Multi-View Vertebra Localization and Identification from CT Images

TL;DR

This paper introduces a multi-view 2D approach for vertebra localization and identification in CT images, leveraging global information and contrastive learning to improve accuracy and efficiency over traditional 3D methods.

Contribution

It proposes converting 3D vertebra localization into multi-view 2D tasks, utilizing contrastive pre-training and a sequence loss for better structural understanding.

Findings

Outperforms state-of-the-art methods in accuracy

Uses only two 2D networks for localization and identification

Reduces computational costs compared to 3D approaches

Abstract

Accurately localizing and identifying vertebrae from CT images is crucial for various clinical applications. However, most existing efforts are performed on 3D with cropping patch operation, suffering from the large computation costs and limited global information. In this paper, we propose a multi-view vertebra localization and identification from CT images, converting the 3D problem into a 2D localization and identification task on different views. Without the limitation of the 3D cropped patch, our method can learn the multi-view global information naturally. Moreover, to better capture the anatomical structure information from different view perspectives, a multi-view contrastive learning strategy is developed to pre-train the backbone. Additionally, we further propose a Sequence Loss to maintain the sequential structure embedded along the vertebrae. Evaluation results demonstrate that, with only two 2D networks, our method can localize and identify vertebrae in CT images accurately, and outperforms the state-of-the-art methods consistently. Our code is available at https://github.com/ShanghaiTech-IMPACT/Multi-View-Vertebra-Localization-and-Identification-from-CT-Images.