Development of CAD scheme for automated detection of intracranial aneurysms in magnetic resonance angiography

Abstract

We have developed a computer-aided diagnostic (CAD) scheme for automated detection of intracranial aneurysms in magnetic resonance angiography (MRA). A total of 29 cases with 36 aneurysms and 31 non-aneurysm cases were processed by use of the selective enhancement filter for dots (aneurysms). All candidates were classified into four types according to the size and local structures based on the effective diameter and the skeleton image of each candidate. A number of false positives were removed in each group by use of rule-based schemes and linear discriminant analysis. All of 36 aneurysms were correctly detected with 0.55 false positives per patient. The CAD scheme would be useful in assisting radiologists for detection of intracranial aneurysms in MRA.