共 50 条
MRI Radiomic Signature of White Matter Hyperintensities Is Associated With Clinical Phenotypes
被引:13
|作者:
Bretzner, Martin
[1
,2
]
Bonkhoff, Anna K.
[1
]
Schirmer, Markus D.
[1
]
Hong, Sungmin
[1
]
Dalca, Adrian, V
[1
,3
,4
]
Donahue, Kathleen L.
[1
]
Giese, Anne-Katrin
[1
]
Etherton, Mark R.
[1
]
Rist, Pamela M.
[1
,5
,6
]
Nardin, Marco
[1
]
Marinescu, Razvan
[1
,4
]
Wang, Clinton
[1
,4
]
Regenhardt, Robert W.
[1
]
Leclerc, Xavier
[2
]
Lopes, Renaud
[2
,7
]
Benavente, Oscar R.
[8
]
Cole, John W.
[9
,10
]
Donatti, Amanda
[11
,12
]
Griessenauer, Christoph J.
[13
,14
]
Heitsch, Laura
[15
,16
,17
]
Holmegaard, Lukas
[18
]
Jood, Katarina
[18
]
Jimenez-Conde, Jordi
[19
]
Kittner, Steven J.
[9
,10
]
Lemmens, Robin
[20
,21
,22
]
Levi, Christopher R.
[23
,24
]
McArdle, Patrick F.
[25
]
McDonough, Caitrin W.
[26
,27
]
Meschia, James F.
[28
]
Phuah, Chia-Ling
[16
,17
]
Rolfs, Arndt
[29
]
Ropele, Stefan
[30
]
Rosand, Jonathan
[31
]
Roquer, Jaume
[32
,33
]
Rundek, Tatjana
[32
,33
]
Sacco, Ralph L.
[32
,33
]
Schmidt, Reinhold
[30
]
Sharma, Pankaj
[34
,35
]
Slowik, Agnieszka
[36
]
Sousa, Alessandro
[9
,10
]
Stanne, Tara M.
[18
]
Strbian, Daniel
[37
]
Tatlisumak, Turgut
[38
,39
]
Thijs, Vincent
[40
]
Vagal, Achala
[41
]
Wasselius, Johan
[42
,43
]
Woo, Daniel
[44
]
Wu, Ona
[3
]
Zand, Ramin
[45
]
Worrall, Bradford B.
[46
]
机构:
[1] Massachusetts Gen Hosp, J Philip Kistler Stroke Res Ctr, Boston, MA 02114 USA
[2] Univ Lille, U1172 LilNCog JPARC Lille Neurosci & Cognit, CHU Lille, INSERM, Lille, France
[3] Harvard Med Sch, Massachusetts Gen Hosp, AA Martinos Ctr Biomed Imaging, Boston, MA 02115 USA
[4] MIT, Comp Sci & Artificial Intelligence Lab, 77 Massachusetts Ave, Cambridge, MA 02139 USA
[5] Brigham & Womens Hosp, Dept Med, Div Prevent Med, 75 Francis St, Boston, MA 02115 USA
[6] Harvard Med Sch, Boston, MA 02115 USA
[7] Inst Pasteur, US 41 UMS 2014 PLBS, CNRS, Lille, France
[8] Univ British Columbia, Dept Med, Div Neurol, Vancouver, BC, Canada
[9] Univ Maryland, Sch Med, Dept Neurol, Baltimore, MD 21201 USA
[10] Vet Affairs Maryland Hlth Care Syst, Baltimore, MD USA
[11] Univ Campinas UNICAMP, Sch Med Sci, Campinas, SP, Brazil
[12] Brazilian Inst Neurosci & Neurotechnol BRAINN, Campinas, Brazil
[13] Geisinger, Dept Neurosurg, Danville, PA USA
[14] Paracelsus Med Univ, Res Inst Neurointervent, Salzburg, Austria
[15] Washington Univ, Sch Med, Div Emergency Med, St Louis, MO USA
[16] Washington Univ, Sch Med, Dept Neurol, St Louis, MO 63110 USA
[17] Barnes Jewish Hosp, St Louis, MO 63110 USA
[18] Univ Gothenburg, Inst Biomed, Sahlgrenska Acad, Gothenburg, Sweden
[19] Univ Autonoma Barcelona, Inst Hosp del Mar Invest Med IMIM, Dept Neurol, Neurovasc Res Grp NEUVAS, Barcelona, Spain
[20] KU Leuven Univ Leuven, Expt Neurol, Dept Neurosci, Leuven, Belgium
[21] KU Leuven Univ Leuven, Leuven Res Inst Neurosci & Dis LIND, Leuven, Belgium
[22] Univ Hosp Leuven, Vesalius Res Ctr, Dept Neurol, Lab Neurobiol, Leuven, Belgium
[23] Univ Newcastle, Sch Med & Publ Hlth, Newcastle, NSW, Australia
[24] John Hunter Hosp, Dept Neurol, Newcastle, NSW, Australia
[25] Univ Maryland, Sch Med, Dept Med, Div Endocrinol Diabet & Nutr, Baltimore, MD 21201 USA
[26] Univ Florida, Dept Pharmacotherapy & Translat Res, Gainesville, FL USA
[27] Univ Florida, Ctr Pharmacogen, Gainesville, FL USA
[28] Mayo Clin, Dept Neurol, Jacksonville, FL USA
[29] Centogene AG, Rostock, Germany
[30] Med Univ Graz, Dept Neurol, Clin Div Neurogeriatr, Graz, Austria
[31] Massachusetts Gen Hosp, Henry & Allison McCance Ctr Brain Hlth, Ctr Genom Med, Boston, MA 02114 USA
