Key result
The direct segmentation approach achieved a Dice score of 52% for false lumen thrombus, significantly outperforming the multi-task approach but highlighting the difficulty of the ImageTBAD dataset.
Why the study?
Accurate segmentation of the true lumen, false lumen, and false lumen thrombus on CTA is crucial for TBAD quantification, but existing works only focus on the true and false lumens without considering false lumen thrombus.
Absolute Event Rate: 0.52% vs 0.44%
p-value: p=<0.05
The release of the ImageTBAD dataset provides the first 3D CTA dataset with annotations for true lumen, false lumen, and false lumen thrombus to facilitate research in automatic segmentation of Type-B Aortic Dissection.
Including FLT in TBAD CTA segmentation may refine prognosis; leaves open prospective validation before clinical adoption.
Type-B Aortic Dissection (TBAD) is one of the most serious cardiovascular events characterized by a growing yearly incidence,and the severity of disease prognosis. Currently, computed tomography angiography (CTA) has been widely adopted for the diagnosis and prognosis of TBAD. Accurate segmentation of true lumen (TL), false lumen (FL), and false lumen thrombus (FLT) in CTA are crucial for the precise quantification of anatomical features. However, existing works only focus on only TL and FL without considering FLT. In this paper, we propose ImageTBAD, the first 3D computed tomography angiography (CTA) image dataset of TBAD with annotation of TL, FL, and FLT. The proposed dataset contains 100 TBAD CTA images, which is of decent size compared with existing medical imaging datasets. As FLT can appear almost anywhere along the aorta with irregular shapes, segmentation of FLT presents a wide class of segmentation problems where targets exist in a variety of positions with irregular shapes. We further propose a baseline method for automatic segmentation of TBAD. Results show that the baseline method can achieve comparable results with existing works on aorta and TL segmentation. However, the segmentation accuracy of FLT is only 52%, which leaves large room for improvement and also shows the challenge of our dataset. To facilitate further research on this challenging problem, our dataset and codes are released to the public.
No takes yet. Share an insight, caveat, or question.
Yao et al. (2021) studied Type-B Aortic Dissection (n=100). Direct segmentation (Approach B) vs. Multi-task segmentation (Approach A) was evaluated on Dice score for False Lumen Thrombus (FLT) segmentation (S=96) (p=<0.05). The direct segmentation approach achieved a Dice score of 52% for false lumen thrombus, significantly outperforming the multi-task approach but highlighting the difficulty of the ImageTBAD dataset.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: