RSNA 2011 

Abstract Archives of the RSNA, 2011


LL-PHS-WE1A

Short T2 Relaxometry and Image Contrast Enhancement of Aqueous and Lipid Containing Gels in Ultrashort TE (UTE) MRI

Scientific Informal (Poster) Presentations

Presented on November 30, 2011
Presented as part of LL-PHS-WE: Physics

Participants

Felix Liu BS, Presenter: Nothing to Disclose
Hernan Jara PhD, Abstract Co-Author: Patent holder, Synthetic-MR algorithms Patent holder, Q-MRI algorithms

PURPOSE

Image contrast in short T2 decay weighted images in ultrashort TE MRI is commonly enhanced with image subtraction of long TE acquisitions from short TE acquisitions, which may suffer from inadequate long T2 signal reduction or low short T2 signal magnitude. An approach is presented in which estimated mean short T2 signal, mean long T2 signal, and chemical shift effects are used to display the proportion of signal from short and long T2 components.

METHOD AND MATERIALS

A multi-vial phantom was scanned on a 3.0T Achieva MRI scanner (Philips Medical Systems) with 3D UTE FFE and T2* weighted 3D FFE pulse sequences. The phantom consisted of 9 vials of water, agarose gel, olive oil, or agarose gel with water-olive oil mixtures in varying concentrations of 10% to 60% oil. UTE scans were performed with 16 short TEs from 0.16 to 2.6 ms, and 4 moderate TEs of 3.45, 6.9, 9.2 and 18 ms. Long T2 maps were generated from 3D FFE images acquired at TR= 50 ms, initial TE= 2.0 ms, and 16 echoes with ΔTE=3.0 ms. Short T2 relaxometry was estimated using nonlinear least squares fitting to a bi-exponential short and long T2 decay with variable chemical shift effects: S = (BScos(ω*TE)+(1-BS))AS*exp(-TE/T2S) + (BLcos(ω*TE)+(1-BL))AL*exp(-TE/T2L) The long T2 times were constrained to estimates from the T2 maps. Chemical shift modulation, signal amplitude and short T2 coefficients were constrained to realistic ranges. Pixel values in the short T2 signal proportion map are the ratios of short T2 signal to short and long T2 signal, calculated at any specified TE.

RESULTS

R2 values for fitting oil and water-oil mixtures of 20-60% oil were 0.98 or above; the R2 value for the 10% oil mixture was 0.91. Liquid water and water gel, with R2 values of 0.77 and 0.3 respectively, were modeled poorly with short and long T2 bi-exponential fitting.

CONCLUSION

The short T2 relaxometry of oil and water-oil systems may be accurately modeled by fitting a bi-exponential decay with chemical shifts. By generating signal proportion maps, short T2 signals with poor image contrast may be identified.

CLINICAL RELEVANCE/APPLICATION

UTE relaxometry could be useful for characterizing semi-solid lesions and fatty infiltrative pathology, or for displaying tissues with significant but low magnitude short T2 signal components.

Cite This Abstract

Liu, F, Jara, H, Short T2 Relaxometry and Image Contrast Enhancement of Aqueous and Lipid Containing Gels in Ultrashort TE (UTE) MRI.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11007336.html