Enhancement of automated blood flow estimates (ENABLE) from arterial spin‐labeled MRI
<strong>Purpose</strong> To validate a multiparametric automated algorithm—ENhancement of Automated Blood fLow Estimates (ENABLE)—that identifies useful and poor arterial spin‐labeled (ASL) difference images in multiple postlabeling delay (PLD) acquisitions and thereby improve clinical A...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Journal article |
| Published: |
Wiley
2017
|
| _version_ | 1826291257137168384 |
|---|---|
| author | Shirzadi, Z Stefanovic, B Chappell, MA Ramirez, J Schwindt, G Masellis, M Black, SE MacIntosh, BJ |
| author_facet | Shirzadi, Z Stefanovic, B Chappell, MA Ramirez, J Schwindt, G Masellis, M Black, SE MacIntosh, BJ |
| author_sort | Shirzadi, Z |
| collection | OXFORD |
| description | <strong>Purpose</strong> To validate a multiparametric automated algorithm—ENhancement of Automated Blood fLow Estimates (ENABLE)—that identifies useful and poor arterial spin‐labeled (ASL) difference images in multiple postlabeling delay (PLD) acquisitions and thereby improve clinical ASL. <strong>Materials and Methods</strong> ENABLE is a sort/check algorithm that uses a linear combination of ASL quality features. ENABLE uses simulations to determine quality weighting factors based on an unconstrained nonlinear optimization. We acquired a set of 6‐PLD ASL images with 1.5T or 3.0T systems among 98 healthy elderly and adults with mild cognitive impairment or dementia. We contrasted signal‐to‐noise ratio (SNR) of cerebral blood flow (CBF) images obtained with ENABLE vs. conventional ASL analysis. In a subgroup, we validated our CBF estimates with single‐photon emission computed tomography (SPECT) CBF images. <strong>Results</strong> ENABLE produced significantly increased SNR compared to a conventional ASL analysis (Wilcoxon signed‐rank test, P < 0.0001). We also found the similarity between ASL and SPECT was greater when using ENABLE vs. conventional ASL analysis (n = 51, Wilcoxon signed‐rank test, P < 0.0001) and this similarity was strongly related to ASL SNR (t = 24, P < 0.0001). <strong>Conclusion</strong> These findings suggest that ENABLE improves CBF image quality from multiple PLD ASL in dementia cohorts at either 1.5T or 3.0T, achieved by multiparametric quality features that guided postprocessing of dementia ASL. |
| first_indexed | 2024-03-07T02:56:43Z |
| format | Journal article |
| id | oxford-uuid:af8a3446-6d65-42eb-a566-6216d781a679 |
| institution | University of Oxford |
| last_indexed | 2024-03-07T02:56:43Z |
| publishDate | 2017 |
| publisher | Wiley |
| record_format | dspace |
| spelling | oxford-uuid:af8a3446-6d65-42eb-a566-6216d781a6792022-03-27T03:50:15ZEnhancement of automated blood flow estimates (ENABLE) from arterial spin‐labeled MRIJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:af8a3446-6d65-42eb-a566-6216d781a679Symplectic Elements at OxfordWiley2017Shirzadi, ZStefanovic, BChappell, MARamirez, JSchwindt, GMasellis, MBlack, SEMacIntosh, BJ<strong>Purpose</strong> To validate a multiparametric automated algorithm—ENhancement of Automated Blood fLow Estimates (ENABLE)—that identifies useful and poor arterial spin‐labeled (ASL) difference images in multiple postlabeling delay (PLD) acquisitions and thereby improve clinical ASL. <strong>Materials and Methods</strong> ENABLE is a sort/check algorithm that uses a linear combination of ASL quality features. ENABLE uses simulations to determine quality weighting factors based on an unconstrained nonlinear optimization. We acquired a set of 6‐PLD ASL images with 1.5T or 3.0T systems among 98 healthy elderly and adults with mild cognitive impairment or dementia. We contrasted signal‐to‐noise ratio (SNR) of cerebral blood flow (CBF) images obtained with ENABLE vs. conventional ASL analysis. In a subgroup, we validated our CBF estimates with single‐photon emission computed tomography (SPECT) CBF images. <strong>Results</strong> ENABLE produced significantly increased SNR compared to a conventional ASL analysis (Wilcoxon signed‐rank test, P < 0.0001). We also found the similarity between ASL and SPECT was greater when using ENABLE vs. conventional ASL analysis (n = 51, Wilcoxon signed‐rank test, P < 0.0001) and this similarity was strongly related to ASL SNR (t = 24, P < 0.0001). <strong>Conclusion</strong> These findings suggest that ENABLE improves CBF image quality from multiple PLD ASL in dementia cohorts at either 1.5T or 3.0T, achieved by multiparametric quality features that guided postprocessing of dementia ASL. |
| spellingShingle | Shirzadi, Z Stefanovic, B Chappell, MA Ramirez, J Schwindt, G Masellis, M Black, SE MacIntosh, BJ Enhancement of automated blood flow estimates (ENABLE) from arterial spin‐labeled MRI |
| title | Enhancement of automated blood flow estimates (ENABLE) from arterial spin‐labeled MRI |
| title_full | Enhancement of automated blood flow estimates (ENABLE) from arterial spin‐labeled MRI |
| title_fullStr | Enhancement of automated blood flow estimates (ENABLE) from arterial spin‐labeled MRI |
| title_full_unstemmed | Enhancement of automated blood flow estimates (ENABLE) from arterial spin‐labeled MRI |
| title_short | Enhancement of automated blood flow estimates (ENABLE) from arterial spin‐labeled MRI |
| title_sort | enhancement of automated blood flow estimates enable from arterial spin labeled mri |
| work_keys_str_mv | AT shirzadiz enhancementofautomatedbloodflowestimatesenablefromarterialspinlabeledmri AT stefanovicb enhancementofautomatedbloodflowestimatesenablefromarterialspinlabeledmri AT chappellma enhancementofautomatedbloodflowestimatesenablefromarterialspinlabeledmri AT ramirezj enhancementofautomatedbloodflowestimatesenablefromarterialspinlabeledmri AT schwindtg enhancementofautomatedbloodflowestimatesenablefromarterialspinlabeledmri AT masellism enhancementofautomatedbloodflowestimatesenablefromarterialspinlabeledmri AT blackse enhancementofautomatedbloodflowestimatesenablefromarterialspinlabeledmri AT macintoshbj enhancementofautomatedbloodflowestimatesenablefromarterialspinlabeledmri |