@article{oai:niigata-u.repo.nii.ac.jp:00004124, author = {猪又, 聖美 and 李, 鎔範 and 蔡, 篤儀 and 横山, 龍二郎 and 原, 武史 and 藤田, 広志 and 兼松, 雅之 and 岩間, 亨 and 星, 博昭}, issue = {9}, journal = {日本放射線技術学会雑誌, 日本放射線技術学会雑誌}, month = {Sep}, note = {We propose an approach for automated detection of cerebral vessels from head CT angiographic images. This approach contains two major features. First, instead of using the well-known image-processing techniques such as thresholding and labeling, a novel Laplacian-like filter is developed and employed in the region of interest in an image to be processed. Second, not only is the axial-view image reconstructed from head CT angiographic images used, but, in addition, the sagittal- and coronal-view images are reconstructed and used. By applying these major features in the process of detection of brain vessels, more accurate results can be achieved. To validate the effectiveness of the proposed method, we applied the method to three clinical cases, all of which were head CT angiograms. Our preliminary results showed that the proposed method has the potential to automatically detect cerebral vessels in head CT angiograms with acceptable accuracy.}, pages = {1325--1331}, title = {頭部CTA画像における脳血管領域の自動抽出の試み}, volume = {60}, year = {2004} }