Abstract
MR imaging is a very important method for the diagnosis of intracranial lesions. A measurement of the volume of the temporal lobe is a very useful method in diagnosing temporal lobe abnormalities. Therefore we develop the algorithm of the automated volumetry of the temporal lobe. However, it is often difficult to segment the temporal lobe. The Sylvian fissure is an important index for the segmentation of the temporal lobe. The purpose of this study is to develop the algorithm of the automated identification of the Sylvian fissure. We studied 10 normal volunteers and 2 patients who were suspected of Alzheimer disease. We performed the following four steps: (1) Segmentation of the brain on coronal images. (2) Edge enhancement of the y-direction on sagittal images. (3) Hough transform on edge-enhanced images. (4) Correction of R-L direction and S-I direction. In results, this method proved to be able to identify the Sylvian fissure on coronal and sagittal MR images in normal volunteers and patients. It is concluded that the Sylvian fissure identified by this method is a very useful index of the segmentation of the temporal lobe.
