MR-based truncation and attenuation correction in integrated PET/MR hybrid imaging using HUGE with continuous table...

Area of Science:

• Medical Imaging
• Radiology
• Nuclear Medicine

Background:

• PET/MR hybrid imaging offers combined functional and anatomical information.
• Accurate PET quantification is crucial for diagnosis and treatment monitoring.
• MR-based attenuation correction is preferred but can suffer from truncation artifacts.

Purpose of the Study:

• To introduce and evaluate the HUGE (B0 Homogenization using gradient enhancement) method for MR-based attenuation and truncation correction.
• To improve PET quantification in whole-body PET/MR hybrid imaging.
• To compare HUGE with standard MR-based methods and PET-based MLAA algorithms.

Main Methods:

• The HUGE method extends the MR field-of-view for truncation correction.
• HUGE was combined with continuously moving table data acquisition for whole-body coverage.
• The method was validated using NEMA standard phantoms and applied to 24 oncologic patients.

Main Results:

• HUGE significantly reduced geometric distortions and signal truncations compared to standard Dixon-VIBE.
• HUGE provided robust outer contour data, outperforming MLAA in cases with arm truncation.
• HUGE showed comparable SUVmean increases to MLAA, with maximal differences up to 14%, and accurately corrected a case where MLAA failed.

Conclusions:

• The HUGE method effectively reduces truncations in whole-body PET/MR imaging by extending the MR field-of-view.
• As a fully MR-based approach, HUGE is independent of the radiotracer, ensuring robust correction for all patients.
• HUGE improves standard MR-based attenuation correction and PET image quantification in PET/MR applications.