Improved image contrast and scan efficiency for fat suppressed T1-weighted imaging at 3T with a spin echo two-point Dixon technique

Introduction: Fat suppressed T1-weighted imaging is often an essential component of a clinical MRI exam. Compared to several relatively newer pulse sequences (e.g., spoiled gradient echo, T1FLAIR, FSE), the conventional spin echo (SE) sequence still produces the best T1-contrast and is preferred by many radiologists for certain applications (such as for imaging of head & neck). However, inclusion of the frequency-selective fat suppression (CHESS) pulses in a SE acquisition substantially reduces the number of slices per TR and therefore the overall scan efficiency. More importantly, the image contrast (as well as the signal strength) can be adversely affected from the incidental magnetization transfer (MT) effect of the CHESS pulses [1,2] . At 3 Tesla, the reduced T1 contrast and image quality are further degraded by the prolonged T1-relaxation times and possible dielectric effects. Consequently, the consensus of the radiologists at our institution and at other institutions that we know of is that the T1-contrast of the fat-suppressed SE images is much worse at 3 Tesla than at 1.5 Tesla. Lack of a satisfactory fat-suppressed T1-weighted sequence is currently the main reason for 3 Tesla scanners to be excluded at our institution from certain MR studies (e.g., of the head & neck). The purpose of this work was to develop a new SE two-point Dixon technique to address these limitations of the conventional fat-suppressed SE acquisition. The substantial improvement in scan efficiency and image quality by the new method is demonstrated in vivo.