Accelerated MRI thermometry by direct estimation of temperature from undersampled k-space data

Area of Science:

• Medical Imaging
• Biophysics
• Ultrasound Technology

Background:

• Magnetic resonance (MR) thermometry is crucial for MR-guided focused ultrasound applications.
• Current methods face limitations in speed, spatial resolution, and volume coverage.
• Accelerated MR thermometry is needed for advanced applications.

Purpose of the Study:

• To develop and validate a constrained reconstruction method for accelerating MR thermometry.
• To enable faster data acquisition without compromising temperature accuracy.
• To improve MR-guided focused ultrasound capabilities.

Main Methods:

• A novel constrained reconstruction method directly estimates temperature from k-space data.
• The method fits a model to undersampled k-space data, removing aliasing artifacts.
• Comparisons were made with parallel imaging (SPIRiT) and conventional thermometry techniques.

Main Results:

• Simulations and phantom experiments showed superior temporal resolution and lower error compared to existing methods.
• The k-space method achieved higher acceleration factors (up to 32×) with minimal temperature error.
• Demonstrated advantage over parallel imaging alone for overcoming undersampling.

Conclusions:

• The proposed k-space reconstruction algorithm allows for high acceleration factors in MR thermometry.
• Temporal resolution and accuracy are maintained even at accelerated rates.
• This advancement supports more demanding MR-guided focused ultrasound applications.