Deep Learning with Anatomical Priors: Imitating Enhanced Autoencoders in Latent Space for Improved Pelvic Bone Segmentation in MRI
This work addresses a domain-specific problem in medical imaging for pelvic bone segmentation, but it appears incremental as it builds on existing U-Net and autoencoder methods.
The authors tackled pelvic bone segmentation in MRI by proposing a 2D encoder-decoder architecture that incorporates anatomical priors through imitating an enhanced autoencoder in latent space, showing promising improvements compared to a standard U-Net.
We propose a 2D Encoder-Decoder based deep learning architecture for semantic segmentation, that incorporates anatomical priors by imitating the encoder component of an autoencoder in latent space. The autoencoder is additionally enhanced by means of hierarchical features, extracted by an U-Net module. Our suggested architecture is trained in an end-to-end manner and is evaluated on the example of pelvic bone segmentation in MRI. A comparison to the standard U-Net architecture shows promising improvements.