Final Report Summary - ICOMATH (Integrated Computational Model Framework for the Study of Atherosclerosis)
Executive Summary:
The project ICOMATH aimed at providing a better understanding of the pathobiology of atherosclerosis, so that a more accurate detection of potential sites of plaque formation could be achieved at early stages of the disease. At the same time, more efficient and individualized patient monitoring and treatment could be sought. Towards these objectives, research based on the acoustic radiation force of ultrasound and shear-based elastography was conducted. Such technology has been relatively recently appeared in the literature and has shown great potential in assessing the viscoelastic behavior of arterial walls and distinguishing between normal and calcified arteries in a non-invasive manner.
Therefore, the initial phase of the project (1st reporting period) was focused on studying novel and more efficient methods for the estimation of the mechanical properties of viscoelastic media, by using the shear waves generated by the acoustic radiation force of ultrasound. Such methods could be directly applied to arterial-wall models, which are known to be viscoelastic deformable tissues. Specifically, a theoretical/computational model was developed in order to study and optimize the radiation-force profile generated by two confocal ultrasound beams of nearly equal frequencies. It was shown that by appropriately selecting their geometry, as well as, other source parameters, the spatial profile of the force can be improved, leading to an optimized shear and longitudinal response in terms of spatial resolution. This, subsequently, can significantly improve the estimated local viscoelastic properties of the propagating medium (e.g. arterial wall).
A second study during the first reporting period aimed at estimating the shear dispersion and in turn, the local shear modulus and viscosity based on the ridges of a time-frequency transformation. This method was shown to be more efficient, accurate and robust to noise than the conventional phase-delay Fourier method in estimating the local viscoelastic properties.
During the second reporting period, emphasis was given on the study of a more realistic and clinically-relevant model of the shear-wave propagation, aiming again at improving the estimated viscoelastic properties. For this reason, a three-dimensional (3-D) finite-element-method (FEM) model was developed and studied for the propagation of shear waves in viscoelastic media containing spherical inclusions. The Voigt model of viscoelasticity was considered and higher source pressures were also assumed, such that the generated shear waves could propagate and be detected further away from the source. An inverse approach was, subsequently, used to extract 3-D maps of the local shear modulus and viscosity, based on both the fundamental and higher harmonics of the radiation force. It was shown that the shear modulus can be successfully reconstructed based on both the fundamental and the second harmonic component and that the shear viscosity can be more accurately extracted based on the second-harmonic shear component. A more efficient frequency-based algorithm was proposed to deal with the estimation of the viscoelastic properties under conditions of increased noise, since the inverse algorithm is known to be vulnerable to noise. The above three generic studies are valid both in one-layer and multilayer arterial-wall models. In this manner, the shear displacement field in healthy and diseased arteries can be computed and finally, more accurate estimates of their viscosity and elasticity can be extracted.
Another study performed toward the end of the project aimed at the characterization of potential atherosclerotic plaques, using the results of the aforementioned FEM shear-propagation model and a stochastic image processing model based on mixtures of Gaussian distributions. In this manner, successful classification of the type of atherosclerotic plaques and their components was possible, thereby, facilitating the prediction of their evolution more efficiently. Finally, a study on intravascular photoacoustic imaging was performed during Dr Giannnoula’s short visit at the Medical Optics group of the Institute of Photonic Sciences (ICFO), Barcelona, Spain. Together with the research in charge (Dr. Aguirre), they showed that photoacoustic imaging, an alternative novel and very powerful medical modality known to work with a combination of ultrasound and light, can be used to image with high resolution within the arterial walls and assess atherosclerotic plaques with respect to their morphology and composition.
In addition, the project has had significant impact on Dr Giannoula’s integration into the Greek and European society and scientific community. She has had the opportunity to meet, discuss, transfer her expertise and at the same time expand her research interests by exchanging ideas with various scientists in Greece, as well as, in other countries of the European Commission. Specifically, her collaboration with ICFO, as described above, and with several other institutes in Barcelona, Spain, has helped in making her work visible in the country where her husband has recently moved (Spain) and has already led to several job opportunities that will allow her to continue her research career.