banner image
No data available.
Please log in to see this content.
You have no subscription access to this content.
No metrics data to plot.
The attempt to load metrics for this article has failed.
The attempt to plot a graph for these metrics has failed.
Accuracy in breast shape alignment with 3D surface fitting algorithms
Rent this article for


Image of FIG. 1.
FIG. 1.

Exemplifying nipple position reproducibility before (left panel) and after (right panel) the application of the constrained fitting algorithm.

Image of FIG. 2.
FIG. 2.

Breast phantom aligned with lasers. The numbered circles (1–7) represent the features used for texture localization validation. Features 1–3 are the BBs used for phantom alignment.

Image of FIG. 3.
FIG. 3.

Spatial configuration of the additional texture features identified in patient 1 (left panel) and patient 2 (right panel).

Image of FIG. 4.
FIG. 4.

Histogram of nipple positioning error for the unconstrained fitting algorithm.

Image of FIG. 5.
FIG. 5.

ROI surface mismatch (boxplots) and nipple position reproducibility (symbol ◼, with errors bars denoting the quartile range) shown as a function of the applied constraint. The average surface mismatch within the ROI (whole breast) is computed along the normal direction to the reference surface model. Boxplots depict the 25th, 50th (median), and 75th percentile values, with notches representing the 95% confidence interval of the median value. Whiskers extend from both sides of the box up to 100% of the quartile range and symbols denote outliers. Brackets denote the significantly different groups according to nonparametric post hoc comparison (see text for details) for both the average surface mismatch (symbol ) and nipple position reproducibility (symbol ).

Image of FIG. 6.
FIG. 6.

Exemplifying representation of residual surface mismatch in the ROI for the unconstrained (left panel) and constrained surface fitting (right panel) algorithms. Please note that the depiction in the left panel is deceiving in this case, as the 3D error in nipple repositioning is 9.5 mm, whereas the surface distance along the normal direction in that area is much smaller (around 3 mm).

Image of FIG. 7.
FIG. 7.

Repositioning error of the additional texture features grouped as a function of the 3D distance from the nipple location, with 30 mm wide equally spaced bins. Each bin comprises five to nine texture features, except the last one (150–180 mm distance) that includes two features only. Unconstrained and constrained fittings (1.5–3–5 mm constraint) are shown in the representation.

Image of FIG. 8.
FIG. 8.

Variation in the distance between texture features as a function of the distance from the nipple location for patient 1. The dot-dashed line depicts the corresponding linear fit.


Generic image for table

Texture localization validation in the breast phantom.


Article metrics loading...


Full text loading...

This is a required field
Please enter a valid email address
752b84549af89a08dbdd7fdb8b9568b5 journal.articlezxybnytfddd
Scitation: Accuracy in breast shape alignment with 3D surface fitting algorithms