Functional lung imaging with transient spoiled gradient echo.

Abstract

To introduce an alternative framework for perfusion and ventilation lung imaging at 3T using transient spoiled gradient echo (tSPGR) acquisitions. Sets of coronal 2D time-resolved lung image series were acquired in 5 healthy volunteers using tSPGR and compared with contemporary SPGR and ultrafast balanced SSFP (uf-bSSFP) implementations at 1.5T and 3T. Sequence parameters and view ordering were optimized for tSPGR to yield maximum signal intensity in the lung tissue. Signal-to-noise ratio and contrast-to-noise ratio analyses were performed in all acquired tSPGR, SPGR, and uf-bSSFP data sets. Matrix pencil decomposition was applied to generate functional parameter maps, including fractional ventilation, relative perfusion, and blood arrival time. For the lung, the signal intensity of tSPGR imaging was maximal for minimal TR and TE settings of 0.99ms and 0.43ms, respectively. Moreover, low RF spoiling increments in combination with a centric view ordering resulted in a further signal-to-noise ratio increase of about 30% to 40%. The average signal-to-noise ratio in the lung parenchyma was 73.3 for uf-bSSFP, 38.1 for tSPGR, 20.7 for SPGR at 1.5T, and 31.2 for uf-bSSFP, 35.6 for tSPGR, and 21.3 for SPGR at 3T. The average ventilation and perfusion contrast-to-noise ratio was 33.2 and 36.2 for uf-bSSFP, 15.4 and 12.5 for tSPGR, 13.5 and 4.1 for SPGR at 1.5T, and 16.5 and 11.3 for uf-bSSFP, 29.7 and 50.8 for tSPGR, and 22.4 and 16.5 for SPGR at 3T, respectively. At 3T, application of balanced SSFP is limited, so tSPGR offers an alternative framework for successful lung function assessment using matrix pencil MRI.