Hybrid Graph Convolutional Neural Networks for Landmark-Based Anatomical Segmentation

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 Hybrid-Net produces accurate and anatomically plausible landmark-based segmentations, by naturally incorporating shape constraints within the decoding process via spectral convolutions.

Automated summary

The study introduces a novel neural architecture HybridGNet, which combines standard convolutions and graph convolutional neural networks for better decoding of anatomical structures and robustness to image occlusions, as well as construction of landmark-based segmentations from pixel level annotations.