Segmentation of bone surface from ultrasound using a lightweight network UBS-Net

Abstract

Objective. Ultrasound-assisted orthopaedic navigation holds promise due to its non-ionizing feature, portability, low cost, and real-time performance. To facilitate the applications, it is critical to have accurate and real-time bone surface segmentation. Nevertheless, the imaging artifacts and low signal-to-noise ratios in the tomographical B-mode ultrasound (B US) images-create substantial challenges in bone surface detection. In this study, we present an end-to-end lightweight US bone segmentation network (UBS-Net) for bone surface detection.
Approach. We present an end-to-end lightweight UBS-Net for bone surface detection, using the U-Net structure as the base framework and a level set loss function for improved sensitivity to bone surface detectability. A dual attention (DA) mechanism is introduced at the end of the encoder, which considers both position and channel information to obtain the correlation between the position and channel dimensions of the feature map, where axial attention (AA) replaces the traditional self-attention (SA) mechanism in the position attention module for better computational efficiency. The position attention and channel attention (CA) are combined with a two-class fusion module for the DA map. The decoding module finally completes the bone surface detection.
Main Results. As a result, a frame rate of 21 frames per second (fps) in detection were achieved. It satisfied the real time requirement. The segmentation quality from the proposed approach outperformed the state-of-the-art method with higher accuracy (Dice similarity coefficient: 88.76% vs. 87.22%) in 612 retrospective testing images.
Significance. The proposed UBS-Net for bone surface detection in ultrasound achieved outstanding accuracy and real-time performance. The new method out-performed the state-of-the-art methods. It had potential in US-guided orthopaedic surgery applications.