Fusing 2D and 3D convolutional neural networks for the segmentation of aorta and coronary arteries from CT images
作者：
Highlights：
• We propose a two-stage strategy that retains the advantages of both 2D and 3D CNNs.
• The strategy is applied to segment the aorta and coronary arteries from CT images.
• The 3D CNN improves the inter-slice continuity and reduces the missed detection rate.
• This strategy can alleviate the class imbalance problem and reduce the training time.
• Extensive experiments are carried out to validate the efficacy of our method.
摘要
•We propose a two-stage strategy that retains the advantages of both 2D and 3D CNNs.•The strategy is applied to segment the aorta and coronary arteries from CT images.•The 3D CNN improves the inter-slice continuity and reduces the missed detection rate.•This strategy can alleviate the class imbalance problem and reduce the training time.•Extensive experiments are carried out to validate the efficacy of our method.
论文关键词：Convolutional neural networks,Human aorta and coronary arteries segmentation,2D and 3D network fusion,Medical images
论文评审过程：Received 1 February 2021, Revised 23 September 2021, Accepted 29 September 2021, Available online 7 October 2021, Version of Record 8 October 2021.
论文官网地址：https://doi.org/10.1016/j.artmed.2021.102189
