Skip to Main Content
In MRI images, the boundary of each encephalic tissue is highly irregular. Sphere-Shaped Support Vector Machine (SSSVM) has the advantage of solving high non-linearity and irregularity problems. In this paper, SSSVM is applied in brain MRI image 3D reconstruction. Selecting parameters for SSSVM, however, is a complicated problem. Appropriate parameters can make the model more flexible and help to obtain more accurate data description. In this paper, Immune Algorithm (IA) is used in searching for optimal SSSVM parameters. Immune SSSVM (ISSSVM) is proposed to reconstruct 3D model of each encephalic tissue. As shown by the experiment of this study, encephalic tissues are reconstructed efficiently and satisfied accuracies and visual effects are obtained.