Surface alignment to unmask scoliotic deformity in surface topography

Area of Science:

• Biomedical Engineering
• Orthopedics
• Medical Imaging

Background:

• Scoliosis monitoring using back surface topography is crucial for treatment assessment.
• Variations in patient stance significantly alter back surface shape, confounding scoliosis measurements.
• Distinguishing scoliosis-induced shape changes from postural variations is a key challenge.

Purpose of the Study:

• To develop a method for separating scoliosis-related back shape changes from stance variations.
• To quantify the noise introduced by postural changes in back surface topography measurements.
• To investigate the potential for modeling stance variations to improve scoliosis monitoring.

Main Methods:

• Collected back surface topography data from patients with scoliosis.
• Employed mathematical alignment techniques to standardize back shapes and quantify similarity.
• Utilized a surface similarity index (root-mean-square of residual differences) to estimate noise and assess modeling effectiveness.

Main Results:

• Postural variation (stance) contributed significant noise, indicated by a surface similarity index averaging 2.0 mm for repeated measurements.
• Back shapes measured during routine clinical visits showed greater variation (average index of 3.5 mm), highlighting the impact of stance.
• The study demonstrated evidence that noise due to stance variations can be modeled.

Conclusions:

• Stance variations represent a significant source of noise in 3D back surface comparisons for scoliosis monitoring.
• Recognizing and quantifying postural noise is essential for accurate scoliosis assessment.
• Modeling stance variations shows promise for improving the reliability of back surface topography in clinical settings.