[32] Univ Miami, Miller Sch Med, Dept Neurol, Miami, FL 33136 USA
[33] Univ Miami, Miller Sch Med, Evelyn F McKnight Brain Inst, Miami, FL 33136 USA
[34] Royal Holloway Univ London ICR2UL, Inst Cardiovasc Res, Egham, Surrey, England
[35] Ashford St Peters Hosp, Chertsey Ashford, England
[36] Jagiellonian Univ Med Coll, Dept Neurol, Krakow, Poland
[37] Helsinki Univ Cent Hosp, Dept Neurol, Div Neurocrit Care Emergency Neurol, Helsinki, Finland
[38] Univ Gothenburg, Inst Neurosci Physiol, Dept Clin Neurosci, Sahlgrenska Acad, Gothenburg, Sweden
[39] Sahlgrens Univ Hosp, Dept Neurol, Gothenburg, Sweden
[40] Austin Hlth, Florey Inst Neurosci Mental Hlth, Stroke Div, Dept Neurol, Heidelberg, Vic, Australia
[41] Univ Cincinnati, Coll Med, Dept Radiol, Cincinnati, OH USA
[42] Lund Univ, Dept Clin Sci Lund, Radiol, Lund, Sweden
[43] Skane Univ Hosp, Dept Radiol, Neuroradiol, Malmo, Sweden
[44] Univ Cincinnati, Coll Med, Dept Neurol Rehabil Med, Cincinnati, OH USA
[45] Geisinger, Dept Neurol, Danville, PA USA
[46] Univ Virginia, Dept Neurol Publ Hlth Sci, Charlottesville, VA USA
[47] Univ Technol Sydney, Fac Hlth, Ultimo, NSW, Australia
[48] Skane Univ Hosp, Dept Neurol Rehabil Med, Lund, Sweden
[49] Lund Univ, Dept Clin Sci Lund, Neurol, Lund, Sweden
关键词:
stroke;
cerebrovascular disease (CVD);
MRI;
radiomics;
machine learning;
brain health;
SMALL VESSEL DISEASE;
INTEGRITY;
OUTCOMES;
IMAGES;
VOLUME;
BRAIN;
D O I:
10.3389/fnins.2021.691244
中图分类号:
Q189 [神经科学];
学科分类号:
071006 ;
摘要:
Objective: Neuroimaging measurements of brain structural integrity are thought to be surrogates for brain health, but precise assessments require dedicated advanced image acquisitions. By means of quantitatively describing conventional images, radiomic analyses hold potential for evaluating brain health. We sought to: (1) evaluate radiomics to assess brain structural integrity by predicting white matter hyperintensities burdens (WMH) and (2) uncover associations between predictive radiomic features and clinical phenotypes. Methods: We analyzed a multi-site cohort of 4,163 acute ischemic strokes (AIS) patients with T2-FLAIR MR images with total brain and WMH segmentations. Radiomic features were extracted from normal-appearing brain tissue (brain mask-WMH mask). Radiomics-based prediction of personalized WMH burden was done using ElasticNet linear regression. We built a radiomic signature of WMH with stable selected features predictive of WMH burden and then related this signature to clinical variables using canonical correlation analysis (CCA). Results: Radiomic features were predictive of WMH burden (R-2 = 0.855 +/- 0.011). Seven pairs of canonical variates (CV) significantly correlated the radiomics signature of WMH and clinical traits with respective canonical correlations of 0.81, 0.65, 0.42, 0.24, 0.20, 0.15, and 0.15 (FDR-corrected p-values(CV1-6) < 0.001, p-value(CV7) = 0.012). The clinical CV1 was mainly influenced by age, CV2 by sex, CV3 by history of smoking and diabetes, CV4 by hypertension, CV5 by atrial fibrillation (AF) and diabetes, CV6 by coronary artery disease (CAD), and CV7 by CAD and diabetes. Conclusion: Radiomics extracted from T2-FLAIR images of AIS patients capture microstructural damage of the cerebral parenchyma and correlate with clinical phenotypes, suggesting different radiographical textural abnormalities per cardiovascular risk profile. Further research could evaluate radiomics to predict the progression of WMH and for the follow-up of stroke patients' brain health.
引用
收藏
页数:11
相关论文