Week #5

Done

• Accomplish both the tutorial and the programing assignments of Course 3: Structuring Machine Learning Projects of deeplearning.ai in Coursera.

• Paper review of:

  • Abhijit Guha Roy et al., “QuickNAT: A Fully Convolutional Network for Quick and Accurate Segmentation of Neuroanatomy,” ArXiv:1801.04161 [Cs], January 12, 2018. link

  • Hongzhi Wang and Paul A. Yushkevich, “Multi-Atlas Segmentation with Joint Label Fusion and Corrective Learning—an Open Source Implementation,” Frontiers in Neuroinformatics 7 (2013). link

  • Christian Wachinger, Martin Reuter, and Tassilo Klein, “DeepNAT: Deep Convolutional Neural Network for Segmenting Neuroanatomy,” NeuroImage 170 (April 2018): 434–45. link

  • Jonathan Long, Evan Shelhamer, and Trevor Darrell, “Fully Convolutional Networks for Semantic Segmentation,” ArXiv:1411.4038 [Cs], November 14, 2014. link

  • Olaf Ronneberger, Philipp Fischer, and Thomas Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation,” ArXiv:1505.04597 [Cs], May 18, 2015. link

  • J. Dolz, C. Desrosiers, and I. Ben Ayed, “3D Fully Convolutional Networks for Subcortical Segmentation in MRI: A Large-Scale Study,” NeuroImage 170 (April 2018): 456–70. Link

• Freesurfer tutorial (40% finished)

To do

• Build a first convNet in PyTorch.
• Be familiar with Pytorch

Ideas

• Can we imitate the QuickNAT by using the output labeled data of QuickNAT/PICSL as the training data for our network.