In vitro Assessment of Aortic Regurgitation Using Four-Dimensional Flow Magnetic Resonance Imaging

The characteristics of valvular diseases can be identified through this protocol, which otherwise is difficult to through in vivo diagnostic study. Further, the in vitro evaluation of the 4D flow MRI is demonstrated in this protocol. This technique can measure the time-resolved 3D velocity field of the in vitro heart valve model.

This includes analysis of the flow rate and stroke volume retrospectively. 4D flow MRI measurement provides the results with DICOM medical format images. However, understanding these medical images and transferring images to physical flow data may be difficult for beginners.

Due to the limited access of MRI for general researchers, awareness of 4D flow MRI measurement is limited in many fields. Visual demonstration of this protocol would increase its application. To begin, determine the parameter values of the aortic root such as valve-based diameter and sinus radius.

Run the three-dimensional modeling software by clicking sketch, then go to tools, sketch, tools and go click on sketch picture. Sketch circles corresponding to R maximum and R minimum using the circle tool to make a sinus mode. Draw a curved line of the sinus using the free curve function, click loft tool and select the sketch area for loft.

Sketch additional circles on the top and bottom of the current model, click extrude tool and select the circles. Set the options as 20 millimeters downward and 30 millimeters upward. Make a hexahedron model in the same way.

From the insert menu, go to features, select combine and click on combine tool. Select subtract in the property manager. Select the hexahedron model and the sinus model.

Fabricate the final design as an acrylic model with a five-axis CNC machine per the manufacturer's instructions. Run 3D modeling software and open a new sketch. Manually draw a square and a circle in the center of the valve base.

Click the extrude tool and adjust the height of the valve base to five millimeters. Extrude the circle with a height of 23.5 millimeters and a thickness of three millimeters. Divide the model into 12 uniform pieces using line tools so that each piece has 30 degrees.

Select three pieces with 120 degrees intervals and extrude with a height of 16.5 millimeters to make three pillars. Click fillet tool and select the pillars. Adjust the fillet radius at the top and bottom as four millimeters and 10 millimeters respectively.

Save it in an STL file format. 3D print the valve frame by setting infill density to 100%and using acrylonitrile butadiene styrene as film material. Run the 3D modeling software and open a new sketch.

Draw a horizontal line of 23 millimeters and a vertical line of 15 millimeters. Click three-point ARC tool from the ARC command manager. Set two points on each end of the horizontal line and the last point on the end of the vertical line and extrude the sketch with a thickness of five millimeters.

Export the model with STL file format and print it. Overlap the ePTFE membrane in two layers. Draw the leaflet borders at intervals of two millimeters using the printed leaflet.

Suture along the drawn lines and side borders at one millimeter intervals with a polyamide suture of 0.1 millimeter diameter. Suture the ePTFE valve from top to bottom on the frame at one millimeter intervals. Cut the outer side of the membrane and suture them with each other.

Perform modifications for three different models. For the dilation model, reduce the ratio of the designated leaflet parameters to 90%Make a circular hole of two millimeters diameter using scissors in the center of one leaflet for the perforation model. For prolapse, fix the two commissures of the valve at a hole with a low post height.

Prepare the experimental system consisting of aortic models, a heart simulation pump, and MRI. Set the experiment models in the MRI room and connect the pump, reservoir, and models using the silicone tube with 25 millimeters inner diameter. Use a 10 centimeter long cable and fasten the connection parts to prevent any leakage.

Locate the model within the field of view of the MRI. Perform a scout scan to observe phantom images in the coronal, axial, and sagittal views in MRI operating console monitor. Locate the two-dimensional image plane in the center of the aorta model.

Run a variable velocity encoding parameter 2D phase contrast imaging to select the most appropriate velocity encoding value for 4D flow MRI. Set VENC to a 10%higher value in 4D flow MRI. Enter the desired spatial resolution and the temporal resolution on the MRI console.

For aortic flow, these values are two to three millimeters and 20 to 40 milliseconds and acquired data for both with and without flow using the three types of AR valves and without valve. Copy raw data files from the scanner to analyze the data. Sort the DICOM files according to the header named series description using the DICOM sort software.

Click sort images in DICOM sort software to sort three-directional phase images and magnitude images in separate folders. Load magnitude image into the ITK-SNAP software. Click brush in the ITK-SNAP and manually paint the internal fluid region of the phantom using the brush tool.

Save segmented image. Optionally, load both phase image data obtained with the flow on and off using MATLAB. Subtract the data with the flow by the data without the flow to remove background errors.

Repeat this for every direction and cardiac cycle. Calculate the velocity of 5D matrix phase data using a vendor-specific pixel-to-velocity equation. Load previously obtained 5D matrix velocity into flow visualization analysis software.

Click the isosurface part and change the data type for 3D analysis by clicking the isovolume button. Drag the speed data in the variables command manager and add it to the isovolume to check the velocity distribution of the model. Click particle trace emitters tool in the main menu.

Check advanced option for a more accurate analysis. Select the desired visualization, such as streamlines or path lines in creation. Set the values for the experiment.

Create and check the results over time. Right-click the particle trace model and click the color by. Select the velocity component to color the streamline with the velocity.

Load the velocity data and segmented image previously obtained onto MATLAB. Set the velocity outside of the segmentation region to zero by multiplying element wise the segmented matrix and the velocity matrix data. Check if the velocity data has phase wrapping using the image show function of MATLAB.

Inversion of the velocity direction indicates phase wrapping. Slice the desired plane of the matrix data. Sum all velocity data within the plane and multiply spatial resolution to calculate the flow rate through the plane.

Sum all flow rates throughout the cardiac cycle and multiply the temporal resolution to calculate the stroke volume. The figure shows the results of 4D flow MRI which streamlines normal and regurgitation jets during systole and diastole. It can be observed that without a valve, an overall forward and backward flow occurred.

The regurgitant jet of the dilation model came out from the center and tended to change directions over time. Also, the forward jet was straight in all models except for the perforation model. A wall-biased jet during the systole phase occurred in the perforation model.

Moreover, the perforation and prolapse model regurgitant jet leaned toward the wall. The figure shows the flow rate for each valve and the forward and regurgitant volumes in a 3D plane away from the valve base. The flow rates showed different waveforms and quantities for each model.

Generally, the positive percentage values indicate underestimation, while the negative percentage values represent overestimation. Following this protocol, researchers can fabricate various in vitro heart valves, including stenosis heart valves and regurgitation heart valves. Also, hemodynamics in these valves can be investigated.

This technique explored in vitro fabrication of diseased heart valves and 4D flow MRI demonstrations.