New dimensions in ultrasound contrast imaging - Visualization of myocardial blood flow

Aims are: 1. To find new bubble excitation strategies for improved contrast/tissue ratio of the ultrasound image 2. To perform ultrasound pulse field and contrast bubble response simulations 3. To peruse corresponding in vitro experiments 4. To deliver the contrast optimized pulse sequences for implementation in echocardiographs (for clinical studies)

• Staff:

Birgitta Janerot Sjöberg , MD, PhD, Docent		Main supervisor		IMH Clin Physiol/IMT LiU
Per Ask , Prof		Co-supervisor,		IMT, LiU
Tomas Jansson , PhD, Docent		Co-supervisor, Ultrasound Physisist		Elmath, LTH
Marcus Ressner , MSc,		PhD-student		IMT, LiU

• Former Staff:

• Project Description:

External expertise involved in the project

Lars-Åke Brodin, MD, prof. Biomedical engineering; mecical images, KI/KTH

Lars Hoff, assoc prof, PhD, Ultrasound contrast, Norway

Arunas Lukosevicius, prof. Ultrasound Physics, Kaunas techn Univ

Sigmund Friegstad, responsible Contrast imagin, GE Healthcare; U/S Vingmed cardiovascular ultrasound, Trondheim, Norway

Background/Methods:
Our long term goal of ultrasound contrast research is to find a new model driven approach for estimation of myocardial perfusion. We work on new bubble excitation strategies for improved contrast/tissue ratio of the backscattered echo and to enhance the blood flow information by a “bubble fingerprint” strategy. Further, our method will also include a strategy of gentle bubble destruction as well as contrast detection at low acoustic power to avoid tissue damage related to cavitation.  Our model driven approach is performed by simulating the linear and non-linear wave propagation in tissue and the resulting echo from the pulse/bubble interaction.. Results from simulation are compared to in vitro experiments. The myocardial perfusion method will in the next step include strategies for visualization of the results in 3 and 4 dimensions and the developed strategies will be clinically evaluated in patients and compared to gold standard methods.
   We have in Linköping built an in vitro ultrasound laboratory with specially designed ultrasound probes, high performance arbitrary waveform generator, transmit and receive amplifiers and a high speed data acquisition system. Tissue mimicking phantoms are designed containing contrast agents either fixed in gel-based solutions or moving in thin fibers. The simulation of the contrast bubble and the ultrasound field is performed in cooperation with Lars Hoff (reknown for his contrast bubble model) and guest researchers from Kaunas University in Lituania specialized on ultrasound physics. The field simulation model "Superposition Attenuation Waves" has been modified to implement ultrasound fields from spherical focused transducers, attenuation, diffraction and non-linear wave propagation in order to simulate the ultrasound fields from the transducers used in our in vitro model.
    The propagation of ultrasound pulses in tissue and their interaction with gas-encapsulated bubbles is a complex physical process. The simulation of the bubble responce has been performed using the Rayleigh-Plesset equations with the addition of radiation damping developed by Hoff. Experimental pulses sampled with a hydrophone in an in vitro setup have been used as input parameters in the bubble response simulations. Some of our results presenting "Modelling of nonlinear effects and the response of ultrasound contrast micro bubbles: simulation and experiment" by us was published in Ultrasonics (Kvikliene A, Jurkonis R, Ressner M, Hoff L, Jansson T, Janerot-Sjoberg B, Lukosevicius A, Ask P. Modelling of nonlinear effects and the response of ultrasound contrast micro bubbles: simulation and experiment. Ultrasonics. 2004 Apr;42(1-9):301-7)
Results
   Effects of pulse length and pulse polarity of transmitted waveforms on the response of ultrasound contrast micro bubbles was investigated using computer simulation and invitro experiments. Waveforms from a clinical ultrasound system (GE Vivid 7) have been sampled in vitro with a hydrophone (Precision Acoustics) and used as input for simulations of the bubble response. Simulations of the bubble response show that the magnitude of the second harmonic will increase with increased pulse length and that the magnitude differences between a long and a short pulse is decreased with increased acoustic power. Further, the effects of pulse polarity on the magnitude of the second harmonic show a complex relation to both frequency and pulse length. The magnitude differences of the second harmonic will generally increase with decreased pulse length but the extent of the magnitude difference will vary with frequency and number of cycles transmitted. Finally, the spectral centroid or first moment of the frequency spectrum is an indicator for the amount of the backscattered signal. The difference of the centroid due to pulse polarity is most noticeable at pulse length below 2.5 cycles and will decrease with increase pulse length and is not affected by frequency in the range of 1.5-3.5 MHz.
   A purpose was to determine the effect of UCA on Tissue Doppler velocity estimation in vivo and in vitro. We performed echocardiography in 12 patients with ischemic heart disease before and immediately after a slow intravenous infusion of 2.7 ml Optison® using color myocardial Doppler imaging (GE Vingmed systemV) to examine the effect of contrast in vivo. Longitudinal basal systolic velocities and their integrals were analyzed in digitally stored cineloops. An in vitro study was carried out to exclude any biological or physiological factors affecting the velocity estimation and determine the effects of contrast concentration and acoustic output for the three different contrast agents Otison®, SonoVue® and Sonazoid®.
   In vivo results show that the peak mean velocity increased 10% during contrast infusion, from mean 5.2±1.8 (SD) to 5.7±2.3 cm/s, (p=0.02, confidence interval 2- 16%). The longitudinal basal systolic velocity integral did not change (0.8±0.4 cm). Contrast had no effect on blood pressure or heart rate in the used dose. In vitro results show a 5-40 % velocity increase in the interval of 4-8 cm/s compared to a tissue mimicking Agar/Graphite background. These effects are consistent with moderate variations of contrast concentration and will increase with higher acoustic power. It can be concluded that these findings can not be explained by biological or physiological related to the administration of UCA. Nor can it solely be explained by the increase of the Doppler signal due to contrast interaction as the increased velocity was noticeable even at low acoustic powers and reduced gain settings.
   The project is now performed also in cooperation with General Electric Vingmed and is also supported by Nimed center of excellence at LiU.

Publikationer:

Ressner M, Jansson T, Cedefamn J, Ask P, Janerot-Sjoberg B. Contrast biases the autocorrelation phase shift estimation in Doppler tissue imaging. Ultrasound Med Biol. 2009;35 (3):447-57. Epub 2009 Jan 18.

Ressner M, Brodin LA, Jansson T, Hoff L, Ask P, Janerot-Sjoberg B. Effects of ultrasound contrast agents on Doppler tissue velocity estimation. J Am Soc Echocardiogr. 2006;19 (2):154-64.