0 avis
Investigation of apple parenchyma microstructure using MRI multi-exponential T2 relaxation and microporosity measurements
Archive ouverte
Edité par CCSD -
International audience. Introduction Quantitative Magnetic Resonance Imaging is an appropriate non-destructive tool to study fruit microstructure [1]. Recent developments [2] allow accessing spatially resolved multi-exponential relaxation of water protons in the intact fruit, providing insights on water status and distribution at the subcellular level. Microporosity distribution in fruit tissues can also be estimated by MRI [3], offering additional information about gas distribution. For the first time, MRI multi-exponential T2 and apparent microporosity were used to investigate the effects of fruit heterogeneity and cell size for different apple cultivars. The MRI data were related to histological measurements. The general objective was to provide a multi-scale approach for apple tissue characterization. Material & Methods Experiments were performed 2-weeks after harvest on Granny Smith, Ariane, Fuji apples and an experimental cultivar (EC). Apples were calibrated as 3 groups: c1, about 120 cm3, c2 about xx cm3 and c3, about 290 cm3. MRI measurements were carried out on a 1.5T MRI scanner (Avanto, Siemens). The 5 mm median planes of fruit were imaged with a pixel size = 1.19 mm² and a TR = 10 s. T2 was obtained from a 512-echoes MSE sequence with ΔTE = 7.1 ms. T2*, used with T2 for microporosity estimation, was obtained from GE sequences with TE1 = 2.77 ms and ΔTE = 1.61 ms. Cell size distribution was estimated on same samples from macrovision images using an erosion/dilation method [4]. Results Apple T2 decay was well fitted by a tri-exponential decay curve with T2 about 15-26 ms, 85-135 ms and 360-550 ms and associated relative intensities about 10, 20 and 70 % respectively. The study of heterogeneity of parenchyma revealed that from outer (near the cuticle) to inner (near the core) regions, relaxation times decreased up to 20% and subcellular water distribution was slightly different. In contrast, microporosity was shown to increase of up to 28 % (see fig. 1). T2 measurements of Fuji apple s of calibers c1 and c3 showed variations only for the longest vacuolar T2 component from 423 ± 25 ms to 469 ± 20 ms, respectively. Cell size distributions, obtained by macrovision, showed larger cells in big apples thus linking cell size to T2 relaxation time. Microporosity increased from 33 ± 3% for small fruits to 39 ± 2 % for big apples. Differences between cultivars of the same caliber were observed. For example, EC had higher porosity and lower relaxation times and Granny Smith had a less heterogeneous parenchyma than other cultivars. Size-contrasted fruits permitted to highlight cell size influence on multi-T2 relaxation time. In apple cortex, T2 relaxation time increased with distance to the core even if apparent microporosity increased. This work showed that MRI can differentiate apple cultivars according to their T2 and microporosity spatial distribution. References [1] Van As, H. and J. van Duynhoven, Journal of Magnetic Resonance, 2013. 229: p. 25-34.. [2] Adriaensen H, et al, Magnetic Resonance Imaging, 2013. [3] Musse, M., et al., Magnetic Resonance Imaging, 2010. 28(10): p. 1525-34 [4] Devaux, M.-F., et al., Postharvest Biology and Technology, 2008. 47(2): p. 199-209
Consulter en ligne
Suggestions
Du même auteur
