Simultaneous vessel segmentation and unenhanced prediction using self-supervised dual-task learning in 3D CTA (SVSUP). (September 2022)

• Record Type: Journal Article
• Authors: Huang, Wenjian; Gao, Weizheng; Hou, Chao; Zhang, Xiaodong; Wang, Xiaoying; Zhang, Jue
• Is Part Of: Computer methods and programs in biomedicine. Volume 224(2022)
• Publication Date: 2022-09
• Subjects: Head-neck CTA -- Dual-task learning -- All-vessel segmentation -- Unenhanced prediction -- Self-supervised learning -- Virtual unenhanced image; Medicine -- Computer programs -- Periodicals; Biology -- Computer programs -- Periodicals; Computers -- Periodicals; Medicine -- Periodicals; Médecine -- Logiciels -- Périodiques; Biologie -- Logiciels -- Périodiques; Biology -- Computer programs; Medicine -- Computer programs; Periodicals; Electronic journals; 610.28
• DOI: 10.1016/j.cmpb.2022.107001
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: Proposed an iterative residual-sharing scheme based dual-task learning framework for vessel segmentation and unenhanced CT prediction. Presented a pseudo-labeling based self-supervised strategy for vessel segmentation using contrasting modalities (pre/post-contrast imaging), which avoids labor-intensive manual labeling of training sets. Verified the feasibility of unenhanced CT prediction, which has the potential to eliminate the pre-contrast CT scan for radiation-dose reduction. The vessel segmentation results obtained by the proposed dual-task method are superior compared to popular 3D vessel segmentation models and deep-learning segmentation methods. Abstract: Background and objective: The vessel segmentation in CT angiography (CTA) provides an important basis for automatic diagnosis and hemodynamics analysis. Virtual unenhanced (VU) CT images obtained by dual-energy CT can assist clinical diagnosis and reduce radiation dose by obviating true unenhanced imaging (UECT). However, accurate segmentation of all vessels in the head-neck CTA (HNCTA) remains a challenge, and VU images are currently not available from conventional single-energy CT imaging. Methods:In this paper, we proposed a self-supervised dual-task deep learning strategy to fully automatically segment all vessels and predict unenhanced CT images from single-energy HNCTA based on a developed iterative residual-sharing scheme. The underlying idea was to use the correlation between the two tasks to improve task performance while avoiding manual annotation for model training. Results: The feasibility of the strategy was verified using the data of 24 patients. For vessel segmentation task, the proposed model achieves a significantly higher average Dice coefficient (84.83%, P -values 10 − 3 in paired t -test) than the state-of-the-art segmentation model, vanilla VNet (78.94%), and several popular 3D vessel segmentation models, including Hessian-matrix based filter (62.59%), optically-oriented flux (66.33%), spherical flux model (66.91%), and deep vessel net (66.47%). For the unenhanced prediction task, the average ROI-based error compared to the UECT in the artery tissue is 6.1 ± 4.5 HU, similar to previously reported 6.4 ± 5.1 HU for VU reconstruction. Conclusions: Results show that the proposed dual-task framework can effectively improve the accuracy of vessel segmentation in HNCTA, and it is feasible to predict the unenhanced image from single-energy CTA, providing a potential new approach for radiation dose saving. Moreover, to our best knowledge, this is the first reported annotation-free deep learning-based full-image vessel segmentation for HNCTA.