Current browse context:
eess.IV
Change to browse by:
References & Citations
Electrical Engineering and Systems Science > Image and Video Processing
Title: Autoencoders for Multi-Label Prostate MR Segmentation
(Submitted on 9 Jun 2018 (v1), last revised 26 Jun 2018 (this version, v2))
Abstract: Organ image segmentation can be improved by implementing prior knowledge about the anatomy. One way of doing this is by training an autoencoder to learn a lowdimensional representation of the segmentation. In this paper, this is applied in multi-label prostate MR segmentation, with some positive results.
Submission history
From: Ard de Gelder [view email][v1] Sat, 9 Jun 2018 13:10:46 GMT (881kb)
[v2] Tue, 26 Jun 2018 11:16:36 GMT (881kb)
Link back to: arXiv, form interface, contact.