Combination of 3D and 4D MRA Image Sequences for Improved Quantitative and Visual Rating of Cerebrovascular Malformations

BACKGROUND Cerebral vascular diseases (CVD) are one of the most common causes of death worldwide. Exact knowledge of the individual anatomy and hemodynamic situation is needed for disease rating. New MRA imaging enables the acquisition of spatial and spatiotemporal image sequences which offer either detailed information about vessel topology or blood flow. The purpose of this work was to develop a software tool named AnToNIa (Analysis Tool for Neuro Imaging Data) which enables a combined analysis and visualization of 3D/4D MRA datasets for an improved diagnosis and therapy planning of CVDs AnToNIa was developed for commonly available MRA datasets like high resolution 3D TOF, 3D contrast enhanced, 4D TREAT and 4D TWIST MRA sequences. AnToNIa consists of four major parts. First, pre-processing methods for reducing MRA artifacts like the slap-boundary-artifact, magnetic field inhomogeneities and noise are implemented. Second, a number of automatic segmentation methods e.g. for brain extraction and vessel segmentation for quantitative image analysis are available. The hemodynamic analysis is the third part. Here, concentration time curves are extracted and analyzed based on 4D MRA datasets. Beside state-of-the-art approaches a new linear curve fit approach was used to determine the bolus arrival time. The last part contains the 3D/4D MRA image registration, which enables the combined analysis and interactive blood flow visualization on a patient-individual 3D vessel surface (see Fig). EVALUATION A retrospective rating of 20 AVM patients was performed to evaluate AnToNIa. The patients were rated using standard DSA datasets and the combined 3D/4D MRA visualization. A comparison of the results showed that the dynamic blood flow visualization offers advanced clinical information and allows comparable rating results. At the same time a reduction of the required rating time was observed. DISCUSSION AnToNIa has so far only been incorporated in retrospective medical studies for rating of CVDs. It is planned to provide AnToNIa under the required common approvals and certificates in near future. CONCLUSION AnToNIa enables the combined analysis and visualization of common 3D/4D MRA datasets for improved diagnosis of CVDs. FIGURE (OPTIONAL) http://media.rsna.org/media/abstract/2010/9001297/9001297_vnij.jpg