Medical imaging techniques such as ultrasound, X-ray, computed tomography (CT), and magnetic resonance imaging (MRI) are integral screening modalities in medicine due to their ability to non-invasively visualise internal organs. Elastography is an emerging imaging modality that complements conventional imaging by mapping the mechanical (elastic) properties of soft tissues alongside anatomical information. Often described as the analogue of palpation for internal organs, elastography works by applying controlled mechanical vibrations to tissue and measuring the resulting deformation or wave propagation. This response is used to infer biomechanical tissue properties, which are displayed as quantitative images for clinical interpretation. Such information is clinically valuable, as many pathological conditions—directly or indirectly—alter tissue biomechanics.
Several elastography techniques have been developed, differing primarily in the imaging modality used to observe tissue response, with ultrasound- and MRI-based approaches being the most prominent. Magnetic Resonance Elastography (MRE) extends conventional MRI by enabling non-invasive measurement of soft tissue stiffness through the generation and imaging of low-frequency shear waves. MRE provides quantitative stiffness maps that have proven valuable in the assessment of diseases such as solid tumours, liver fibrosis, and neurological disorders. Importantly, unlike ultrasound elastography, MRE is not limited by air or bone, allowing reliable assessment of deep organs, including the brain.
The BioMagnus project aims to develop an advanced in silico tool for non-invasive reconstruction of soft tissue stiffness using Magnetic Resonance Elastography. By integrating a high-fidelity numerical solver for viscoelastic wave propagation combined with physics-informed deep learning for the inversion process of tissue stiffness reconstruction, the project will enable fast and accurate estimation of tissue biomechanical properties. The proposed methodology will address key limitations of existing MRE reconstruction techniques by accounting for realistic tissue mechanics, complex geometries, and enhanced spatial resolution. The in silico tool developed will be validated using brain MRE data provided by the partner institute Charité in Berlin, and has strong potential to enhance the diagnosis and staging of solid tumours and neurological diseases across multiple organs.
This BioMagnus project has received the Seal of Excellence from the MSCA Postdoctoral Fellowships 2024 programme and is funded by the RESTART 2016–2020 Programmes for Research, Technological Development and Innovation of Cyprus’ Research & Innovation Foundation.
Related publications:
Lilaj et al. (2021). Magnetic Resonance in Medicine, https://doi.org/10.1002/mrm.28898
Galarce et al. (2023). SIAM Journal on Imaging Sciences, https://doi.org/10.1137/22M149363X Hiscox et al. (2021). NeuroImage, https://doi.org/10.1016/j.neuroimage.2021.117889
