P045] Neonatal brain: Contrast to noise ratio (CNR) optimization of T1-weighted (T1W) and T2-weighted (T2W) sequences at multiple 1.5T MRI systems, utilizing a neonatal brain tissue mimicking phantom

Abstract

Purpose To optimize clinical T1w and T2w sequences at multiple 1.5T MRI systems of different vendors, using an in house developed phantom simulating neonatal brain relaxometric characteristics. Materials Methods A neonatal brain phantom, fabricated from paramagnetically doped agarose gel solutions, underwent quantitative MR relaxometry utilizing an Inversion Recovery Spin Echo (IRTSE) method to estimate T1 relaxation times and a Car-Purcell-Meiboom-Gill (CPMG) method to estimate T2 relaxation times at multiple 1.5T MRI systems. Multi-TI turbo spin-echo and multi-TE spin-echo sequences were utilized for direct T1 and T2 relaxation time measurements. Results Signal data S(TR, TE) were reproduced mathematically according to the standard SE signal acquisition equation: S(TR, TE) = PD*(1-exp(-TR/T1)*(exp(-TE/T2). Consequently, TR and TE values at which the maximum CNR between neonatal gray and white matter was recorded were used for the final optimization of T1w and T2w clinical sequences respectively. Based on the above values, TR and TE parameters were modified resulting in an optimized T1wSE sequence TR/TE = 1200 ms/10 ms and an optimized T2wTSE sequence TR/TE = 9500 ms/280 ms as estimated on a clinical MAGNETOM Sonata 1.5T system. Clinical MRI inter-equipment variability for T1 and T2 measurements remained TR and TE parameter variation in optimized T1w and T2w sequences remained less than 7% amongst the selected MRI systems included in this study. Conclusions CNR optimization of T1w and T2w sequences for neonatal brain imaging at multiple 1.5T clinical MRI systems was accomplished with the aid of an in house developed tissue-mimicking phantom.