Patient-Specific Skeletal Muscle Fiber Modeling from Structure Tensor Field of Clinical CT Images

摘要: We propose an optimization method for estimating patient-specific muscle fiber arrangement from clinical CT. Our approach first computes the structure tensor field to estimate local orientation, then a geometric template representing is fitted using B-spline deformation by maximizing fitness of orientation smoothness penalty. The initialization computed with previously proposed algorithm that takes account only muscle’s surface shape. Evaluation was performed CT volume (1.0 mm\(^\text {3}\)/voxel) and high resolution optical images serial cryo-section (0.1 {3}\)/voxel). mean distance error at 6.00 mm decreased 2.78 after gluteus maximus muscle, 5.28 3.09 medius muscle. result 20 patient suggested reconstructed anatomically more plausible than previous method.