BME PhD Proposal Seminar: Manoj Menon

Resolution Estimation and Bias Reduction in Acoustic Radiation Force Impulse Imaging

Supervised by Professor Stephen McAleavey

Abstract

Acoustic Radiation Force Impulse (ARFI) imaging ultrasonically measures the micronscale displacements induced in tissue by local acoustic radiation forces using high intensity ultrasound pulses generated by a standard diagnostic ultrasound scanner. Generated displacement images can exhibit improved contrast of diseased tissue than conventional ultrasound modalities. In this thesis, novel simulation and experimental techniques are developed and used to quantify the spatial resolution limits of an ARFI imaging system. The full-width, half-maximum (FWHM) of the point-spread function (PSF), a measure of the resolution limit of an imaging system, was extracted by imaging a tissue-mimicking phantom composed of two bonded materials of differing modulus. The ARFI image of the material interface was an estimate of the step response of the system. Resolution was further explored using FEM/acoustic field simulations and linear shift invariant (LSI) models. Our ARFI imaging system demonstrated submillimeter resolution, which was highly dependent on imaging parameters. Axial resolution was limited by the correlation window length and tracking pulse parameters. Lateral resolution corresponded to the two-way beamwidth of the tracking beam. Measuring ARFI imaging resolution capabilities on small phantom inclusions and tissue ablation lesions demonstrated the validity of linear-shift invariant (LSI) ARFI imaging models, and the allowed comparison of step-response based estimated resolution limits to those obtained from objects of clinically relevant circular geometry. To improve the ability of ARFI imaging to resolve targets near bright boundaries, a method called envelope weighted normalization (EWN) was developed to reduce amplitude modulation of ultrasound signals, thereby reducing displacement estimation bias.