Learned Correction Methods for Ultrasound Computed Tomography Imaging Using Simplified Physics Models

TL;DR

This paper systematically compares learning-based correction methods for ultrasound computed tomography using simplified physics models, highlighting the importance of physics incorporation and domain-specific correction strategies for improved image quality and task performance.

Contribution

It introduces a comprehensive assessment of data-driven and model-based learning approaches for USCT reconstruction with simplified physics models, revealing their strengths and limitations.

Findings

Correction in measurement domain yields accurate images with fewer artifacts.

Correction in image domain is more robust to noise but prone to hallucinations.

Combining both corrections improves image metrics but may reduce task accuracy.

Abstract

Ultrasound computed tomography (USCT) is an emerging modality for breast imaging. Image reconstruction methods that incorporate accurate wave physics produce high resolution quantitative images of acoustic properties but are computationally expensive. The use of a simplified linear model in reconstruction reduces computational expense at the cost of reduced accuracy. This work aims to systematically compare different learning approaches for USCT reconstruction utilizing simplified linear models. This work considered various learning approaches to compensate for errors stemming from a linearized wave propagation model: correction in the data and image domains. The resulting image reconstruction methods are systematically assessed, alongside data-driven and model-based methods, in four virtual imaging studies utilizing anatomically realistic numerical phantoms. Image quality was assessed utilizing relative root mean square error (RRMSE), structural similarity index measure (SSIM), and a task-based assessment for tumor detection. Correction in the measurement domain resulted in images with minor visual artifacts and highly accurate task performance. Correction in the image domain demonstrated a heavy bias on training data, resulting in hallucinations, but greater robustness to measurement noise. Combining both forms of correction performed best in terms of RRMSE and SSIM, at the cost of task performance. This work systematically assessed learned reconstruction methods incorporating an approximated physical model for USCT imaging. Results demonstrated the importance of incorporating physics, compared to data-driven methods. Learning a correction in the data domain led to better task performance and robust out-of-distribution generalization compared to correction in the image domain.