A physical phantom for amine chemical exchange saturation transfer (CEST) MRI - PubMed (original) (raw)

. 2021 Aug;34(4):569-580.

doi: 10.1007/s10334-020-00902-z. Epub 2021 Jan 23.

Jingwen Yao 1 2 3, Catalina Raymond 1 2, Blake Bergstrom 3, Xing Chen 3, Kaveri Das 3 4, Huy Dinh 3, Zoe S Kim 3, Angela N Le 3, Matthew W J Lim 3, Jane A N Pham 3, Joseph D Prusan 3, Sriram S Rao 3, David A Nathanson 5, Benjamin M Ellingson 6 7 8

Affiliations

A physical phantom for amine chemical exchange saturation transfer (CEST) MRI

Jingwen Yao et al. MAGMA. 2021 Aug.

Abstract

Objective: To develop a robust amine chemical exchange saturation transfer (CEST) physical phantom, validate the temporal stability, and create a supporting software for automatic image processing and quality assurance.

Materials and methods: The phantom was designed as an assembled laser-cut acrylic rack and 18 vials of phantom solutions, prepared with different pHs, glycine concentrations, and gadolinium concentrations. We evaluated glycine concentrations using ultraviolet absorbance for 70 days and measured the pH, relaxation rates, and CEST contrast for 94 days after preparation. We used Spearman's correlation to determine if glycine degraded over time. Linear regression and Bland-Altman analysis were performed between baseline and follow-up measurements of pH and MRI properties.

Results: No degradation of glycine was observed (p > 0.05). The pH and MRI measurements stayed stable for 3 months and showed high consistency across time points (R2 = 1.00 for pH, R1, R2, and CEST contrast), which was further validated by the Bland-Altman plots. Examples of automatically generated reports are provided.

Discussion: We designed a physical phantom for amine CEST-MRI, which is easy to assemble and transfer, holds 18 different solutions, and has excellent short-term chemical and MRI stability. We believe this robust phantom will facilitate the development of novel sequences and cross-scanners validations.

Keywords: Chemical exchange saturation transfer (CEST); MRI phantom; Physical phantom; Temporal stability.

© 2021. European Society for Magnetic Resonance in Medicine and Biology (ESMRMB).

PubMed Disclaimer

Conflict of interest statement

Compliance with ethical standards

Conflict of interest Author B.M.E. is a paid consultant at companies Medicenna, MedQIA, Neosoma, Agios Pharmaceuticals, Siemens, Imaging Endpoints, Kazia/Novogen, and NW Biopharmaceuticals. He is a consultant at companies Oncoceutics, BeiGene, BBI, Tocagen, and non-profit organizations Global Coalition for Adaptive Research (GCAR) and NIH/NCI Cancer Imaging Steering Committee. Author B.M.E. also receives research grant from companies Siemens, Janssen Pharmaceuticals, VBL, and non-profit organizations National Brain Tumor Society and ACS. Author D.A.N. is a co-founder and consultant at company Katmai Pharmaceuticals and a co-founder of company Trethera Corporation. He is also a shareholder in Sofie Biosciences. Authors J.Y., C.R.G., X.C., A.N.L., H.D., C.W., K.D., M.W.J.L, B.B., Z.S.K., J.A.N.P., J.D.P., S.S.R. has no conflict of interest to disclose.

Figures

Fig.1

Fig.1

The physical phantom rack design. The rack is composed of two platforms holding the solution vials (a), two platforms with Cartesian grids (b), and four stands. In platform a, the bigger circular holes were designed to hold bottles containing phantom solutions, smaller holes were for water filling and placing the platforms with fingers. An “L” shape was carved out to help orienting the phantom and registering MR images. The components can be assembled through fit-in notches. The 3D rendering of the rack design and the actual photo of an assembled phantom are demonstrated in c and d, respectively

