Metadata

Name

4D flow MRI data with varied isotropic voxel size and acceleration factor

Repository

ZENODO

Identifier

doi:10.5281/zenodo.4882572

Description

4D flow magnetic resonance imaging (MRI) can non-invasively quantify flow in blood vessels. However, the accuracy of 4D flow MRI can be impaired with partial volume effects. We aimed to assess 4D flow MRI precision in small vessels under limited spatial resolution in vitro at 3 T. Here we deposit the flow data acquired with 4D flow MRI accelerated up to 13 times with three isotropic voxel sizes (0.5, 1, and 1.5 mm). 4D flow MRI was applied to quantify flow in a set of silicone tubes and a patient-specific aneurysm model.

***************************************************************************
1_Flow_in_silicone_tubes
---------------------------------------------------------------------------------------------------------------
In this folder raw MRI data in .PAR/.REC format is located.
A subfolder name, such as "0.5_cs2.5", means "(isotropic voxel size of a 4D flow MRI)_cs(acceleration factor of the compressed sensing scheme implemented by Philips)". 
In our work, we used GTflow (Version 3.1.12, Gyrotools, Switzerland) to evaluate the data.
---------------------------------------------------------------------------------------------------------------
Bent silicone tubes with inner diameters (ID) of 2, 3, 4, and 5 mm were wound around a 50 ml falcon tube (Fig. 1 a). Thus, a model of curved neurovascular vessels was created. The flow was analyzed in regions perpendicular to the flow direction (regions of interest, ROI A-C). Tubes were supplied with the glycerol-water mixture (40/60 by volume) at a mean flow rate of 0.6, 1.2, 2.3, and 3.1 ml/s, respectively, to have a similar peak velocity of about 60 cm/s (PD-1100, BDC Laboratories, USA). The flow in the tubes was controlled by a transonic flow sensor (US sensor, ME8PXL-M12). The US sensor was located 10 cm downstream from the model outlets.

***************************************************************************
2_Velocity_in_an_aneurysm_model
---------------------------------------------------------------------------------------------------------------
In this folder 3D velocity field .raw format is located. Three binary files are included, containing the X, Y, and Z components of the velocity vector in each voxel. The value format is single-precision floating-point and represents velocity in [cm/s]. In addition, a text file with the extension "_Descr.txt" is located, it contains in text format the dimension sizes of the stored velocity field.
The velocity data can be open and processed with Matlab. An example of the Matlab-based program is provided.
A subfolder name, such as "0.5_cs2.5", means "(isotropic voxel size of a 4D flow MRI)_cs(acceleration factor of the compressed sensing scheme implemented by Philips)". 
The velocity field is generated with GTflow (Version 3.1.12, Gyrotools, Switzerland).
---------------------------------------------------------------------------------------------------------------
Patient-specific model of a 17.5-mm infraophthalmic extradural internal carotid artery (ICA) aneurysm of a female patient was produced in-house [24] using 3D printing technology (Form 3, Formlabs, USA), based on 3D rotational angiographic data from the patient (Fig. 1 b). The US sensor was located 10 cm downstream from the model inlet. The aneurysm flow model was supplied with tap water at a flow rate of 3.8 ml/s (Ismaltec MCP Standart, Cole Parmer, USA). The flow was monitored using a US sensor (ME6PXN325). Additionally, we measured the pressure using a Luer-lock pressure sensor (PRESS-N-000, PendoTech, USA) at the inlet and outlet of the model. The velocity maps were visualized at the 2D plane (Fig. 1 c, left). Velocity magnitude was calculated over the 3D aneurysm ROI (ROI, Fig. 1 c, right).

The tubes and aneurysm model were submerged in 3 % agarose gel before the MRI experiment.
***************************************************************************
magnetic resonance imaging
---------------------------------------------------------------------------------------------------------------
The imaging was performed using a whole state-of-the-art body 3 T MRI system using a clinical 32-channel volume head coil (Ingenia CX, R5 V6.1, Philips Healthcare, Best, Netherland). Three angiographic pulse sequences were used in the study: time-of-flight (TOF), 2D PC, and 4D flow MRI.
2D PC and 4D flow MRI are based on spoiled gradient-echo sequence with Cartesian sampling [1]. Both sequences were accelerated with a compressed‐sensing technique implemented by the vendor (Philips). Venc was set to 60 and 80 cm/cm for the experiment with silicon tubes and an aneurysm model, respectively. The cardiac cycle was equal to 800 and 816 ms, and 24 heart phases were obtained. The resulting temporal resolution was 33 – 34 ms.
2D PC MRI comprised following parameters: repetition time (TR)/echo time (TE) – 9.4/6.2 ms; field of view – 180 x 180 mm2; voxel size – 0.5 x 0.5 x 4 mm3; acceleration factor – 2.5. The 2D PC MRI acquisition time was 2 min. 
4D flow MR images were acquired using a balanced symmetric 4-point PC velocity-encoding scheme (Hadamard)[2] with three different isotropic voxel sizes: 0.5, 1.0, and 1.5 mm3. 
Silicon tubes were imaged by 4D flow MRI, where TR/TE we in range of (6.6 - 10)/(4 - 6.3); imaging volume – 110 x 78 x 30 mm3; acceleration factor – 2.5, 4.5, 6.5, and 13. The 4D flow MRI examination time was varied from 1.5 to 57.5 min.
Aneurysm flow model was imaged by 4D flow MRI, where TR/TE – (6.5 - 10.6)/(3.9 - 6.4); imaging volume – 100 x 100 x 20 mm3; acceleration factor – 2.5, 4.5, and 6.5. The 4D flow MRI examination time was varied from 3.2 to 73.2 min. Also, an aneurysm model was imaged with TOF MRI that comprised the following parameters: TR/TE – 25/5.8; imaging volume – 180 x 180 x 160 mm3; voxel size – 0.25 x 0.40 x 0.50; acceleration factor – 4.7; examination time – 20 min.
***************************************************************************
References
---------------------------------------------------------------------------------------------------------------
[1] Turski P, Scarano A, Hartman E, Clark Z, Schubert T, Rivera L, et al. Neurovascular 4DFlow MRI (Phase Contrast MRA): emerging clinical applications. Neurovascular Imaging. 2016 Apr 22;2(1):8.
[2] Pelc NJ, Bernstein MA, Shimakawa A, Glover GH. Encoding strategies for three-direction phase-contrast MR imaging of flow. Journal of Magnetic Resonance Imaging. 1991;1(4):405–13. 


 

Data or Study Types

multiple

Source Organization

Unknown

Access Conditions

available

Year

2021

Access Hyperlink

https://doi.org/10.5281/zenodo.4882572

Distributions

• Encoding Format: HTML ; URL: https://doi.org/10.5281/zenodo.4882572