BREAST-E: A PHOTOGRAMMETRY-BASED PREOPERATIVE METHOD FOR QUANTITATIVE BREAST MORPHOMETRIC ANALYSIS AND VOLUME ESTIMATION IN RECONSTRUCTIVE SURGER smhoong76@um.edu.my

225-05

BREAST-E: A PHOTOGRAMMETRY-BASED PREOPERATIVE METHOD FOR QUANTITATIVE BREAST MORPHOMETRIC ANALYSIS AND VOLUME ESTIMATION IN RECONSTRUCTIVE SURGER

7

Malaysia

Oral or Poster

Accurate breast volume measurement in low-resource settings faces challenges like limited technology and financial constraints. Photogrammetry-based methods provide a cost-effective, non-invasive alternative by reconstructing 3D models from 2D images. This study aims to develop and validate the BREAST-E method, focusing on standardized protocols, accuracy, and clinical validation for surgical planning workflows.

Breast cancer patients scheduled for mastectomy were recruited following ethical approval. Inclusion criteria included age ≥18, mobility, and imaging consent. Photogrammetry captured 2D breast images at three angles: 0° (straight-on), 20° above, and 20° below. Markers (20, 40, 60) were placed on anatomical landmarks for alignment during 3D reconstruction using Photomodeler software. Accuracy, consistency, and reliability were validated against mastectomy specimen volumes. Statistical analysis (SPSS, MedCal) included correlation coefficients, Bland-Altman analysis, and error metrics (MAE, RMSE). Subgroup analysis identified optimal imaging configurations for clinical application.

Photogrammetry-based breast volume estimation showed strong accuracy in 3D phantom models (R = 0.899, R² = 80.8%, MAE = 98.12 ml) but slightly reduced in real patients (R = 0.789, R² = 62.2%, MAE = 104.69 ml). Reliability demonstrated good internal consistency (Cronbach’s alpha = 0.8749) but lower ICC in real patients (-0.0568). Consistency varied, with minimal variability in phantom models (SD = 92.04 ml) and increased variability in real patients (SD = 99.57 ml). Anatomical variability, ptosis, asymmetry, and marker placement caused discrepancies.

Photogrammetry is a cost-effective, reliable method for breast volume estimation. Workflow refinements and validation are needed to address anatomical variability, improving applicability in resource-limited clinical settings.