Deep learning for efficient reconstruction of highly accelerated 3D FLAIR MRI in neurological deficits.
Accelerated imaging
Deep learning
Image resolution
MRI
Neurological deficits
Journal
Magma (New York, N.Y.)
ISSN: 1352-8661
Titre abrégé: MAGMA
Pays: Germany
ID NLM: 9310752
Informations de publication
Date de publication:
30 Aug 2024
30 Aug 2024
Historique:
received:
30
04
2024
accepted:
08
08
2024
revised:
07
08
2024
medline:
31
8
2024
pubmed:
31
8
2024
entrez:
30
8
2024
Statut:
aheadofprint
Résumé
To compare compressed sensing (CS) and the Cascades of Independently Recurrent Inference Machines (CIRIM) with respect to image quality and reconstruction times when 12-fold accelerated scans of patients with neurological deficits are reconstructed. Twelve-fold accelerated 3D T2-FLAIR images were obtained from a cohort of 62 patients with neurological deficits on 3 T MRI. Images were reconstructed offline via CS and the CIRIM. Image quality was assessed in a blinded and randomized manner by two experienced interventional neuroradiologists and one experienced pediatric neuroradiologist on imaging artifacts, perceived spatial resolution (sharpness), anatomic conspicuity, diagnostic confidence, and contrast. The methods were also compared in terms of self-referenced quality metrics, image resolution, patient groups and reconstruction time. In ten scans, the contrast ratio (CR) was determined between lesions and white matter. The effect of acceleration factor was assessed in a publicly available fully sampled dataset, since ground truth data are not available in prospectively accelerated clinical scans. Specifically, 451 FLAIR scans, including scans with white matter lesions, were adopted from the FastMRI database to evaluate structural similarity (SSIM) and the CR of lesions and white matter on ranging acceleration factors from four-fold up to 12-fold. Interventional neuroradiologists significantly preferred the CIRIM for imaging artifacts, anatomic conspicuity, and contrast. One rater significantly preferred the CIRIM in terms of sharpness and diagnostic confidence. The pediatric neuroradiologist preferred CS for imaging artifacts and sharpness. Compared to CS, the CIRIM reconstructions significantly improved in terms of imaging artifacts and anatomic conspicuity (p < 0.01) for higher resolution scans while yielding a 28% higher SNR (p = 0.001) and a 5.8% lower CR (p = 0.04). There were no differences between patient groups. Additionally, CIRIM was five times faster than CS was. An increasing acceleration factor did not lead to changes in CR (p = 0.92), but led to lower SSIM (p = 0.002). Patients with neurological deficits can undergo MRI at a range of moderate to high acceleration. DL reconstruction outperforms CS in terms of image resolution, efficient denoising with a modest reduction in contrast and reduced reconstruction times.
Identifiants
pubmed: 39212832
doi: 10.1007/s10334-024-01200-8
pii: 10.1007/s10334-024-01200-8
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Informations de copyright
© 2024. The Author(s).
Références
Toledano-Massiah S et al (2018) Accuracy of the compressed sensing accelerated 3D-FLAIR sequence for the detection of ms plaques at 3T. Am J Neuroradiol 39(3):454–458. https://doi.org/10.3174/AJNR.A5517
doi: 10.3174/AJNR.A5517
pubmed: 29348137
pmcid: 7655323
Yamamoto T et al (2022) Validation of a denoising method using deep learning- based reconstruction to quantify multiple sclerosis lesion load on fast FLAIR imaging. Am J Neuroradiol 43(8):1099–1106. https://doi.org/10.3174/ajnr.A7589
doi: 10.3174/ajnr.A7589
pubmed: 35902124
pmcid: 9575422
Vranic JE, Cross NM, Wang Y, Hippe DS, De Weerdt E, Mossa-Basha MMD (2019) Compressed sensing-sensitivity encoding (CS-SENSE) Accelerated brain imaging: reduced scan time without reduced image quality. Am J Neuroradiol 40(1):92–98. https://doi.org/10.3174/AJNR.A5905
doi: 10.3174/AJNR.A5905
pubmed: 30523142
pmcid: 7048595
Okano Y et al (2018) Clinical features of stroke mimics in the emergency department. Acute Med Surg 5(3):241–248. https://doi.org/10.1002/AMS2.338
doi: 10.1002/AMS2.338
pubmed: 29988676
pmcid: 6028791
