Validation of T1 and T2 algorithms for quantitative MRI: performance by a vendor-independent software

Abstract

Background

Determination of the relaxation time constants T1 and T2 with quantitative magnetic resonance imaging is increasingly used for both research and clinical practice. Recently, groups have been formed within the Society of Cardiovascular Magnetic Resonance to address issues with relaxometry. However, so far they have avoided specific recommendations on methodology due to lack of consensus and current evolving research. Standardised widely available software may simplify this process.

The purpose of the current study was to develop and validate vendor-independent T1 and T2 mapping modules and implement those in the versatile and widespread software Segment, freely available for research and FDA approved for clinical applications.

Results

The T1 and T2 mapping modules were developed and validated in phantoms at 1.5 T and 3 T with reference standard values calculated from reference pulse sequences using the Nelder-Mead Simplex optimisation method. The proposed modules support current commonly available MRI pulse sequences and both 2- and 3-parameter curve fitting. Images acquired in patients using three major vendors showed vendor-independence. Bias and variability showed high agreement with T1 and T2 reference standards for T1 (range 214–1752 ms) and T2 (range 45–338 ms), respectively.

Conclusions

The developed and validated T1 and T2 mapping and quantification modules generated relaxation maps from current commonly used MRI sequences and multiple signal models. Patient applications showed usability for three major vendors.