test patient’s body mass index (BMI) and the template patient’s BMI. A small DSC indicates a reduced agreement between the physician-approved contours and the unedited deformably mapped contours.