Medical Image Segmentation via Unsupervised Convolutional Neural Network
Preprint
Abstract
For the majority of the learning-based segmentation methods, a large quantity of high-quality training data is required. In this paper, we present a novel learning-based segmentation model that could be trained semi- or un- supervised. Specifically, in the unsupervised setting, we parameterize the Active contour without edges (ACWE) framework via a convolutional neural network (ConvNet), and optimize the parameters of the ConvNet using a self-supervised method. In another setting (semi-supervised), the auxiliary segmentation ground truth is used during training. We show that the method provides fast and high-quality bone segmentation in the context of single-photon emission computed tomography (SPECT) image.Keywords
All Related Versions
This publication has 8 references indexed in Scilit:
- Incorporating CT prior information in the robust fuzzy C-means algorithm for QSPECT image segmentation2019
- Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification2015
- Recurrent convolutional neural network for object recognition2015
- Population of anatomically variable 4D XCAT adult phantoms for imaging research and optimizationMedical Physics, 2013
- 4D XCAT phantom for multimodality imaging researchMedical Physics, 2010
- Active contours without edgesIEEE Transactions on Image Processing, 2001
- An SVD investigation of modeling scatter in multiple energy windows for improved SPECT imagesIEEE Transactions on Nuclear Science, 1996
- A practical method for incorporating scatter in a projector-backprojector for accurate scatter compensation in SPECTIEEE Transactions on Nuclear Science, 1993