Purpose: We prospectively quantified the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of integrated parallel acquisition technique (PAT) and simultaneous multislice (SMS) acceleration and various combinations thereof, and we further compared two 4-fold-accelerated (PAT2-SMS2) high-resolution turbo spin echo (TSE) magnetic resonance imaging (MRI) protocols of the knee against a clinical 2-fold-accelerated (PAT2-SMS1) TSE standard.
Materials and methods: Institutional review board approval was obtained, and all subjects gave informed consent. Fourteen knee MRI examinations were obtained (8 men, 6 women; mean age, 46 years; age range, 28-62 years) using a 3 T MRI system and a TSE pulse sequence prototype that allowed for the combination of PAT and SMS acceleration. Predicted whole-body specific absorption rates were recorded for all pulse sequences. For quantitative analysis, the difference method was used to calculate SNR and CNR analysis of 6 different TSE acceleration schemes (PAT2-SMS1, PAT3-SMS1, PAT1-SMS2, PAT1-SMS3, PAT2-SMS2, and PAT2-SMS3). For qualitative analysis, sagittal intermediate-weighted and axial fat-suppressed T2-weighted MR images were obtained with PAT2-SMS1 and PAT2-SMS2 acceleration schemes using similar parameters. One faster PAT2-SMS2 acceleration scheme with decreased repetition time and longer echo train was labeled with the addition SPEED for the purpose of this report. Two readers rated the data sets for image quality, structural visibility, and overall observer satisfaction using equidistant 5-point Likert scales. Readers additionally noted the presence of cartilage defects, meniscal tears, tendons and ligament tears, and bone marrow edema pattern. Friedman and Kruskal-Wallis tests were used. P values of less than 0.01 were considered significant.
Results: All pulse sequences were successfully executed and reconstructed inline. Whole-body specific absorption rates ranged between 1.4 and 3.9 W/kg for all acquisitions and remained within mandated limits. Quantitatively, mean SNR and CNR were significantly higher for SMS than those for PAT and similar for PAT2-SMS2 and PAT2-SMS1. Fluid was brightest on PAT2-SMS1, whereas noise, edge sharpness, contrast resolution, and fat suppression were similar on PAT2-SMS1 and SMS2-PAT2 and mildly inferior on PAT2-SMS2 SPEED. Articular cartilage received mildly higher ratings on PAT2-SMS1, whereas visibility of menisci was mildly inferior on PAT2-SMS2 SPEED. Observer preferences were similarly high for PAT2-SMS1 and PAT2-SMS2 and mildly inferior for the faster PAT2-SMS2 SPEED images. Four cartilage defects and 2 meniscal tears were seen by both readers on all sequences.
Conclusions: We demonstrate the signal preservation capabilities of SMS over PAT acceleration, which allow for similar SNR and CNR of 4-fold PAT2-SMS2 and 2-fold PAT2-SMS1 acceleration. Four-fold-accelerated TSE through the combination of PAT2 and SMS2 enables approximately 50% shorter acquisition times compared with regular PAT2 acceleration, similar quantitative and qualitative image quality, and holds promise for a meaningful increase of the efficiency of clinical 2-dimensional MRI of the knee.