Abstract
Multi-frequency Electrical Impedance Tomography (mfEIT) is an emerging biomedical imaging modality to reveal frequency-dependent conductivity distributions in biomedical ap-plications. Conventional model-based image reconstruction meth-ods suffer from low spatial resolution, unconstrained frequency correlation and high computational cost. Deep learning has been extensively applied in solving the EIT inverse problem in biomed-ical and industrial process imaging. However, most existing learning-based approaches deal with the single-frequency setup, which is inefficient and ineffective when extended to the multi-frequency setup. This paper presents a Multiple Measurement Vector (MMV) model based learning algorithm named MMV-Net to solve the mfEIT image reconstruction problem. MMV-Net considers the correlations between mfEIT images and unfolds the update steps of the Alternating Direction Method of Multipliers for the MMV problem (MMV-ADMM). The non-linear shrinkage operator associated with the weighted l2,1 regularization term of MMV-ADMM is generalized in MMV-Net with a cascade of a Spatial Self-Attention module and a Convolutional Long Short-Term Memory (ConvLSTM) module to better capture intra- and inter-frequency dependencies. The proposed MMV-Net was validated on our Edinburgh mfEIT Dataset and a series of comprehensive experiments. The results show superior image quality, convergence performance, noise robustness and computational efficiency against the conventional MMV-ADMM and the state-of-the-art deep learning methods.
Original language | English |
---|---|
Pages (from-to) | 1-12 |
Number of pages | 12 |
Journal | IEEE Transactions on Neural Networks and Learning Systems |
Early online date | 9 Mar 2022 |
DOIs | |
Publication status | E-pub ahead of print - 9 Mar 2022 |
Keywords / Materials (for Non-textual outputs)
- Deep Learning
- Electrical impedance tomography (EIT)
- image reconstruction
- multifrequency
- multiple measurement vectors
- MMV
- EIT
- Correlation
- Conductivity
- multiple measurement vector (MMV)
- Frequency measurement
- electrical impedance tomography (EIT)
- Image reconstruction
- Deep learning
- Electrical impedance tomography
- Biomedical imaging
Fingerprint
Dive into the research topics of 'MMV-Net: A Multiple Measurement Vector Network for Multi-frequency Electrical Impedance Tomography'. Together they form a unique fingerprint.Datasets
-
Edinburgh mfEIT Dataset
Yang, Y. (Creator), Chen, Z. (Creator) & Bagnaninchi, P. (Creator), The University of Edinburgh, 26 May 2022
DOI: 10.7488/ds/3464
Dataset