GPU-Based Ray-Casting of Spherical Functions Applied to High Angular Resolution Diffusion Imaging.

Area of Science:

• Computer Graphics
• Scientific Visualization
• Medical Imaging

Background:

• Smooth, real-valued functions on a sphere can be represented using spherical harmonics.
• Visualizing complex data, such as high angular resolution diffusion imaging (HARDI), requires efficient rendering techniques.

Purpose of the Study:

• To develop a fast and efficient algorithm for rendering radial graphs of functions on a sphere.
• To enable interactive visualization of large HARDI datasets.

Main Methods:

• Utilized Laplace expansion to represent functions on a sphere as sums of spherical harmonics.
• Developed a CPU and GPU-based algorithm employing ray-casting within a GPU fragment shader.
• Implemented the rendering of glyphs representing the function's radial graph.

Main Results:

• The algorithm achieves fast and efficient rendering of spherical functions.
• The implementation can simultaneously display thousands of glyphs.
• The rendering speed allows for interactive manipulation of large HARDI datasets.

Conclusions:

• The proposed rendering algorithm is highly effective for visualizing HARDI data.
• The GPU-accelerated approach facilitates real-time analysis of complex diffusion imaging datasets.
• This method enhances the interactive exploration of high-dimensional scientific data.