🤖 AI Summary
High-quality annotations are scarce for 3D segmentation of Type B Aortic Dissection (TBAD), and existing methods lack robustness across multiple anatomical structures—true lumen (TL), false lumen (FL), and flap (FLT).
Method: We propose a multi-output semi-supervised framework based on Slim UNETR, integrating multi-branch decoders, pseudo-labeling, and synergistic strong-weak data augmentation. Crucially, we introduce hypothesis-free probabilistic response consistency regularization—via rotation and flipping—for the first time in multi-output medical image segmentation, enabling end-to-end, post-processing-free semi-supervised training.
Results: On the ImageTBAD dataset, our method achieves Dice scores of 89.7% (TL), 84.3% (FL), and 76.5% (FLT) using only 30% labeled data—surpassing both fully supervised baselines and state-of-the-art semi-supervised approaches. This demonstrates the efficacy of non-probabilistic consistency modeling for multi-structure segmentation in TBAD.
📝 Abstract
Convolutional neural networks (CNN) for multi-class segmentation of medical images are widely used today. Especially models with multiple outputs that can separately predict segmentation classes (regions) without relying on a probabilistic formulation of the segmentation of regions. These models allow for more precise segmentation by tailoring the network's components to each class (region). They have a common encoder part of the architecture but branch out at the output layers, leading to improved accuracy.
These methods are used to diagnose type B aortic dissection (TBAD), which requires accurate segmentation of aortic structures based on the ImageTBDA dataset, which contains 100 3D computed tomography angiography (CTA) images. These images identify three key classes: true lumen (TL), false lumen (FL), and false lumen thrombus (FLT) of the aorta, which is critical for diagnosis and treatment decisions. In the dataset, 68 examples have a false lumen, while the remaining 32 do not, creating additional complexity for pathology detection.
However, implementing these CNN methods requires a large amount of high-quality labeled data. Obtaining accurate labels for the regions of interest can be an expensive and time-consuming process, particularly for 3D data. Semi-supervised learning methods allow models to be trained by using both labeled and unlabeled data, which is a promising approach for overcoming the challenge of obtaining accurate labels. However, these learning methods are not well understood for models with multiple outputs.
This paper presents a semi-supervised learning method for models with multiple outputs. The method is based on the additional rotations and flipping, and does not assume the probabilistic nature of the model's responses. This makes it a universal approach, which is especially important for architectures that involve separate segmentation.