Premium
Initial validation of equilibrium contrast imaging for extracellular volume quantification using a three‐dimensional engineered tissue model
Author(s) -
Bandula Steve,
Magdeldin Tarig,
Stevens Nicola,
Yeung Jason,
Moon James C.,
Taylor Stuart A.,
Cheema Umber,
Punwani Shonit
Publication year - 2016
Publication title -
journal of magnetic resonance imaging
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.563
H-Index - 160
eISSN - 1522-2586
pISSN - 1053-1807
DOI - 10.1002/jmri.25066
Subject(s) - extracellular fluid , relaxometry , biomedical engineering , linear regression , limits of agreement , magnetic resonance imaging , nuclear medicine , contrast (vision) , gadolinium , medicine , extracellular , chemistry , mathematics , radiology , computer science , statistics , biochemistry , organic chemistry , spin echo , artificial intelligence
Background To test the principles underpinning equilibrium contrast imaging estimation of tissue extracellular volume (ECV) fraction, using a three‐dimensional (3D) engineered tissue model with known cellular and extracellular volumes. Methods Six 3D tissue models (tumoroids) consisting of cell cultures within a collagen containing hydrogel were constructed after culture centrifugation and direct measurement of the cell component volume. Measured tumoroid ECV ranged from 0.89 to 1. ECV was calculated after measuring the T1 relaxation time at 3 Tesla using inversion recovery relaxometry (TI 100–1500 ms) within the tumoroids and surrounding medium before and 375 min after spiking the medium with Gadolinium (to achieve a concentration of 1.4 mM/L). Linear regression model prediction of directly measured ECV (ECV m ) by EQ‐MRI measured ECV (ECV eq ); and Bland‐Altman agreement between measures was assessed. Results The fractional cellular volume measured by EQ‐MRI (ECV eq ) within the tumoroids ranged from 0.821 to 0.963. ECV eq was a good predictor of ECV m (R2 = 0.77, P = 0.02). The regression line Y‐axis intercept (when X = 0) was 0.045 ± 0.019 with a slope of 1.28 ± 0.35. Bland‐Altman comparison demonstrated 95% limits of agreement between ‐0.002 and 0.114 with a bias (SD) of 0.056 (0.03). Conclusion This study supports the principles of ECV estimation using equilibrium contrast MRI, but future development of this model may allow validation over a wider, more physiological ECV range and a greater understanding of the effect of tissue extracellular protein burden on ECV. J. Magn. Reson. Imaging 2016;43:1224–1229.