Traditional velocity measurements with ultrasound are based on the Doppler principle, which states that sound emitted from a moving source or sound reflected from a moving target will lead to a shift in the frequency of the sound. This so-called Doppler shift can be measured directly from the received signal through a continuous wave ultrasound emission (CW-Doppler), or sampled through the emission of several ultrasound pulses (PW-Doppler).

Today, Doppler ultrasound measurement is an integral part of commercial scanner systems. Conventional blood flow imaging modalities include spectral Doppler, in which the complete spectrum of velocities within one specific region is displayed. Another modality estimates the mean velocity and direction of blood in many points in a distributed region, which is encoded as a parametric color image, displayed overlaid an image of the anatomy. This latter color flow imaging (CFI) modality has proven very useful for the detection of areas of abnormal blood flow, which can be investigated further using spectral Doppler techniques. In Figure 1, the operation of both CFI and spectral Doppler techniques are shown.

Figure 1: PW Doppler and Color Flow Imaging are established methods for blood flow imaging.

Some blood flow related student assignments

1. Model-based estimation of complex blood flow in congenital heart disease (fetus, neonates and children)

Cardiac flow patterns may reveal several kinds of cardiovascular disease. Well known examples include the detection and quantification of leaky heart valves and poor systolic and diastolic function. Conventional flow imaging with ultrasound is however limited to only measuring the velocity component along the ultrasound beam, i.e. it is a one-dimensional and angle-dependent measurement. This discrepancy limits the usefulness of Doppler ultrasound in diagnostic settings. In this work we will focus on further developing multi-dimensional flow velocity estimators based on speckle tracking, i.e. image pattern matching techniques. The main clinical application will be pediatric cardiology, with the aim to improve the depiction of complex flow patterns such as vortex and shunt flow.

The proposed multidimensional approaches proposed are however not as robust as conventional methods. Thus, the aims of this student project will be to further develop and optimize tracking algorithms within a robust framework based on the predicted motion of flow using a Kalman filter (model-based estimation).

Aims:

• Further develop and validate robust tracking algorithms that optimally weight measurement and modelling errors (model-based estimation)
• Test the proposed methods on simplified simulations as well as in vivo data from pediatric cardiology

Qualifications:

Knowledge of digital signal processing and Matlab or Python programming, GPU-programming (optional) 

Contact:

• Professor Lasse Løvstakken
• Postdoc Solveig Fadnes

2. Robust blood velocity estimation based on speckle tracking

We see a great potential for providing the medical doctors with more detailed information of blood patterns by estimating the full 2D/3D blood flow velocity vector, where the current methods relying on Doppler principles are only able to give a 1D velocity component. This will be used in many different medical applications such as vascular imaging and pediatric and adult cardiology. One relevant velocity estimator is the blood speckle tracking estimator which relies on an initial (fast) block matching procedure, and a refined subsample displacement estimator step. We would like to know more about this estimator and we want to make it more robust. We are looking for a candidate to 1) Evaluate the statistical properties of the current (simple) algorithms for speckle tracking, and to develop new and more robust speckle tracking algorithms based for instance on robust least squares estimator principles for the initial block matching, and optical flow principles for the subsequent subsample estimation. Also important, we need to provide an uncertainty map to be able to mask out the more uncertain measurements.

Preferred qualifications:

• Signal processing, estimation theory
• Programming in either Matlab / CUDA / Python / C++

Contact:

• Professor Lasse Løvstakken

3. Bedside computational fluid dynamics based on ultrasound imaging

There is increasing interest in using advanced computational models for blood flow based on computational fluid dynamics (CFD). By utilizing state-of-the-art GPU's it is possible to significantly speed up computations. Further, by accelerating recent mesh-less methods based on Smoothed Particle Hydrodynamics (SPH) and Lattice-Boltzmann approaches, one can more easily couple ultrasound measurements and simulations in an easier way. Our overall goal of this task is to be able to reconstruct cardiac flow based purely on measurements from ultrasound, including input from real-time segmentation tools and state-of-the-art blood flow measurements developed in our lab. To achieve high computational speed, some trade-offs are inferred, and a central task will be to investigate the right level of accuracy and speed. This assignment is a collaboration between the ultrasound group and the biomechanics group at NTNU.

