Constructing nasal prosthesis morphological data based on a nonrigid registration algorithm.

Area of Science:

• Medical Engineering
• Computer-Aided Design
• Biomedical Imaging

Background:

• Nasal prostheses are crucial for restoring facial morphology in patients with nasal defects.
• Digital design and manufacture of nasal prostheses require accurate morphological data, but relevant studies are scarce.
• Developing efficient methods for constructing nasal prosthesis morphology is critical for successful repair outcomes.

Purpose of the Study:

• To evaluate the Procrustes Analysis-Nonrigid Iterative Closest Point (PA-NICP) algorithm for rapid construction of nasal prosthesis morphological data.
• To compare the efficacy of the PA-NICP algorithm against the MeshMonk program in generating nasal prosthesis data.
• To assess the nonrigid registration principles for digital prosthetic design.

Main Methods:

• Collected 3D facial data from 30 adult males using a 3D facial scanner.
• Constructed 30 total nasal defect 3D facial datasets using Geomagic Wrap 2021.
• Applied the PA-NICP algorithm (experimental) and MeshMonk program (control) for nasal prosthesis data generation, comparing 3D deviation, edge tightness, and surface continuity.

Main Results:

• No significant difference in 3D morphological deviation (RMS) between PA-NICP (1.51 ±0.45 mm) and MeshMonk (1.34 ±0.31 mm) (P=.054).
• PA-NICP demonstrated significantly better edge tightness (RMS deviation 0.22 ±0.05 mm vs. 0.38 ±0.09 mm; P<.001).
• PA-NICP showed significantly higher edge surface continuity (95.47% vs. 92.20%; P=.001).

Conclusions:

• Both PA-NICP and MeshMonk effectively construct nasal prosthesis data.
• The PA-NICP algorithm provides superior edge tightness and morphological transition compared to MeshMonk.
• PA-NICP is a promising tool for digital nasal prosthesis design and fabrication.