Hextrum S, Biller J (2018) Clinical distinction of cerebral ischemia and triaging of patients in the emergency department: mimics, wake-ups, late strokes, and chameleons. Neuroimaging Clin North Am 28(4):537–549. https://doi.org/10.1016/j.nic.2018.06.001
doi: 10.1016/j.nic.2018.06.001
Thomalla G et al (2011) DWI-FLAIR mismatch for the identification of patients with acute ischaemic stroke within 4·5 h of symptom onset (PRE-FLAIR): a multicentre observational study. Lancet Neurol 10(11):978–986. https://doi.org/10.1016/S1474-4422(11)70192-2
doi: 10.1016/S1474-4422(11)70192-2
pubmed: 21978972
Jiang L et al (2020) FLAIR hyperintensities-DWI mismatch in acute stroke: associations with DWI volume and functional outcome. Brain Imaging Behav 14(4):1230–1237. https://doi.org/10.1007/s11682-019-00156-x
doi: 10.1007/s11682-019-00156-x
pubmed: 31250263
Garciá-Tornel Á et al (2021) Ischemic core overestimation on computed tomography perfusion. Stroke. https://doi.org/10.1161/STROKEAHA.120.031800
doi: 10.1161/STROKEAHA.120.031800
pubmed: 33682453
Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P (1999) SENSE: sensitivity encoding for fast MRI. Magn Reson Med 42(5):952–962. https://doi.org/10.1002/(SICI)1522-2594(199911)42:5%3c952::AID-MRM16%3e3.0.CO;2-S
doi: 10.1002/(SICI)1522-2594(199911)42:5<952::AID-MRM16>3.0.CO;2-S
pubmed: 10542355
Griswold MA et al (2002) Generalized autocalibrating partially parallel acquisitions (GRAPPA). Magn Reson Med 47(6):1202–1210. https://doi.org/10.1002/mrm.10171
doi: 10.1002/mrm.10171
pubmed: 12111967
Lustig M and Donoho D (2008) Compressed sensing MRI,” Signal Process. …, no. pp. 72–82, 2008, Accessed 07 Feb 2015. [Online]. Available: http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4472246
Ding J et al (2021) Acceleration of brain TOF-MRA with compressed sensitivity encoding: a multicenter clinical study. Am J Neuroradiol 42(7):1208–1215. https://doi.org/10.3174/AJNR.A7091
doi: 10.3174/AJNR.A7091
pubmed: 33858820
pmcid: 8324268
Lu SS et al (2018) Clinical evaluation of highly accelerated compressed sensing time-of-flight MR angiography for intracranial arterial stenosis. Am J Neuroradiol 39(10):1833–1838. https://doi.org/10.3174/AJNR.A5786
doi: 10.3174/AJNR.A5786
pubmed: 30213812
pmcid: 7410741
Bash S et al (2021) Deep learning enables 60% accelerated volumetric brain MRI while preserving quantitative performance: a prospective, multicenter, multireader trial. Am J Neuroradiol 42(12):2130–2137. https://doi.org/10.3174/AJNR.A7358
doi: 10.3174/AJNR.A7358
pubmed: 34824098
pmcid: 8805755
Shaul R, David I, Shitrit O, Riklin Raviv T (2020) Subsampled brain MRI reconstruction by generative adversarial neural networks. Med Image Anal 65:101747. https://doi.org/10.1016/j.media.2020.101747
doi: 10.1016/j.media.2020.101747
pubmed: 32593933
Ursprung S et al (2023) Accelerated diffusion-weighted imaging of the prostate using deep learning image reconstruction: a retrospective comparison with standard diffusion-weighted imaging. Eur J Radiol 165:110953. https://doi.org/10.1016/j.ejrad.2023.110953
doi: 10.1016/j.ejrad.2023.110953
pubmed: 37399667
Aggarwal HK, Mani MP, Jacob M (2019) MoDL: model-based deep learning architecture for inverse problems. IEEE Trans Med Imaging 38(2):394–405. https://doi.org/10.1109/TMI.2018.2865356
doi: 10.1109/TMI.2018.2865356
pubmed: 30106719
Hammernik K et al (2018) Learning a variational network for reconstruction of accelerated MRI data. Magn Reson Med 79(6):3055–3071. https://doi.org/10.1002/mrm.26977
doi: 10.1002/mrm.26977
pubmed: 29115689
Lønning K, Putzky P, Sonke J-J, Reneman L, Caan MWA, Welling M (2019) Recurrent inference machines for reconstructing heterogeneous MRI data. Med Image Anal. https://doi.org/10.1016/j.media.2019.01.005
doi: 10.1016/j.media.2019.01.005
pubmed: 30703579
Karkalousos D, Noteboom S, Hulst HE, Vos FM, Caan MWA (2022) Assessment of data consistency through cascades of independently recurrent inference machines for fast and robust accelerated MRI reconstruction. Phys Med Biol 67(12):124001. https://doi.org/10.1088/1361-6560/ac6cc2
doi: 10.1088/1361-6560/ac6cc2
M. Uecker M, Ong F, Tamir JI, Bahri D, Virtue P, Cheng JY, Zhang T, Lustig M (2015) Berkeley advanced reconstruction toolbox. In Proc. Intl. Soc. Mag. Reson. Med., p. 23.
