Predictive Modeling for Personalized Three-Dimensional Burn Injury Assessments.

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Computational Anatomy

Background:

• Accurate burn size estimation is critical for effective patient treatment and minimizing complications.
• Current clinical methods for estimating burn size (percent Total Body Surface Area - TBSA) often lack precision and reliability.

Purpose of the Study:

• To develop and validate a novel 3D-based approach for accurate burn size estimation.
• To utilize personalized 3D models derived from limited anthropometric measurements for improved TBSA assessment.

Main Methods:

• A reliability study identified key anthropometric measurements.
• Machine learning algorithms were employed to determine the most predictive measurements for Body Surface Area (BSA).
• 3D models were constructed using a limited set of measurements and evaluated against gold-standard 3D scans using error metrics and 3D color maps.

Main Results:

• Height and circumferences of the bust, neck, hips, and waist were identified as optimal predictors for BSA.
• 3D models generated from just three to four anthropometric measurements demonstrated high accuracy.
• These models showed close geometric similarity to gold-standard 3D scans.

Conclusions:

• The proposed 3D modeling approach offers a more accurate and reliable method for TBSA assessment in burn patients.
• Personalized 3D models derived from minimal anthropometric data can significantly reduce estimation errors.
• Improved TBSA assessment accuracy is expected to decrease medical/surgical complications and enhance patient outcomes through personalized care.