An Adaptive Enhancement Algorithm for CT Brain Images.

We present an adaptive smoothing filter aiming to improve the visibility and detectability of the obscuration of the lentiform nucleus, which is one of the early signs of acute cerebral artery infraction, in CT brain images. The proposed method is able to enhance image data by removing noise without significantly blurring the structures in the images. This method is tested both on computer-simulated image data and on clinical CT brain images. Moreover, a comparison of the proposed method to two commonly used techniques is made. Preliminary results demonstrate the superiority of the proposed method and its usefulness for detecting the obscuration of the lentiform nucleus, which could not be detected previously without using the proposed method.