Rudie JD et al (2022) Clinical assessment of deep learning–based super-resolution for 3D volumetric brain MRI. Radiol Artif Intell. https://doi.org/10.1148/ryai.210059
doi: 10.1148/ryai.210059
pubmed: 36523646
pmcid: 9748624
Esteban O, Birman D, Schaer M, Koyejo OO, Poldrack RA, Gorgolewski KJ (2017) MRIQC: advancing the automatic prediction of image quality in MRI from unseen sites. PLoS ONE 12(9):e0184661. https://doi.org/10.1371/journal.pone.0184661
doi: 10.1371/journal.pone.0184661
pubmed: 28945803
pmcid: 5612458
Ganzetti M, Wenderoth N, Mantini D (2016) Intensity inhomogeneity correction of structural MR images: a data-driven approach to define input algorithm parameters. Front Neuroinform. https://doi.org/10.3389/fninf.2016.00010
doi: 10.3389/fninf.2016.00010
pubmed: 27014050
pmcid: 4791378
Mortamet B et al (2009) Automatic quality assessment in structural brain magnetic resonance imaging. Magn Reson Med 62(2):365–372. https://doi.org/10.1002/mrm.21992
doi: 10.1002/mrm.21992
pubmed: 19526493
pmcid: 2780021
Atkinson D et al (1999) Automatic compensation of motion artifacts in MRI. Magn Reson Med 41(1):163–170. https://doi.org/10.1002/(SICI)1522-2594(199901)41:1%3c163::AID-MRM23%3e3.0.CO;2-9
doi: 10.1002/(SICI)1522-2594(199901)41:1<163::AID-MRM23>3.0.CO;2-9
pubmed: 10025625
Zarrar S et al (2015) The preprocessed connectomes project quality assessment protocol - a resource for measuring the quality of MRI data. Front Neurosci. https://doi.org/10.3389/conf.fnins.2015.91.00047
doi: 10.3389/conf.fnins.2015.91.00047
Forman SD, Cohen JD, Fitzgerald M, Eddy WF, Mintun MA, Noll DC (1995) Improved assessment of significant activation in functional magnetic resonance imaging (fMRI): use of a cluster-size threshold. Magn Reson Med 33(5):636–647. https://doi.org/10.1002/mrm.1910330508
doi: 10.1002/mrm.1910330508
pubmed: 7596267
Muckley MJ et al (2021) Results of the 2020 fastMRI challenge for machine learning MR image reconstruction. IEEE Trans Med Imaging 40(9):2306–2317. https://doi.org/10.1109/TMI.2021.3075856
doi: 10.1109/TMI.2021.3075856
pubmed: 33929957
pmcid: 8428775
Zhao R et al (2022) fastMRI+, clinical pathology annotations for knee and brain fully sampled magnetic resonance imaging data. Sci Data 9(1):152. https://doi.org/10.1038/s41597-022-01255-z
doi: 10.1038/s41597-022-01255-z
pubmed: 35383186
pmcid: 8983757
Virtanen P et al (2020) SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat Methods 17(3):261–272. https://doi.org/10.1038/s41592-019-0686-2
doi: 10.1038/s41592-019-0686-2
pubmed: 32015543
pmcid: 7056644
Kiryu S et al (2023) Clinical impact of deep learning reconstruction in MRI. Radiographics. https://doi.org/10.1148/rg.220133
doi: 10.1148/rg.220133
pubmed: 37200221
Estler A et al (2024) Deep learning accelerated image reconstruction of fluid-attenuated inversion recovery sequence in brain imaging: reduction of acquisition time and improvement of image quality. Acad Radiol 31(1):180–186. https://doi.org/10.1016/j.acra.2023.05.010
doi: 10.1016/j.acra.2023.05.010
pubmed: 37280126