A Knowledge-Based Automatic Segmentation of the Cerebral Computerized Tomography Images
LI Chuan-f · Chinese journal of medical instrumentation · 2004
Based on the prior knowledge that there are distinct difference of CT values and morphologic characteristics among different regions of cerebral CT images, and some information stored in the DICOM file, we have been successful in automatically segmenting the different regions of cerebral CT images, including the headrest, scalp, skull, CSF and the brain. This method could be used as the pre-process of more advanced algorithm, 3D reconstruction, quantification of cerebral atrophy and digital brain atlas.