Automatic Segmentation of Optic Nerve Fibers

The cross-sectional morphometry of different fiber types is an important tool in studying the structure and function of healthy and diseased human optic nerve. However, such studies are hampered by the thousands of fibers involved when manual segmentation has to be used. We have developed an automatic segmentation method that combines several image processing techniques. First, using region growing segments the axon and myelin sheath by combining the feature information together with spatial information, and obtains a binary image. Next, identify the axon candidates by labeling closed region and remove false axons. Then the connected myelin sheaths are separated from each other using the maximum gradient magnitude of the outer annulus. Our results indicate that this approach for segmenting optic nerve fiber images is fast, accurate, and reproducible compared with manual segmentation.