3-D surface reconstruction of multiple sclerosis lesions using spherical harmonics.

Area of Science:

• Medical imaging
• Computational geometry
• Neurology

Background:

• Multiple sclerosis (MS) is a chronic neurological disease characterized by lesions in the central nervous system.
• Accurate quantification of MS lesion volume and shape is crucial for disease monitoring and treatment assessment.
• Current methods for MS lesion volume estimation have limitations in precision and geometrical accuracy.

Purpose of the Study:

• To introduce a new computational method for approximating the three-dimensional (3-D) shape and calculating the volume of multiple sclerosis (MS) lesions.
• To evaluate the accuracy and consistency of this novel method compared to existing techniques.

Main Methods:

• The method employs spherical harmonics to approximate the 3-D surfaces of MS lesions based on contours from segmented magnetic resonance (MR) images.
• The approach was applied to both in vivo and simulated MS lesions to generate 3-D reconstructions and calculate volumes.

Main Results:

• The spherical harmonics method provided good geometrical approximations of MS lesion 3-D shapes.
• Volume estimations demonstrated consistency across different lesion sizes.
• The average volume estimation error was significantly lower (15.5% and 13.1%) compared to the slice stacking technique (25.0%).

Conclusions:

• The presented method offers a robust tool for analyzing the 3-D geometry and volume of MS lesions.
• This approach may provide valuable geometrical information that can serve as an additional clinical index for monitoring MS disease progression.
• Improved accuracy in lesion volume quantification can aid in better patient management and therapeutic evaluations.