Featured Discovery

Home > Featured Discovery > SJTU Research Team Published Findings in IEEE Transactions on Medical Imaging

SJTU Research Team Published Findings in IEEE Transactions on Medical Imaging

March 11, 2021      Author:

Recent years have seen growing emphasis and wider application of the diagnosis and treatment of lung diseases via natural lumen for its non-invasive and flexible characteristics. Doctors would locate nidus based on the CT images ahead of the surgery and make plans accordingly. During the surgery, however, it is rather time consuming to reconstruct the 3D image of pulmonary airway and blood vein, and its accuracy is in question.

In order to solve the problem, research on computer-aided diagnosis, especially premedical planning based on deep learning, is gaining momentum, among which convolutional neural networks (CNNs) entails many technical difficulties in the segmentation of pulmonary airway and artery-vein due to sparse supervisory signals caused by the severe class imbalance between tubular targets and background. Hence, team led by Prof. Yang Guangzhong from Institute of Medical Robotics, SJTU put forward a CNNs-based method for accurate airway and artery-vein segmentation in non-contrast computed tomography, which enjoys superior sensitivity to tenuous peripheral bronchioles, arterioles, and venules.

The basic algorithm

The accuracy of pulmonary airway and artery-vein segmentation has been demonstrated by a large amount of research on CT images generated from real cases.

Findings of the research is compiled into a research article titled "Learning Tubule-Sensitive CNNs for Pulmonary Airway and Artery-Vein Segmentation in CT" and published on the website of IEEE Transactions on Medical Imaging. (doi: 10.1109/TMI.2021.3062280)

 

Author: Gu Yun

Source: School of Biomedical Engineering, SJTU

Translated by: Zhang Wenying

Proofread by: Xiao Yangning, Fu Yuhe

 

ABSTRACT:

Training convolutional neural networks (CNNs) for segmentation of pulmonary airway, artery, and vein is challenging due to sparse supervisory signals caused by the severe class imbalance between tubular targets and background. We present a CNNs-based method for accurate airway and artery-vein segmentation in non-contrast computed tomography. It enjoys superior sensitivity to tenuous peripheral bronchioles, arterioles, and venules. The method first uses a feature recalibration module to make the best use of features learned from the neural networks. Spatial information of features is properly integrated to retain relative priority of activated regions, which benefits the subsequent channel-wise recalibration. Then, attention distillation module is introduced to reinforce representation learning of tubular objects. Fine-grained details in high-resolution attention maps are passing down from one layer to its previous layer recursively to enrich context. Anatomy prior of lung context map and distance transform map is designed and incorporated for better artery-vein differentiation capacity. Extensive experiments demonstrated considerable performance gains brought by these components. Compared with state-of-the-art methods, our method extracted much more branches while maintaining competitive overall segmentation performance. Codes and models are available at http://www.pami.sjtu.edu.cn/News/56.