Skip to main content
To KTH's start page To KTH's start page

Exploring Polymer-Shelled Microbubbles: Detection Modeling and Application

Time: Fri 2020-12-11 09.00

Location:, (English)

Subject area: Technology and Health

Doctoral student: Hongjian Chen , Medicinsk avbildning

Opponent: Professor Tomas Jansson, Department of Biomedical Engineering, Lund University

Supervisor: Associate Professor Dmitry Grishenkov, ; Professor Birgitta Janerot-Sjöberg, KI, CLINTEC

Export to calendar


Ultrasound imaging (US) is widely used in clinical practice. Given the low cost and easy access to the ultrasound machine, US has a great potential to improve the health care condition for the majority of the population in the world. The US could be significantly improved by injecting ultrasound contrast agents to opacify the bloodstream. The polymer-shelled microbubbles (MB) are promising candidates for the next generation ultrasound contrast agent. In the current doctoral work, one of the polymer-shelled MBs, the polyvinyl alcohol (PVA) MB was investigated.

In Study I and Study II, I developed a novel contrast pulse sequence, CPS4, to efficiently detect the PVA MBs. The CPS4 is a combination of the sub-harmonic (SH), ultra-harmonic, and pulse inversion techniques. The comparison of the performance of each individual technique and CPS4 was carried out in a tissue-mimicking phantom. The CPS4 demonstrated the highest contrast-to-tissue ratio among all four imaging techniques. However, the SH response of the CPS4 was not fully excited. The high SH pressure threshold, above which the SH response is generated, was suspected to be the reason for the weak SH signal. Therefore, I wanted to optimize the performance of the CPS4 for the PVA MBs detection by boosting the SH signal. The optimization strategy was to lower the frequency-dependent SH threshold by setting the SH excitation frequency, which is the frequency of the ultrasound wave that excites the SH response, at the damped resonance frequency of the PVA MBs. To estimate the damped resonance frequency, a mathematical model based on the Church’s model with frequency-dependent material properties was proposed. The mechanical parameters of the new model were estimated by fitting the measured attenuation coefficient of the PVA MBs suspension with the simulated one. The calibrated model was employed to predict the damped resonance frequency of the PVA MBs, i.e., the optimized SH excitation frequency for the CPS4. The performance of the CPS4 was evaluated in-vitro, driving the system at four SH excitation frequencies in the proximity of the damped resonance frequency of the PVA MBs suspension. The best performance was observed at the SH excitation frequency of 11.25 MHz, which is in line with the simulated damped resonance frequency of 10.85 MHz. The in vitro experiment also revealed that the small particles constituting the artificial blood solution might interact with the PVA MBs and decreased the response echoes in a nonlinear and frequency-dependent fashion. Thus, more efforts are needed to move our model-guided optimization methods for the CPS4 towards clinical application.

In Study III, I modified the PVA MBs to support the dual-modal imaging of CT and US. The main idea of the modification is to incorporate the gold nanoparticles with the PVA MBs. The success of the modification is dependent on the amount of the gold nanoparticles carried by the modified PVA MBs. Two routes were proposed to fabricate candidates that support dual-modal imaging. In the first route, the gold nanoparticles were added during the fabrication of PVA MBs. Thus, the gold nanoparticles were embedded in the PVA shell during its formation (candidate named AuNP-S-MB). In the second route, the gold nanoparticles were loaded into the core of the PVA MBs, substituting air by increasing the permeability (candidate named AuNP-Capsule). The CT revealed an insignificant amount of gold nanoparticles was embedded in the shell of AuNP-S-MB, while detectable gold nanoparticles were loaded into AuNP-Capsule. Moreover, the CT-number of the surrounding liquid of AuNP-Capsule is low, i.e., the gold nanoparticles were locked in the AuNP-Capsule, making the second route a promising step towards the further development of the dual-modal contrast agent for CT and US.

In Study IV, I studied the effect of PVA MBs on the cavitation flows in microscale. The cavitation in clinical practices generates great pressure, which might be harmful and damage cells or beneficial and facilitate the treatment. A better understanding of cavitation generation mechanisms could avoid harmful cavitation, increase the safety of the clinical protocol, and increase the therapeutic cavitation, empower the treatments. Therefore, the effect of PVA MBs on cavitation is of great interest. More specifically, the effect of PVA MBs on the hydrodynamic cavitation was studied. Three microfluidic devices with different wall roughness and structure were fabricated. Two working fluids, PVA MBs suspension and water, were driven with controlled pressure through different microfluidic devices. The high-speed visualization revealed that the PVA MBs trigger the inception of hydrodynamic cavitation at a lower upstream pressure and enhance the cavitation flow in all three microfluidic devices. Furthermore, it takes a longer time for the cavitation bubbles to disappear in the PVA MB suspension.

To conclude the doctoral work, I developed a novel detection sequence, CPS4, optimized it for PVA MBs with a model-guided method, modified the PVA MB to extend its application, and studied the effect of PVA MB on hydrodynamic cavitation. The work promotes the PVA MBs for pre-clinical study, as well as provides an insight into the studies of other clinically approved ultrasound contrast agents. The methodology developed and presented within the thesis can be transferred to other clinically approved ultrasound contrast agents. For instance, the CPS4 and model-guided optimization method could be employed to improve CPS4 to other ultrasound contrast agents.