Brain segmentation is a challenging problem due to the complexity of the brain. In this paper, we propose an automated brain segmentation method for 3D magnetic resonance (MR) brain images which are represented as a sequence of 2D brain images. The proposed method consists of three steps: pre-processing, removal of non-brain regions (e.g., the skull, meninges, other organs, etc), and spinal cord restoration. In pre-processing, we perform adaptive thresholding which takes into account variable intensities of MR brain images corresponding to various image acquisition conditions. In segmentation process, we iteratively apply 2D morphological operations and masking for the sequences of 2D sagittal, coronal, and axial planes in order to remove non-brain tissues. Next, final 3D brain regions are obtained by applying OR operation for segmentation results of three planes. Finally we reconstruct the spinal cord truncated during the previous processes. Experiments are performed with fifteen 3D MR brain image sets with 8-bit gray-scale. Experiment results show the proposed algorithm is fast, and provides robust and satisfactory results.
Perfusion CT has been successfully used as a functional imaging technique for diagnosis of patients with hyperacute stroke. However, the commonly used methods based on curve-fitting are time consuming. Numerous researchers have investigated to what extent Perfusion CT can be used for the quantitative assessment of cerebral ischemia and to rapidly obtain comprehensive information regarding the extent of ischemic damage in acute stroke patients. The aim of this study is to propose an alternative approach to rapidly obtain the brain perfusion mapping and to show the proposed cerebral flow imaging of the vessel and tissue in human brain be reliable and useful. Our main design concern was algorithmic speed, robustness and automation in order to allow its potential use in the emergency situation of acute stroke. To obtain a more effective mapping, we analyzed the signal characteristics of Perfusion CT and defined the vessel-around model which includes the vessel and tissue. We proposed a nonparametric vessel-around approach which automatically discriminates the vessel and tissue around vessel from non-interested brain matter stratifying the level of maximum enhancement of pixel-based TAC. The stratification of pixel-based TAC was executed using the mean and standard deviation of the signal intensity of each pixel and mapped to the cerebral flow imaging. The defined vessel-around model was used to show the cerebral flow imaging and to specify the area of markedly reduced perfusion with loss of function of still viable neurons. Perfusion CT is a fast and practical technique for routine clinical application. It provides substantial and important additional information for the selection of the optimal treatment strategy for patients with hyperacute stroke. The vessel-around approach reduces the computation time significantly when compared with the perfusion imaging using the GVF. The proposed cerebral imaging shows reliable results which are validated by physicians and medical staff. Moreover the vessel-around approach was found to be comprehensive and easy-to-interpret by physicians and medical staff, hence we conclude that our proposed vessel-around technique can be used for brain perfusion mapping.