A multi-sound-path approach for modeling thermal-induced artifacts in ultrasound imaging

Ultrasound (US) imaging is essential for monitoring thermotherapies, but accurate simulation of heating-induced artifacts remains challenging. Conventional single-sound-path (SSP) models approximate thermal effects as a channel-independent axial shift, failing to reproduce thermo-acoustic lens (TAL) artifacts observed in thermotherapy experiments. This study proposes a physically realistic multi-sound-path (MSP) simulation framework that captures TAL artifacts arising from temperature-induced speed-of-sound inhomogeneity in array-based US imaging. Transient temperature fields are computed using an alternating direction implicit solver, and US images are reconstructed via delay-and-sum beamforming. Compared to SSP, the MSP model successfully reproduces characteristic TAL artifacts, including lateral resolution degradation (up to 52.0% increase in full width at half maximum) and speckle brightness reduction (up to 49.1%). When applied to speckle tracking and thermal strain imaging, MSP also reveals substantial decorrelation and displacement errors consistent with experimental observations. By providing a computationally efficient imaging simulation framework that avoids the prohibitive cost of full-wave solvers, the proposed framework is well-suited for large-scale parametric studies and evaluation of US-based thermometry and other speckle-based functional imaging methods.