This repository contains part of the work we conduct at LIVIA that can be made publicly available. The main focus on our research to segment medical images is on deep learning models and optimization techniques.
This is the current content of this repository:
- LiviaNET. 3D fully Convolutional Neural Network for semantic image segmentation Link
This repository contains the code of LiviaNET, a 3D fully convolutional neural network that was employed in our work: 3D fully convolutional networks for subcortical segmentation in MRI: A large-scale study Neuroimage, April,17th 2017.
- SemiDenseNet. Semi-dense connectivity to segment infant brain tissue in multi-modal images Link
This repository contains the code of the network that we employed in the iSEG Grand MICCAI Challenge 2017, infant brain segmentation. This network extends out previous work in 3D fully convolutional neural network that was employed in our work: 3D fully convolutional networks for subcortical segmentation in MRI: A large-scale study.
This repository will contain the code of HyperDense-Net, a hyper-densely connected network that we proposed to segment medical images in a multi-modal images scenario. It achieved state-of-the-art performance in some multi-modal image based segmentations. HyperDense-Net: A densely connected CNN for multi-modal image segmentation.