Highly undersampled phase-contrast flow measurements using compartment-based k-t principal component analysis.

Area of Science:

• Cardiovascular Imaging
• Medical Physics
• Biomedical Engineering

Background:

• Phase-contrast imaging for cine blood flow measurements is limited by long scan times, hindering clinical applicability.
• Existing k-t principal component analysis (PCA) and k-t SENSE methods face challenges in achieving high accuracy with accelerated imaging.

Purpose of the Study:

• To develop an advanced k-t PCA method for accurate and accelerated cine blood flow imaging.
• To improve reconstruction accuracy by defining spatial compartment-dependent temporal basis functions.
• To enable faster acquisition of 2D velocity and pulse wave velocity data in the aorta.

Main Methods:

• An extension of the k-t PCA method was developed using spatial compartment-dependent temporal basis functions.
• The method was evaluated for its reconstruction accuracy compared to original k-t PCA and k-t SENSE.
• Prospective undersampling was applied, achieving a net acceleration factor of 8.

Main Results:

• The proposed method demonstrated significant improvements in reconstruction accuracy.
• Stroke volume error was maintained below 5% with prospective nominal undersampling of up to 16 (net acceleration factor of 8).
• Successful acquisition of 2D cine flow data in the aorta was achieved at 100 frames per second.

Conclusions:

• The enhanced k-t PCA method allows for significantly accelerated cine blood flow measurements with high accuracy.
• This advancement facilitates clinical applications, such as 2D velocity and pulse wave velocity assessment in the aorta, within a single breathhold.
• The method overcomes limitations of traditional phase-contrast imaging by reducing scan times while maintaining diagnostic quality.