An Improved Ray Casting Method for Rendering Radial Anterior Segment Optical Coherence Tomography Images

Medical image three-dimensional reconstruction technology has been widely used in medical diagnosis and treatment planning. Previous studies have mainly focused on parallel image sequence. Along with the optical coherence tomography images applied in ophthalmic diagnosis, radial scan pattern was used for anterior segment 3D-OCT images. This paper proposes a novel volume rendering algorithm to render radial image sequence. Our method uses image processing technology and adaptive threshold segmentation to segment cornea and iris in individual 2D images and subsequently uses coordinate transformation to align the preprocessing images to the same coordinate system. Finally, the principle of ray casting algorithm is used for rendering these rearranged images, and a trilinear interpolation method aiming at radial image sequence is used to reconstruct 3D model. The results show that our method can indeed realize volume rendering reconstruction for radial image sequence. Our obtained results can be further used in anterior segment diagnose and surgical planning and positioning.