The impact of cranial bones and fontanels-sutures on neuronal source localization in infants by either high-density electroencephalography or high-density functional optical imaging has been proven. However, the main issue in pre-matures and neonates is accurate identification and extraction of the fontanels-sutures which induce inhomogeneity in the skull and show significant differences in specific conductivities and optical properties. Hence it is crucial to provide a realistic model of the skull that identifies the cranial bones and fontanels-sutures. Such a model can be created by segmenting archived CT images which have been acquired for clinical purposes.
In this study, a new framework based on variational coupled level set has been developed for the extraction of new born skull including fontanels and sutures from CT images. Beside segmentation, the method is designed to have surface reconstruction properties. This approach applies a pair of interior/exterior surfaces as geodesic active regions propagating towards and interacting with each other. The moving surfaces are forced to stop alongside of the outer (convex) and inner (concave) surface of cranial bones using edge information. In locations corresponding to fontanels-sutures, these moving surfaces touch each other without crossing over. The proposed method utilizes hard tissue contrast in CT image, prior information of head shape integrated in level sets initialization, and a predefined constraint to impose surface reconstruction properties. The proposed method was evaluated using eighteen neonatal CT images. The segmentation results achieved by the suggested method have been compared with manual segmentations by two different raters, performed to establish a reliable reference. The comparison of the two segmentation results using the Dice similarity coefficient and modified Hausdorff distance shows that the proposed approach provides satisfactory results.