Hybrid graph convolutional neural networks for landmark-based anatomical segmentation
This addresses anatomical segmentation in medical imaging, particularly for chest x-rays, with incremental improvements in robustness.
The authors tackled landmark-based anatomical segmentation by proposing HybridGNet, an encoder-decoder architecture combining standard convolutions with graph convolutional networks, which was more robust to image occlusions in chest x-ray images and could construct landmark-based segmentations from pixel-level annotations.
In this work we address the problem of landmark-based segmentation for anatomical structures. We propose HybridGNet, an encoder-decoder neural architecture which combines standard convolutions for image feature encoding, with graph convolutional neural networks to decode plausible representations of anatomical structures. We benchmark the proposed architecture considering other standard landmark and pixel-based models for anatomical segmentation in chest x-ray images, and found that HybridGNet is more robust to image occlusions. We also show that it can be used to construct landmark-based segmentations from pixel level annotations. Our experimental results suggest that HybridGNet produces accurate and anatomically plausible landmark-based segmentations, by naturally incorporating shape constraints within the decoding process via spectral convolutions.