Subject-Specific Breast Biomechanics Dataset: Prone MRI, Standing 3D Scans, and Finite-Element Parameter Identification (Natural vs. Post-Augmentation)

Overview
This repository accompanies the manuscript Evaluation of Natural Breasts and Post-Augmentation Breasts with Silicone Implants Using Subject-Specific Finite Element Modeling. It provides de-identified tabular data and a complete set of visualization figures that document the optimal parameter fits for 12 subjects (6 natural; 6 post-augmentation).
Cohorts
Group 1 (1-L to 6-L): healthy volunteers with natural breasts (n=6). Group 2 (7-L to 12-L): participants after silicone-implant augmentation (n = 6).
Acquisition & Modeling summary
Prone MRI provides anatomy for segmentation and meshing; standing 3D surface scans capture posture-dependent morphology under gravity; the left breast of each subject. Finite-element simulations use a first-order Ogden model. During optimization, cohort-level density (ρ) and shear modulus (μ) are fixed; the subject-specific strain-hardening coefficient (α) is identified by minimizing the point-to-surface mismatch quantified by the Hausdorff distance (HDF, mm).
Files included
dataset.xlsx — subject-level summary table and analysis outputs. Columns include subject index, group, material settings (ρ, μ), α, and the corresponding HDF.
figures.jpg — 12 point-cloud deviation maps (JPG, high resolution), one per subject, e.g., 1-L.jpg … 2-L.jpg. Each figure shows: scatter of surface nodes in a 2D projection (X–Y, mm); colorbar = pointwise deviation (mm) between the FE-predicted standing shape and the 3D surface scan; header text listing the optimized HDF, ρ, μ, and α for that subject.
Units
ρ: T/mm³; μ: MPa; α: dimensionless; HDF: mm.
Quality/validation
Across all subjects, simulations reproduced standing morphology with millimeter-scale error (typical HDF ≈ 1–3 mm). Augmented breasts generally required higher effective μ and showed altered, more variable α compared with natural breasts.
Ethics & de-identification
All data are fully de-identified. Imaging/scan outputs contain no personal identifiers. Study procedures were approved by the local ethics committee/IRB with written informed consent.