Spherical harmonics to quantify cranial asymmetry in deformational plagiocephaly

Area of Science:

• Medical Imaging
• Computational Geometry
• Pediatric Health

Background:

• Infant deformational plagiocephaly (DP) is common, affecting a significant percentage of infants.
• Current manual measurement methods for DP diagnosis lack inter-user accuracy.
• Three-dimensional (3D) models offer rich data but often rely on manual measurements for analysis.

Purpose of the Study:

• To develop an automatic and accurate methodology for detecting DP using 3D head models.
• To leverage the extensive information within 3D models for improved DP assessment.
• To introduce spherical harmonics as a novel technique for DP identification.

Main Methods:

• Computed an ideal fitted ellipsoid for each infant's 3D head model.
• Calculated orthogonal distances between the head model and the fitted ellipsoid.
• Modeled these distances using spherical harmonics, focusing on coefficients of degree 2 and order -2.

Main Results:

• Spherical harmonic coefficients of degree 2 and order -2 effectively represent the asymmetry characteristic of DP.
• This novel coefficient demonstrated superior performance compared to traditional anthropometric indexes (e.g., Asymmetry Index).
• A maximum degree of 4 for the spherical harmonic analysis yielded the best results in DP detection.

Conclusions:

• Spherical harmonics provide an accurate and automated method for detecting infant deformational plagiocephaly from 3D head models.
• This technique surpasses the diagnostic capabilities of current manual measurements and standard anthropometric deformation indexes.
• The proposed method enhances the precision and reliability of DP diagnosis in clinical practice.