Multimodal imaging acquisitions were performed on each vine, from rootstocks to the beginning of branches, by Magnetic Resonance Imaging (MRI) and X-ray Computed Tomography (CT). MRI acquisitions were performed with Tridilogy, Groupe CRP—Imaneo (Tridilogy.com">http://www.Tridilogy.com) and the help of radiologists from CRP/Groupe Vidi at the Clinique du Parc (Castelnau-le-Lez, France), using a Siemens Magnetom Aera 1,5 Tesla and a human head antenna. Three acquisition sequences, T1-weighted(-w), T2-w, and PD-w were performed on each specimen, respectively:

3D T1 Space TSE Sagittal (Thickness 0.6 mm, DFOV 56.5 × 35 cm, 320 images, NEx 1, EC 1, FA 120, TR 500, TE 4.1, AQM 256/256).

3D T2 Space Sagittal (Thickness 0.9 mm, DFOV 57.4 × 35.5 cm, 160 images, NEx 2, EC 1, FA 160, TR 1100, TE 129, AQM 384/273).

Axial Proton Density Fat Sat TSE Dixon (Ep 5 mm, Sp 6.5, DFOV 57.2 × 38 cm, 40 images, NEx 1, EC 1, FA 160, TR 3370, TE 21, AQM 314/448).

X-ray CT acquisitions were performed at the Montpellier RIO Imaging platform (Montpellier, France, http://www.mri.cnrs.fr/en/) on an EasyTom 150 kV microtomograph (RX Solution). 3D volumes were reconstructed using XAct software (RX solution), resulting in approximately 2500 images per specimen at 177 µm/voxel resolution. Geometry, spot, and ring artifacts were corrected using the default correction settings when necessary.
Fernandez R., Le Cunff L., Mérigeaud S., Verdeil J.L., Perry J., Larignon P., Spilmont A.S., Chatelet P., Cardoso M., Goze-Bac C, & Moisy C. (2024). End-to-end multimodal 3D imaging and machine learning workflow for non-destructive phenotyping of grapevine trunk internal structure. Scientific Reports, 14, 5033.
+ Open protocol