MR image reconstruction using deep learning: evaluation of network structure and loss functions
Quantitative Imaging in Medicine and Surgery2019Vol. 9(9), pp. 1516–1527
Citations Over TimeTop 10% of 2019 papers
Vahid Ghodrati, Jiaxin Shao, Mark Bydder, Ziwu Zhou, Wotao Yin, Kim‐Lien Nguyen, Yingli Yang, Peng Hu
Abstract
CNN image reconstruction using Resnet yields comparable image quality to Unet with 10X the number of parameters. This has implications for training with significantly lower data requirements. Network training using the perceptual loss function was found to better agree with radiologist scoring compared to L1, L2 or Dssim loss functions.
Related Papers
- → Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning(2017)4,484 cited
- → Depth-Specific Variational Scaling Method to Improve Accuracy of ResNet(2021)2 cited
- → Inception-v4, Inception-ResNet and the Impact of Residual Connections on\n Learning(2016)199 cited
- Residual Attention Network for Brain Tumour Classification(2019)