Key result
ADSeg improves true lumen segmentation accuracy to ~91% in Stanford type B aortic dissection.
Why the study?
Accurate and rapid lumen segmentation in aortic dissection is critical, but existing methods generally neglect the intimal flap structure separating true and false lumens.
Does the flap-attention-based deep learning framework (ADSeg) improve the accuracy and efficiency of automated true and false lumen segmentation on CTA images compared to existing deep learning methods in patients with Stanford type B aortic dissection?
Observational (n=108)
Yes
Does the flap-attention-based deep learning framework (ADSeg) improve the accuracy and efficiency of automated true and false lumen segmentation on CTA images compared to existing deep learning methods in patients with Stanford type B aortic dissection?
ADSeg provides rapid (0.53 seconds per case), highly accurate, and center-robust automated CTA segmentation of the true and false lumens in Stanford type B aortic dissection by incorporating intimal flap attention and long-distance z-axis interactions, facilitating timely surgical planning and 3D modeling.
Answers are grounded in this paper and cite their sources.
May aid automated CTA segmentation for type B dissection planning; leaves open prospective clinical validation.
Accurate and rapid segmentation of the lumen in an aortic dissection (AD) is an important prerequisite for risk evaluation and medical planning for patients with this serious condition. Although some recent studies have pioneered technical advances for the challenging AD segmentation task, they generally neglect the intimal flap structure that separates the true and false lumens. Identification and segmentation of the intimal flap may simplify AD segmentation, and the incorporation of long-distance z axis information interaction along the curved aorta may improve segmentation accuracy. This study proposes a flap attention module that focuses on key flap voxels and performs operations with long-distance attention. In addition, a pragmatic cascaded network structure with feature reuse and a two-step training strategy are presented to fully exploit network representation power. The proposed ADSeg method was evaluated on a multicenter dataset of 108 cases, with or without thrombus; ADSeg outperformed previous state-of-the-art methods by a significant margin and was robust against center variation.
No takes yet. Share an insight, caveat, or question.
Xiang et al. (2023) conducted an observational in Stanford type B aortic dissection (n=108). ADSeg (flap-attention-based deep learning segmentation framework) vs. Previous state-of-the-art deep learning segmentation methods (including Chen et al., Cao et al., Lyu et al., and traditional 3D U-Net) was evaluated on Dice coefficient score for true lumen (TL) segmentation on CTA (MD 4.1%). ADSeg improved true lumen and false lumen segmentation Dice coefficients to 91.1% and 88.4% in Stanford type B aortic dissection.
Synapse has enriched 10 related papers. Consider them for comparative context: