Manuscript Due

Sept. 2, 2019 (closed)

Description

While signal processing and image recovery can achieve high-quality images from measured dataset with low SNR, medical image reconstruction can generate tomography image for disease diagnosis. In fact, there are various image features of medical images, such as piecewise constant, non-local similarity, low-rank, and so on. The regularization-based, multiscale transforms-based, Learning-based and machine learning-based (especially deep learning-based) methods are being actively developed worldwide for image reconstruction. Hence, in order to efficiently extract the useful information within medical image, advanced image features should be explored. Incorporating the medical image feature prior to formulating optimized models and deep-learning network can improve reconstructed image quality from measured datasets with low SNR. In this respect, this special issue titled as "Medical Image Reconstruction with Low SNR" can serve as a platform to preform better reconstruction results. To evaluate and validate the outperformances of the proposed algorithm, it is neccessary to compare quantitatively the reconstrution results in a systematic and reproducible strategy.


Lead Guest Editor

  • Yang Chen, Southeast University, China

Guest Editors

  • Jean-Claude Nunes, University of Rennes 1, France
  • Yudong Zhang, University of Leicester, UK
  • Weiwen Wu, Chongqing University, China
  • Hao Zhang, Stanford University, USA
  • Jiri Jan, Brno University of Technology, Czech
EURASIP Journal on Advances in Signal Processing