SA flow estimation project

Project work: Monday, May 29, 13.00-17.00; Tuesday, May 30, 08.00-12.00 and 13.00-17.00; Wednesday, May 31, 13.00-17.00 in the group rooms Building 349, Ground floor. Project presentation preparation is Thursday from 8:00-12:00, and the presentations are from 13:00-17:00.

Purpose

Interleaved synthetic aperture sequences can be employed for motion and flow imaging. Estimation of the blood velocity for both stationary and pulsating flow is implemented for an interleaved sequence with 12 emissions.

Preparation

1. Read the notes on SA imaging and the flow papers below.
2. Implement the cross-correlation estimator and employ it on the acquired data.
3. Acquiring data with the Verasonics scanner and use the GE 9L-D probe on the circulating flow rig.
4. Process the images for the acquired data and find the axial velocity.
5. Make a transverse oscillation in the frequency domain and implement the estimator to find the transverse and axial velocity components.
6. Are the velocities as expected?

Papers and reading material:

• J. A. Jensen, B. G. Tomov, L. E. Haslund, N. S. Panduro and C. M. Sørensen: Universal Synthetic Aperture Sequence for Anatomic, Functional and Super Resolution Imaging, Accepted for publication in IEEE UFFC, 2023
• J. A. Jensen: Estimation of High Velocities in Synthetic-Aperture Imaging—Part I: Theory, IEEE Transactions on Ultrasonics, Ferroelectrics and Frequency Control, Vol. 66, No. 6, pp. 1024-1031, June 2019
• J. A. Jensen: Estimation of High Velocities in Synthetic-Aperture Imaging—Part II: Experimental Investigation, IEEE Transactions on Ultrasonics, Ferroelectrics and Frequency Control, Vol. 66, No. 6, pp. 1032-1038, June 2019
• J. A. Jensen, S. Nikolov, K. L. Gammelmark and M. H. Pedersen, (2006). Synthetic Aperture Ultrasound Imaging. Ultrasonics, 44, e5-e15.
• S. Nikolov: Synthetic aperture tissue and flow ultrasound imaging, PhD dissertation, Technical University of Denmark, 2001.

Data:

A number of data sets can be found in the Dropbox under assignments/SA_flow/data/ for both fixed and pulsating flow. Data has been pre-beamformed and the individual images are stored in files named: Beamformed_HRIs_fr_00xxx_to_00yyy.mat for frames xxx to yyy.The file processing_parameters.mat hold variables for the processing. The data sets are:

1. flow_rig: constant flow in the flow rig
2. carotis_phantom_fixed: constant flow in a flow phantom
3. carotis_phantom_bifurcation: Pulsating flow in the bifurcation.

The data file processing_parameters.mat contains several variables and structures. The main variables are:

1. N_images: Number of images beamformed. In this case two
2. no_frames: Number of images in total
3. x_values: Values in the lateral x direction
4. z_values: Values in the axial z direction
5. fr: Frame rate of the images

A more detailed description of the beamforming is given in:

1. beamform_definitions: Definition of the beamforming
2. geometry: Geometry of the image
3. beamform_para: Beamforming parameters

The data is from the 12 x 2 emission sequence. So the array with data has dimensions of: Nz x Nx x 2 x Nfr, where Nz is the number of samples in depth, Nx is the number of image line, and Nfr is the number of frames loaded. The time between the two high resolution images is Tprf=1/(fr*24), because 24 emissions are emitted for the two images.

Dissemination:

Make a 10 minutes presentation of your project, which will be discussed on Thursday, June 1 from 13:00. There is 15 minutes available for each project group.