hide
Free keywords:
-
Abstract:
Longitudinal (T1) and transverse (T2) relaxation are the most fundamental physical processes governing the signal intensity and the soft tissue contrast in magnetic resonance imaging (MRI). The T1 and T2 relaxation times are characteristic properties of living tissues and they were observed to be altered in the context of several diseases. Commonly, the MR images acquired in the clinical routine for medical assessment provide qualitative rather than quantitative information about the T1 and T2 relaxation times of the tissues of interest. The signal intensity generated by conventional qualitative MR imaging may be predominantly T1- or T2-weighted but is generally dependent on other tissue-specific parameters such as the proton density, the acquisition protocol, and the used MR hardware.
It is beneficial to directly measure the T1 and T2 relaxation times by acquiring typically two or more MR images of the same anatomical region but of different contrasts and to represent the results in quantitative maps. While these maps have the same structural appearance as the acquired base images, the individual pixel values have a physical meaning (i.e. the values of T1 and T2 in milliseconds) instead of displaying the signal intensity in arbitrary units. Quantitative maps of tissue-specific MR parameters are superior to conventional MR images since they are ideally independent of the MR protocol and hardware, and thus offer the possibility to directly compare the results from studies across multiple subjects, time-points, and imaging sites. Relaxation time measurements (also referred to as relaxometry) have demonstrated increased specificity and sensitivity to detect pathological tissue changes compared to conventional T1- and T2-weighted MR imaging.
In practice, T1 and T2 quantification techniques which are based on the acquisition of purely T1- and T2-weighted images without residual sensitivity on T2 and T1, respectively, require prohibitively long scan times making them not suited for the clinical practice. As a consequence, fast MR imaging using steady-state free precession (SSFP) sequences has come into the research focus of quantitative MRI. The images obtained from SSFP acquisitions show generally a mixed T1 and T2 contrast. Contemporary SSFP-based techniques for relaxation time measurements are impaired by a T2-related bias in the T1 quantification and by a T1-related bias in the T2 quantification. In addition, their speed comes at the expense of increased sensitivity to extrinsic instrumental factors (e.g. static or transmit field inhomogeneities).
In this thesis, new robust SSFP-based relaxometry methods are developed which offer clinically acceptable scan times and considerably reduce or even eliminate the T2- and T1-bias in the T1 and T2 calculation, respectively, of targets such as the human brain and the musculoskeletal system.
