OpenBreastUS: Benchmarking Neural Operators for Wave Imaging Using Breast Ultrasound Computed Tomography

2025-07-20

AI Summary

Traditional wave-equation solvers for ultrasound computed tomography (USCT) of the breast suffer from high computational cost and numerical instability, while existing neural operator methods exhibit poor generalizability due to oversimplified training data. Method: We introduce the first large-scale, anatomically realistic benchmark dataset for clinical USCT—comprising 8,000 high-fidelity breast phantoms and over 16 million frequency-domain wavefield simulations, rigorously calibrated to real USCT system configurations. Contribution/Results: Using this dataset, we systematically evaluate state-of-the-art neural operators on forward modeling and inverse imaging tasks, assessing accuracy, scalability, and generalization. Our evaluation demonstrates, for the first time, the feasibility of near-real-time reconstruction in *in vivo* breast imaging using neural PDE solvers—marking a significant step toward clinical deployment of physics-informed deep learning in complex biological tissues.

Technology Category

• Computer Vision: Medical and Biological Imaging
• Search and Optimization: Mixed Discrete/Continuous Search
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications

Abstract

Accurate and efficient simulation of wave equations is crucial in computational wave imaging applications, such as ultrasound computed tomography (USCT), which reconstructs tissue material properties from observed scattered waves. Traditional numerical solvers for wave equations are computationally intensive and often unstable, limiting their practical applications for quasi-real-time image reconstruction. Neural operators offer an innovative approach by accelerating PDE solving using neural networks; however, their effectiveness in realistic imaging is limited because existing datasets oversimplify real-world complexity. In this paper, we present OpenBreastUS, a large-scale wave equation dataset designed to bridge the gap between theoretical equations and practical imaging applications. OpenBreastUS includes 8,000 anatomically realistic human breast phantoms and over 16 million frequency-domain wave simulations using real USCT configurations. It enables a comprehensive benchmarking of popular neural operators for both forward simulation and inverse imaging tasks, allowing analysis of their performance, scalability, and generalization capabilities. By offering a realistic and extensive dataset, OpenBreastUS not only serves as a platform for developing innovative neural PDE solvers but also facilitates their deployment in real-world medical imaging problems. For the first time, we demonstrate efficient in vivo imaging of the human breast using neural operator solvers.

Problem

• Accurate and efficient wave equation simulation for USCT imaging.
• Overcoming computational intensity and instability of traditional solvers.
• Bridging gap between neural operators and real-world imaging complexity.

Innovation

• Neural operators accelerate wave equation solving
• Large-scale dataset bridges theory and practice
• Enables efficient in vivo breast imaging