Utilization of MR angiography in perfusion imaging for identifying arterial input function

Autor: Mehmed Ozkan, Bora Buyuksarac
Rok vydání: 2017
Předmět:
Middle Cerebral Artery
Perfusion Imaging
Biophysics
Partial volume
Contrast Media
Perfusion scanning
computer.software_genre
Magnetic resonance angiography
030218 nuclear medicine & medical imaging
03 medical and health sciences
0302 clinical medicine
Imaging
Three-Dimensional
Voxel
medicine.artery
medicine
Cluster Analysis
Humans
Radiology
Nuclear Medicine and imaging
Arterial input function
cardiovascular diseases
Mathematics
Radiological and Ultrasound Technology
medicine.diagnostic_test
business.industry
Brain Neoplasms
Magnetic resonance imaging
Cerebral blood flow
Cerebrovascular Circulation
Middle cerebral artery
cardiovascular system
biological phenomena
cell phenomena
and immunity
Nuclear medicine
business
computer
030217 neurology & neurosurgery
Magnetic Resonance Angiography
circulatory and respiratory physiology
Zdroj: Magma (New York, N.Y.). 30(6)
ISSN: 1352-8661
Popis: This research utilizes magnetic resonance angiography (MRA) to identify arterial locations during the parametric evaluation of concentration time curves (CTCs), and to prevent shape distortions in arterial input function (AIF). We carried out cluster analysis with the CTC parameters of voxels located within and around the middle cerebral artery (MCA). Through MRA, we located voxels that meet the AIF criteria and those with distorted CTCs. To minimize partial volume effect, we re-scaled the time integral of CTCs by the time integral of venous output function (VOF). We calculated the steady-state value to area under curve ratio (SS:AUC) of VOF and used it as a reference in selecting AIF. CTCs close to this reference value (selected AIF) and those far from it were used (eliminated AIF) to compute cerebral blood flow (CBF). Eliminated AIFs were found to be either on or anterior to MCA, whereas selected AIFs were located superior, inferior, posterior, or anterior to MCA. If the SS:AUC of AIF was far from the reference value, CBF was either under- or over-estimated by a maximum of 41.1 ± 14.3 and 36.6 ± 19.2%, respectively. MRA enables excluding voxels on the MCA during cluster analysis, and avoiding the risk of shape distortions.
Databáze: OpenAIRE
Externí odkaz: