상용 자기 공명 영상 기법을 이용하여 측정한 양성자 밀도 지방 분획의 일치도와 재현성 평가를 위한 다제조사, 다기관 팬텀 연구

Abstract

Purpose: To evaluate the agreement and reproducibility of proton density fat fraction (PDFF) measurement using commercial magnetic resonance (MR) sequences across different imagers, vendors, and field strengths through a phantom experiment. Materials and Methods: Eleven fat-water emulsion phantoms of varying fat proportion (i.e., 0-50 weight %) were constructed. Phantom PDFF was estimated using commercial chemical shift-based MR imaging (CS-MRI) sequences with Siemens 1.5T and 3.0T, Philips 3.0T, and GE 1.5T and 3.0T imagers and using MR spectroscopic sequence (HISTO) with Siemens 1.5T and 3.0T imagers. Agreement in estimated PDFF values between commercial sequences was evaluated with Bland-Altman analysis. Reproducibility of PDFF measurement across commercial sequences was evaluated using the reproducibility coefficient. Test-retest repeatability of PDFF measurement was evaluated using the repeatability coefficient. Results: Repeatability coefficients of PDFF measurement was 0.31% - 1.58% in absolute PDFF value for commercial sequences. Statistically significant biases in the estimated PDFF were noted in 19 of 21 pairwise comparisons of commercial sequences (range of mean biases, -4.48% to 8.15% in absolute PDFF value). Reproducibility coefficient of PDFF measurement was 9.0% in absolute PDFF value (95% confidence interval, 8.4%, 9.7%) over all commercial sequences and 10.6% in absolute PDFF value (95% CI, 9.8%, 11.6%) over all CS-MRI sequences. Conclusions: The measurement of PDFF with commercial MR sequences is highly repeatable. However, it is not reproducible across sequences, imager vendors, and field strengths. For longitudinal follow-up of hepatic steatosis using commercial MR sequences for PDFF measurement, the use of the same sequence and imager is recommended.