Evaluation of an Automated Identification System of Unruptured Intracranial Aneurysm Candidates Based on a Blob Structure Enhancement Filter at 3.0 T Magnetic Resonance Angiography

概要

論文の詳細を見る
An automated identification system of unruptured intracranial aneurysm candidates based on a three-dimensional (3D) blob structure enhancement (BSE) filter was evaluated for 3.0 T magnetic resonance angiography (MRA). Ten cases with 10 unruptured aneurysms (diameter, 2.4 to 8.4 mm; mean, 6.6 mm) were used in this study. The isotropic 3D MRA images with 512 x 512 x (224-258) voxels (voxel size, 0.35 mm) were processed by use of the BSE filter. The initial candidates were identified by use of a multiple gray-level thresholding technique on the BSE images and a region growing technique with monitoring some image features. A number of false-positives were removed by using different rules on 17 image features associated with morphology and gray levels. Support vector machine, which can learn the image features of true positives and false positives, was used for further removal of false-positives. With this system, 100% of aneurysms were correctly detected with 3.8 false-positives per patient based on a leave-one-out-by-patient test method. This automated identification system would be useful in assisting radiologists for the detection of intracranial aneurysms in 3.0 T MRA.
一般社団法人電子情報通信学会の論文
2014-01-19