Comparison of Arterial Spin-labeling Perfusion Images at Different Spatial Normalization Methods Based on Voxel-based...

Area of Science:

• Neuroimaging
• Medical Image Analysis
• Statistical Parametric Mapping

Background:

• Spatial normalization is crucial for statistical parametric mapping (SPM) analysis of neuroimaging data.
• Arterial spin-labeling (ASL) perfusion imaging provides valuable functional information.
• Optimizing spatial normalization is key to improving diagnostic accuracy in SPM analysis of ASL images.

Purpose of the Study:

• To determine the most effective spatial normalization method for enhancing diagnostic accuracy in SPM analysis of ASL perfusion images.
• To compare the performance of different spatial normalization techniques, including template-based and DARTEL-based approaches.

Main Methods:

• Evaluation of five spatial normalization methods: 3DT1-conventional, 3DT1-DARTEL, 3DT1-conventional template, 3DT1-DARTEL template, and ASL-DARTEL template.
• Comparison of SPM results in patients with Alzheimer's disease or normal pressure hydrocephalus and healthy controls.
• Utilizing anatomical (3DT1) and ASL images for spatial normalization.

Main Results:

• The ASL-DARTEL template method yielded inaccurate SPM results and a smaller template size.
• No significant differences were observed between 3DT1-conventional and 3DT1-DARTEL template methods.
• The 3DT1-DARTEL method demonstrated superior detection sensitivity and precise anatomical localization.

Conclusions:

• Spatial normalization using the DARTEL technique with anatomical images (3DT1-DARTEL) is recommended for ASL perfusion imaging in SPM analysis.
• This method improves diagnostic accuracy and anatomical precision compared to other evaluated techniques.