Asynchronous magnetic resonance elastography: Shear wave speed reconstruction using noise correlation of incoherent waves

Abstract

Purpose The noninvasive measurement of biological tissue elasticity is an evolving technology that enables the robust characterization of soft tissue mechanics for a wide array of biomedical engineering and clinical applications. We propose, design, and implement here a new MRI technique termed asynchronous magnetic resonance elastography (aMRE) that pushes the measurement technology toward a driverless implementation. This technique can be added to clinical MRI scanners without any additional specialized hardware. Theory Asynchronous MRE is founded on the theory of diffuse wavefields and noise correlation previously developed in ultrasound to reconstruct shear wave speeds using seemingly incoherent wavefields. Unlike conventional elastography methods that solve an inverse problem, aMRE directly reconstructs a pixel‐wise mapping of wave speed using the spatial–temporal statistics of the measured wavefield. Methods Incoherent finger tapping served as the wave‐generating source for all aMRE measurements. Asynchronous MRE was performed on a phantom using a Siemens Prismafit as an experimental validation of the theory. It was further performed on thigh muscles as a proof‐of‐concept implementation of in vivo imaging using a Siemens Skyra scanner. Results Numerical and phantom experiments show an accurate reconstruction of wave speeds from seemingly noisy wavefields. The proof‐of‐concept thigh experiments also show that the aMRE protocol can reconstruct a pixel‐wise mapping of wave speeds. Conclusion Asynchronous MRE is shown to accurately reconstruct shear wave speeds in phantom experiments and remains at the proof‐of‐concept stage for in vivo imaging. After further validation and improvements, it has the potential to lower both the technical and monetary barriers of entry to measuring tissue elasticity.