Preferred qualifications:

• Programming in Python and C++

Contact:

• Professor Lasse Løvstakken

4. Navigated ultrasound imaging – 3-D reconstruction of (pulsatile) artery geometry and flow

Conventional ultrasound imaging of blood flow in central and peripheral arteries is today based on 2-D imaging, while pathology related to atherosclerosis is inherently three-dimensional. While real-time 3-D ultrasound is available for cardiac imaging, transducers for vascular imaging are not yet available. However, by utilizing highly accurate position sensors during scanning, it is possible to reconstruct the 3-D geometry of arteries based on multiple 2-D flow and B-mode images. In this project we will utilize recently installed navigation system based on optical and magnetic sensors to reconstruct 3-D flow in the carotid artery. This flow is highly pulsatile, and we will also incorporate information from ECG (electro-cardiogram) to also get timing information. The imaging approach will follow a recent plane-wave imaging scheme, where a high frame rate and high image quality can be achieved. Investigations will first be done using in vitro setup of known stationary and pulsatile flow. In vivo imaging in healthy volunteers will further be tries to show the potential of mapping arterial geometry and pulsatile 3-D flow patterns. 

Preferred qualifications:

• Programming in Matlab and C++

Contact:

• Professor Lasse Løvstakken
• Postdoc Daniel H. Iversen

5. Accelerating 2D blood flow imaging

In medical ultrasound imaging, blood velocity measurements are important for the diagnostics of cardiovascular disease. Conventional blood velocity imaging is limited as it only estimates the velocity component in the ultrasound beam direction. The ultrasound group at ISB is developing a new method for estimation of 2D blood flow patterns, producing a much more intuitive visualization of blood flow. However, ultrasound is a real-time imaging modality, and the current implementation of the method is not optimized and too slow for clinical use. 

Project aims

• Accelerating blood vector flow estimation by using GPUs, multiple CPUs or through modification of the algorithms.
• Real-time 2D vector flow imaging!

Preferred qualifications

• Knowledge of C/C++/Python and experience with GPU programming.

Figure: Blood flow imaging in the carotid bifurcation. These are the blood vessels that supply the brain (lower branch) and face (upper branch) with blood.

Contact:

• Postdoc Jørgen Avdal
• Postdoc Ingvild Kinn Ekroll

6. Dual probe 3D blood flow imaging

Conventional ultrasound imaging of blood flow in central and peripheral arteries is today based on 2D imaging, with 1D or 2D measurements of the blood velocity. However, only measurements in 3D are able to give accurate estimates of the true velocity magnitude. By utilizing two 2D probes, it may be possible to get 3D measurements of blood flow in an overlapping image plane, enabling more accurate velocity estimates and more intuitive visualization of blood flow in this plane.

Project aims:

• Investigate the feasibility of 3D velocity estimation using two linear probes!

Preferred qualifications:

• Knowledge of Matlab programming and digital signal processing.

Contact:

• Postdoc Jørgen Avdal
• Postdoc Ingvild Kinn Ekroll

7. Robust clutter filtering by image morphing

Cardiovascular diseases (CVDs) are disorders of the heart and blood vessels and include coronary heart disease, cerebrovascular disease, rheumatic heart disease and other conditions.CVDs are the number 1 cause of death globally: more people die annually from CVDs than from any other cause. An estimated 17.5 million people died from CVDs in 2012, representing 31% of all global deaths.

Most cardiovascular diseases can be prevented by early detection and management using counselling and medicines, as appropriate. Recent research shows that early identification of asymptomatic individuals can reduce mortality from myocardial infarction and stroke by 50%.

Ultrasound is the most widely used imaging modality to screen for CVD, since, contrary to MRI, it is a non-ionizing method. Ultrasound screening for CVDs often involves the application of Color Flow Imaging (CFI) a technique that makes it possible to visualize the blood stream velocity field. This is only possible thanks to process called clutter filtering which removes everything but the signal reflected by the blood.

Clutter filtering in the heart is a very challenging problem. The heart tissue moves with velocities which are similar to those of the blood. As a consequence the clutter filter is not able to remove all the tissue in the image or, even worse, removes some of the signal reflected by the blood.

Here at ISB, we have envisioned a technique that can potentially improve the robustness of conventional clutter filters. The technique involves the estimation of the tissue movement with a standard tissue doppler technique and image morphing.

Example of image morphing via the mesh warping: Recent Advances in Image Morphing

Aim

• Implement the technique and test its performance on in-silico, in-vitro and in-vivo data.

Profile

• If you like image processing and programing,
• if you want to be involved into research,
• if you like to boldly go where no one has gone before:

...this is your project

Requirements

• Background in Matlab or C++.

Contact

• Alfonso Rodriguez-Molares