DefCor-Net: Physics-aware ultrasound deformation correction

Zhongliang Jiang; Yue Zhou; Dongliang Cao; Nassir Navab

doi:10.1016/j.media.2023.102923

DefCor-Net: Physics-aware ultrasound deformation correction

Med Image Anal. 2023 Dec:90:102923. doi: 10.1016/j.media.2023.102923. Epub 2023 Aug 18.

Authors

Zhongliang Jiang¹, Yue Zhou², Dongliang Cao³, Nassir Navab⁴

Affiliations

¹ Computer Aided Medical Procedures, Technical University of Munich, Munich, Germany. Electronic address: zl.jiang@tum.de.
² Computer Aided Medical Procedures, Technical University of Munich, Munich, Germany.
³ University of Bonn, North Rhine-Westphalia, Germany.
⁴ Computer Aided Medical Procedures, Technical University of Munich, Munich, Germany; Computer Aided Medical Procedures, Johns Hopkins University, Baltimore, MD, USA.

PMID: 37688982
DOI: 10.1016/j.media.2023.102923

Abstract

The recovery of morphologically accurate anatomical images from deformed ones is challenging in ultrasound (US) image acquisition, but crucial to accurate and consistent diagnosis, particularly in the emerging field of computer-assisted diagnosis. This article presents a novel physics-aware deformation correction approach based on a coarse-to-fine, multi-scale deep neural network (DefCor-Net). To achieve pixel-wise performance, DefCor-Net incorporates biomedical knowledge by estimating pixel-wise stiffness online using a U-shaped feature extractor. The deformation field is then computed using polynomial regression by integrating the measured force applied by the US probe. Based on real-time estimation of pixel-by-pixel tissue properties, the learning-based approach enables the potential for anatomy-aware deformation correction. To demonstrate the effectiveness of the proposed DefCor-Net, images recorded at multiple locations on forearms and upper arms of six volunteers are used to train and validate DefCor-Net. The results demonstrate that DefCor-Net can significantly improve the accuracy of deformation correction to recover the original geometry (Dice Coefficient: from 14.3±20.9 to 82.6±12.1 when the force is 6N). Code:https://github.com/KarolineZhy/DefCorNet.

Keywords: AI for medicine; Anatomy-aware ultrasound imaging; Dense displacement field estimation; Force sensing; Medical image analysis; Physics-aware ultrasound imaging; Robotic ultrasound; Stiffness estimation in ultrasound imaging; Ultrasound elastography; Ultrasound image analysis; Ultrasound image deformation correction.