Multi-contrast reconstruction with Bayesian compressed sensing
Clinical imaging with structural MRI routinely relies on multiple acquisitions of the same region of interest under several different contrast preparations. This work presents a reconstruction algorithm based on Bayesian compressed sensing to jointly reconstruct a set of images from undersampled k-s...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Wiley Blackwell
2014
|
| Online Access: | http://hdl.handle.net/1721.1/85886 https://orcid.org/0000-0002-7637-2914 |
| _version_ | 1826207062795747328 |
|---|---|
| author | Bilgic, Berkin Adalsteinsson, Elfar Goyal, Vivek K. |
| author2 | Harvard University--MIT Division of Health Sciences and Technology |
| author_facet | Harvard University--MIT Division of Health Sciences and Technology Bilgic, Berkin Adalsteinsson, Elfar Goyal, Vivek K. |
| author_sort | Bilgic, Berkin |
| collection | MIT |
| description | Clinical imaging with structural MRI routinely relies on multiple acquisitions of the same region of interest under several different contrast preparations. This work presents a reconstruction algorithm based on Bayesian compressed sensing to jointly reconstruct a set of images from undersampled k-space data with higher fidelity than when the images are reconstructed either individually or jointly by a previously proposed algorithm, M-FOCUSS. The joint inference problem is formulated in a hierarchical Bayesian setting, wherein solving each of the inverse problems corresponds to finding the parameters (here, image gradient coefficients) associated with each of the images. The variance of image gradients across contrasts for a single volumetric spatial position is a single hyperparameter. All of the images from the same anatomical region, but with different contrast properties, contribute to the estimation of the hyperparameters, and once they are found, the k-space data belonging to each image are used independently to infer the image gradients. Thus, commonality of image spatial structure across contrasts is exploited without the problematic assumption of correlation across contrasts. Examples demonstrate improved reconstruction quality (up to a factor of 4 in root-mean-square error) compared with previous compressed sensing algorithms and show the benefit of joint inversion under a hierarchical Bayesian model. |
| first_indexed | 2024-09-23T13:43:37Z |
| format | Article |
| id | mit-1721.1/85886 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T13:43:37Z |
| publishDate | 2014 |
| publisher | Wiley Blackwell |
| record_format | dspace |
| spelling | mit-1721.1/858862022-09-28T15:45:13Z Multi-contrast reconstruction with Bayesian compressed sensing Bilgic, Berkin Adalsteinsson, Elfar Goyal, Vivek K. Harvard University--MIT Division of Health Sciences and Technology Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology. Research Laboratory of Electronics Bilgic, Berkin Goyal, Vivek K. Adalsteinsson, Elfar Clinical imaging with structural MRI routinely relies on multiple acquisitions of the same region of interest under several different contrast preparations. This work presents a reconstruction algorithm based on Bayesian compressed sensing to jointly reconstruct a set of images from undersampled k-space data with higher fidelity than when the images are reconstructed either individually or jointly by a previously proposed algorithm, M-FOCUSS. The joint inference problem is formulated in a hierarchical Bayesian setting, wherein solving each of the inverse problems corresponds to finding the parameters (here, image gradient coefficients) associated with each of the images. The variance of image gradients across contrasts for a single volumetric spatial position is a single hyperparameter. All of the images from the same anatomical region, but with different contrast properties, contribute to the estimation of the hyperparameters, and once they are found, the k-space data belonging to each image are used independently to infer the image gradients. Thus, commonality of image spatial structure across contrasts is exploited without the problematic assumption of correlation across contrasts. Examples demonstrate improved reconstruction quality (up to a factor of 4 in root-mean-square error) compared with previous compressed sensing algorithms and show the benefit of joint inversion under a hierarchical Bayesian model. 2014-03-21T19:07:08Z 2014-03-21T19:07:08Z 2011-06 2011-03 Article http://purl.org/eprint/type/JournalArticle 07403194 1522-2594 http://hdl.handle.net/1721.1/85886 Bilgic, Berkin, Vivek K Goyal, and Elfar Adalsteinsson. “Multi-Contrast Reconstruction with Bayesian Compressed Sensing.” Magnetic Resonance in Medicine 66, no. 6 (December 2011): 1601–1615. https://orcid.org/0000-0002-7637-2914 en_US http://dx.doi.org/10.1002/mrm.22956 Magnetic Resonance in Medicine Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Wiley Blackwell PMC |
| spellingShingle | Bilgic, Berkin Adalsteinsson, Elfar Goyal, Vivek K. Multi-contrast reconstruction with Bayesian compressed sensing |
| title | Multi-contrast reconstruction with Bayesian compressed sensing |
| title_full | Multi-contrast reconstruction with Bayesian compressed sensing |
| title_fullStr | Multi-contrast reconstruction with Bayesian compressed sensing |
| title_full_unstemmed | Multi-contrast reconstruction with Bayesian compressed sensing |
| title_short | Multi-contrast reconstruction with Bayesian compressed sensing |
| title_sort | multi contrast reconstruction with bayesian compressed sensing |
| url | http://hdl.handle.net/1721.1/85886 https://orcid.org/0000-0002-7637-2914 |
| work_keys_str_mv | AT bilgicberkin multicontrastreconstructionwithbayesiancompressedsensing AT adalsteinssonelfar multicontrastreconstructionwithbayesiancompressedsensing AT goyalvivekk multicontrastreconstructionwithbayesiancompressedsensing |