In vivo evaluation of population-specific inversion pulses in parallel transmission

Area of Science:

• Neuroimaging
• Magnetic Resonance Imaging (MRI)
• Pulse Sequence Optimization

Background:

• The MP2RAGE sequence is a key technique for brain imaging at 7 Tesla (7T).
• Standard universal pulses (UP) in MP2RAGE may not optimally suit all individuals due to anatomical variations.
• Optimizing radiofrequency (RF) pulses is crucial for maximizing image quality and diagnostic information in high-field MRI.

Purpose of the Study:

• To evaluate the in vivo performance of population-specific RF pulses for the MP2RAGE sequence at 7T.
• To assess the effectiveness of a novel automated pipeline for selecting tailored pulses based on head anatomy.

Main Methods:

• A cohort of 39 volunteers was analyzed, and their head shapes were clustered into five groups.
• Population-specific inversion pulses were computed for each cluster, alongside a universal pulse (UP).
• A Python pipeline integrated with image reconstruction classified head shapes and selected appropriate pulses, including UP as a fallback, without increasing scan time.

Main Results:

• The automated pipeline successfully classified head shapes and selected suitable population-specific pulses, enhancing MP2RAGE image contrast.
• Population-specific pulses reduced performance loss compared to the UP, achieving results closer to fully customized solutions.
• Benefits were most pronounced for smaller head sizes, while larger or more atypical head shapes showed less improvement.

Conclusions:

• A novel head clustering and pulse selection pipeline enables practical implementation of population-specific pulses in clinical 7T brain MRI.
• This automated approach tailors RF pulse selection to individual head anatomy without extending acquisition time.
• The method enhances MP2RAGE image quality by mitigating limitations of a one-size-fits-all pulse strategy.