Reference-Driven Undersampled MR Image Reconstruction Using Wavelet Sparsity-Constrained Deep Image Prior

Autor: Jincun Zheng, Binyi Qin, Feng Zhao, Di Zhao, Yanhu Huang
Rok vydání: 2021
Předmět:
Databases
Factual
Article Subject
Computer science
Computer applications to medicine. Medical informatics
Wavelet Analysis
R858-859.7
02 engineering and technology
Iterative reconstruction
Regularization (mathematics)
General Biochemistry
Genetics and Molecular Biology
030218 nuclear medicine & medical imaging
Image (mathematics)
03 medical and health sciences
Reference image
Deep Learning
0302 clinical medicine
Wavelet
Image Interpretation
Computer-Assisted
Prior probability
0202 electrical engineering
electronic engineering
information engineering
Humans
General Immunology and Microbiology
business.industry
Applied Mathematics
Deep learning
Brain
Computational Biology
020206 networking & telecommunications
Pattern recognition
General Medicine
Data Compression
Image Enhancement
Magnetic Resonance Imaging
Modeling and Simulation
Artificial intelligence
Mr images
business
Algorithms
Research Article
Zdroj: Computational and Mathematical Methods in Medicine, Vol 2021 (2021)
Computational and Mathematical Methods in Medicine
ISSN: 1748-6718
1748-670X
DOI: 10.1155/2021/8865582
Popis: Deep learning has shown potential in significantly improving performance for undersampled magnetic resonance (MR) image reconstruction. However, one challenge for the application of deep learning to clinical scenarios is the requirement of large, high-quality patient-based datasets for network training. In this paper, we propose a novel deep learning-based method for undersampled MR image reconstruction that does not require pre-training procedure and pre-training datasets. The proposed reference-driven method using wavelet sparsity-constrained deep image prior (RWS-DIP) is based on the DIP framework and thereby reduces the dependence on datasets. Moreover, RWS-DIP explores and introduces structure and sparsity priors into network learning to improve the efficiency of learning. By employing a high-resolution reference image as the network input, RWS-DIP incorporates structural information into network. RWS-DIP also uses the wavelet sparsity to further enrich the implicit regularization of traditional DIP by formulating the training of network parameters as a constrained optimization problem, which is solved using the alternating direction method of multipliers (ADMM) algorithm. Experiments on in vivo MR scans have demonstrated that the RWS-DIP method can reconstruct MR images more accurately and preserve features and textures from undersampled k -space measurements.
Databáze: OpenAIRE
Externí odkaz:
Nepřihlášeným uživatelům se plný text nezobrazuje