Y-net: 3D intracranial artery segmentation using a convolutional autoencoder
This work addresses segmentation for cerebrovascular analysis, which is incremental as it applies an existing autoencoder approach to a specific medical imaging task.
The authors tackled automated 3D segmentation of intracranial arteries from MRA data using an optimized convolutional autoencoder called Y-net, achieving better performance than three traditional methods in binary classification and visual evaluation.
Automated segmentation of intracranial arteries on magnetic resonance angiography (MRA) allows for quantification of cerebrovascular features, which provides tools for understanding aging and pathophysiological adaptations of the cerebrovascular system. Using a convolutional autoencoder (CAE) for segmentation is promising as it takes advantage of the autoencoder structure in effective noise reduction and feature extraction by representing high dimensional information with low dimensional latent variables. In this report, an optimized CAE model (Y-net) was trained to learn a 3D segmentation model of intracranial arteries from 49 cases of MRA data. The trained model was shown to perform better than the three traditional segmentation methods in both binary classification and visual evaluation.