Premium
Quantitative imaging metrics derived from magnetic resonance fingerprinting using ISMRM/NIST MRI system phantom: An international multicenter repeatability and reproducibility study
Author(s) -
Shridhar Konar Amaresha,
Qian Enlin,
Geethanath Sairam,
Buonincontri Guido,
Obuchowski Nancy A.,
Fung Maggie,
Gomez Pedro,
Schulte Rolf,
Cencini Matteo,
Tosetti Michela,
Schwartz Lawrence H,
ShuklaDave Amita
Publication year - 2021
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1002/mp.14833
Subject(s) - reproducibility , repeatability , imaging phantom , nist , nuclear medicine , standard deviation , magnetic resonance imaging , scanner , nuclear magnetic resonance , medicine , mathematics , statistics , physics , computer science , radiology , optics , natural language processing
Purpose To compare the bias and inherent reliability of the quantitative (T 1 and T 2 ) imaging metrics generated from the magnetic resonance fingerprinting (MRF) technique using the ISMRM/NIST system phantom in an international multicenter setting. Method ISMRM/NIST MRI system phantom provides standard reference T 1 and T 2 relaxation values (vendor‐provided) for each of the 14 vials in T 1 and T 2 arrays. MRF‐SSFP scans repeated over 30 days on GE 1.5 and 3.0 T scanners at three collaborative centers. MRF estimated T 1, and T 2 values averaged over 30 days were compared with the phantom vendor‐provided and spin‐echo (SE) based convention gold standard (GS) method. Repeatability and reproducibility were characterized by the within‐case coefficient of variation (wCV) of the MRF data acquired over 30 days, along with the biases. Result For the wide ranges of MRF estimated T 1 values, vials #1‐8 (T 1 relaxation time between 2033 and 184 ms) exhibited a wCV less than 3% and 4%, respectively, on 3.0 and 1.5 T scanners. T 2 values in vials #1‐8 (T 2 relaxation, 1044‐45 ms) have shown wCV to be <7% on both 3.0 and 1.5 T scanners. A stronger linear correlation overall for T 1 (R 2 = 0.9960 and 0.9963 at center‐1 and center‐2 on 3.0 T scanner, and R 2 = 0.9951 and 0.9988 at center‐1 and center‐3 on 1.5 T scanner) compared to T 2 (R 2 = 0.9971 and 0.9972 at center‐1 and center‐2 on 3.0 T scanner, and R 2 = 0.9815 and 0.9754 at center‐1 and center‐3 on 1.5 T scanner). Bland–Altman (BA) analysis showed MRF based T 1 and T 2 values were within the limit of agreement (LOA) except for one data point. The mean difference or bias and 95% lower bound (LB) and upper bound (UB) LOA are reported in the format; mean bias: 95% LB LOA: 95% UB LOA. The biases for T 1 values were 21.34: −50.00: 92.69, 21.32: −47.29: 89.94 ms, and for T 2 values were −19.88: −42.37: 2.61, −19.06: −43.58: 5.45 ms on 3.0 T scanner at center‐1 and center‐2, respectively. Similarly, on 1.5 T scanner biases for T 1 values were 26.54: −53.41: 106.50, 9.997: −51.94: 71.94 ms, and for T 2 values were −23.84: −135.40: 87.76, −37.30: 134.30: 59.73 ms at center‐1 and center‐3, respectively. Additionally, the correlation between the SE based GS and MRF estimated T 1 and T 2 values (R 2 = 0.9969 and 0.9977) showed a similar trend as we observed between vendor‐provided and MRF estimated T 1 and T 2 values (R 2 = 0.9963 and 0.9972). In addition to correlation, BA analysis showed that all the vials are within the LOA between the GS and vendor‐provided for the T 1 values and except one vial for T 2 . All the vials are within the LOA between GS and MRF except one vial in T 1 and T 2 array. The wCV for reproducibility was <3% for both T 1 and T 2 values in vials #1‐8, for all the 14 vials, wCV calculated for reproducibility was <4% for T 1 values and <5% for T 2 . Conclusion This study shows that MRF is highly repeatable (wCV <4% for T 1 and <7% for T 2 ) and reproducible (wCV < 3% for both T 1 and T 2 ) in certain vials (vials #1‐8).