Key points are not available for this paper at this time.
A stroke occurs if the flow of oxygen-rich blood to a portion of the brain is blocked. Without oxygen, brain cells start to die after a few minutes. Sudden bleeding in the brain also can cause a stroke if it damages brain cells. Stroke is a harmful disease in which most of the cases of this disease bring an effect of the physical defect on patients. One of the methods of examining this disease is by means of head CT scan. However, there is a weakness of head CT scan for consisting of some unclear or invisible parts. Hence, it needs a technique for improving the image quality to again emerge the invisible parts. Adaptive histogram equalization (AHE) is a technique that can cope with the weakness of HE by enhancing the contrast in local area. The contrast enhancement, however, can be excessive sometimes. Using the contrast limited adaptive histogram equalization (CLAHE), the excessive contrast enhancement in AHE can be coped with by giving the limit value on histogram. In this research, a stroke identification system has been built comprising image enhancement with CLAHE, feature extraction statistically with the mean value, deviation standard, skewness, kurtosis, and segmentation using the statistical region merging method. The result then showed that the method of image processing significantly conducted was able to be used as a tool to identify the stroke disease in order to distinguish the type of CT scan to the normal or sick state.
Nurhayati et al. (Thu,) studied this question.