Related Experiment Videos
Automated method for generating the arterial input function on perfusion-weighted MR imaging: validation in patients
Michael Mlynash1, Irina Eyngorn, Roland Bammer
1Department of Neurology and Neurological Sciences, Stanford Stroke Center, Palo Alto, CA, USA.
AJNR. American Journal of Neuroradiology
|June 16, 2005
Summary
An automated arterial input function (AIF) program significantly improves the speed and consistency of perfusion-weighted MR imaging (PWI) analysis for stroke evaluation. This automated method enhances the accuracy of blood flow maps, reducing operator dependency.
Area of Science:
- Medical Imaging
- Radiology
- Neurology
Background:
- Arterial input function (AIF) selection critically impacts perfusion-weighted MR imaging (PWI) blood flow map accuracy.
- Manual AIF identification is operator-dependent, inconsistent, and time-consuming.
Purpose of the Study:
- To develop and validate an automated AIF identification program (auto-AIF).
- To assess the performance of auto-AIF compared to manual methods in PWI analysis.
- To evaluate the impact of automated AIF on Tmax and cerebral blood flow (CBF) map generation.
Main Methods:
- Comparison of auto-AIF with manually derived AIFs in 22 stroke patients' PWIs.
- Analysis of parameters: time to peak, curve width, curve height, and voxel location.
- Calculation of Tmax and CBF maps on a per-pixel basis.
- Spatial correlation analysis of map pairs using Pearson correlation coefficients.
Main Results:
- Auto-AIF-derived PWI map parameters were consistently superior to manual methods.
- Excellent reproducibility for auto-AIF-based Tmax maps (r = 1.0).
- Well-correlated Tmax and CBF maps between auto-AIF and manual techniques (r = 0.82).
- Significant reduction in AIF identification time with auto-AIF (mean difference: 72 seconds).
Conclusions:
- Automated AIF identification is feasible, producing reproducible and accurate Tmax and CBF maps.
- Automation reduces PWI analysis time and enhances consistency.
- Auto-AIF holds potential for more effective acute stroke evaluation using PWI.