Fig.2

Fig.2

Sequence diagram of multi-slice multi-echo CEST-EPI sequence

Fig.3

Fig.3

Schematic diagram of automatic image-processing and quality assurance software. First, the T1 mapping, T2 mapping, and CEST images obtained from the scanner were processed to calculate parametric maps. The maps were then registered to high-resolution T1-weighted image. The entire dataset was subsequently registered to template data and the MRI measurements within each vial were extracted. The means and standard deviations of all phantom samples were calculated and used to generate tables and plots for the quality assurance report. The final report included scanner and sequence information obtained from DICOM image header, MRI measurement summary, assessment of registration, and detection of deviations from standard parametric values

Fig.4

Fig.4

Example images of phantom MRI scan. The middle axial, coronal, and sagittal slices of the T1-weighted image were demonstrated in the first row of a. The second row showed the coronal slices at the level of three platforms, as indicated by the three colored dash lines in the first row. b Same slices of T2-weighted image. One example coronal slice of R1 map, R2 map, MTRasym map at 3.0 ppm, B0 map, and relative B1 map is demonstrated in c

Fig.5

Fig.5

Relaxation and CEST properties of the phantom solutions. a, b plotted the relationship between relaxation properties and gadolinium concentration, under the same glycine concentration and pH (sample #1–#4). R1 and R2 showed a linear correlation with gadolinium concentration. The correlation between MTRasym and gadolinium concentration, T1, and T2 is plotted in c. MTRasym showed a linear correlation with T1 and T2. d MTRasym at 3.0 ppm from sample #3 and #5–#18 were plotted against pH, showing an increase in MTRasym with increasing glycine concentration and decreasing pH

Fig.6

Fig.6

The temporal stability of the chemical properties and MRI properties of phantom solutions. The stability of glycine concentration was evaluated using UV absorbance measurement at 210 nm wavelength. The measurement is calibrated right after preparation using diluted glycine solutions (a). The calibration parameters were then used to calculate the glycine concentrations at follow-up measurements, as plotted in b. c shows the pH measurement at preparation and at follow-up time points up to 94 days after preparation is plotted. The longitudinal measurements of MTRasym, R1, and R2 are plotted in d, e, and f, respectively

Fig.7

Fig.7

Linear regression and Bland–Altman plots of pH and MRI measurements. The pH measurements at follow-up time points are plotted against the pH at preparation (a). The longitudinal and baseline measurements showed excellent linear correlation, which is further validated by the Bland–Altman plot in b. The correlation plots and Bland–Altman plots of longitudinal and baseline MRI measurements, including R1, R2, and MTRasym, are demonstrated in c–h, which showed great consistency in all three measurements

References

    1. Vinogradov E, Sherry AD, Lenkinski RE (2013) CEST: from basic principles to applications, challenges and opportunities. J Magn Reson 229:155–172 -PMC -PubMed
    1. van Zijl PCM, Yadav NN (2011) Chemical exchange saturation transfer (CEST): what is in a name and what isn’t? Magn Reson Med 65(4):927–948 -PMC -PubMed
    1. Ling W, Regatte RR, Navon G, Jerschow A (2008) Assessment of glycosaminoglycan concentration in vivo by chemical exchange-dependent saturation transfer (gagCEST). P Natl Acad Sci USA 105(7):2266–2270 -PMC -PubMed
    1. Zaiss M, Anemone A, Goerke S, Longo DL, Herz K, Pohmann R, Aime S, Rivlin M, Navon G, Golay X, Scheffler K (2019) Quantification of hydroxyl exchange of D-Glucose at physiological conditions for optimization of glucoCEST MRI at 3, 7 and 9.4 Tesla. NMR Biomed 32(9):4113 -PMC -PubMed
    1. Nasrallah FA, Pages G, Kuchel PW, Golay X, Chuang KH (2013) Imaging brain deoxyglucose uptake and metabolism by glucoCEST MRI. J Cerebr Blood Flow Metab 33(8):1270–1278 -PMC -PubMed

MeSH terms

Substances

Grants and funding

LinkOut - more resources