Big Data Approaches to Phenotyping Acute Ischemic Stroke Using Automated Lesion Segmentation of Multi-Center Magnetic Resonance Imaging Data
-
- Ona Wu
- From Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown (O.W., S.W., B.L.H., M.J.R.J.B., R.E.I., S.J.T.M., E.C.M., R.B.)
-
- Stefan Winzeck
- From Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown (O.W., S.W., B.L.H., M.J.R.J.B., R.E.I., S.J.T.M., E.C.M., R.B.)
-
- Anne-Katrin Giese
- Department of Neurology, JP Kistler Stroke Research Center, MGH, Boston, MA (A.-K.G., M.R.E., K.D., M.D.S., N.S.R.)
-
- Brandon L. Hancock
- From Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown (O.W., S.W., B.L.H., M.J.R.J.B., R.E.I., S.J.T.M., E.C.M., R.B.)
-
- Mark R. Etherton
- Department of Neurology, JP Kistler Stroke Research Center, MGH, Boston, MA (A.-K.G., M.R.E., K.D., M.D.S., N.S.R.)
-
- Mark J.R.J. Bouts
- From Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown (O.W., S.W., B.L.H., M.J.R.J.B., R.E.I., S.J.T.M., E.C.M., R.B.)
-
- Kathleen Donahue
- Department of Neurology, JP Kistler Stroke Research Center, MGH, Boston, MA (A.-K.G., M.R.E., K.D., M.D.S., N.S.R.)
-
- Markus D. Schirmer
- Department of Neurology, JP Kistler Stroke Research Center, MGH, Boston, MA (A.-K.G., M.R.E., K.D., M.D.S., N.S.R.)
-
- Robert E. Irie
- From Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown (O.W., S.W., B.L.H., M.J.R.J.B., R.E.I., S.J.T.M., E.C.M., R.B.)
-
- Steven J.T. Mocking
- From Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown (O.W., S.W., B.L.H., M.J.R.J.B., R.E.I., S.J.T.M., E.C.M., R.B.)
-
- Elissa C. McIntosh
- From Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown (O.W., S.W., B.L.H., M.J.R.J.B., R.E.I., S.J.T.M., E.C.M., R.B.)
-
- Raquel Bezerra
- From Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown (O.W., S.W., B.L.H., M.J.R.J.B., R.E.I., S.J.T.M., E.C.M., R.B.)
-
- Konstantinos Kamnitsas
- Department of Computing, Imperial College London, United Kingdom (K.K.)
-
- Petrea Frid
- Department of Clinical Sciences Lund, Lund University, Sweden (P.F., J.W., A.G.L.)
-
- Johan Wasselius
- Department of Clinical Sciences Lund, Lund University, Sweden (P.F., J.W., A.G.L.)
-
- John W. Cole
- Department of Neurology, University of Maryland School of Medicine and Veterans Affairs Maryland Health Care System, Baltimore, MD (J.W.C., S.J.K.)
-
- Huichun Xu
- Division of Endocrinology, Diabetes and Nutrition, Department of Medicine, University of Maryland School of Medicine, Baltimore, MD (H.X., P.F.M., B.D.M.)
-
- Lukas Holmegaard
- Institute of Neuroscience and Physiology, the Sahlgrenska Academy at University of Gothenburg, Sweden (L.H.)
-
- Jordi Jiménez-Conde
- Department of Neurology, Neurovascular Research Group (NEUVAS), IMIM-Hospital del Mar (Institut Hospital del Mar d’Investigacions Mèdiques), Universitat Autonoma de Barcelona, Spain (J.J.-C., J.R.)
-
- Robin Lemmens
- Department of Neurosciences, Experimental Neurology, KU Leuven—University of Leuven (R.L.)
-
- Eric Lorentzen
- Department of Laboratory Medicine, Institute of Biomedicine, the Sahlgrenska Academy at University of Gothenburg, Sweden (E.L., T.M.S., C.J.)
-
- Patrick F. McArdle
- Division of Endocrinology, Diabetes and Nutrition, Department of Medicine, University of Maryland School of Medicine, Baltimore, MD (H.X., P.F.M., B.D.M.)
-
- James F. Meschia
- Department of Neurology, Mayo Clinic, Jacksonville, FL (J.F.M.)
-
- Jaume Roquer
- Department of Neurology, Neurovascular Research Group (NEUVAS), IMIM-Hospital del Mar (Institut Hospital del Mar d’Investigacions Mèdiques), Universitat Autonoma de Barcelona, Spain (J.J.-C., J.R.)
-
- Tatjana Rundek
- Department of Neurology, Miller School of Medicine, University of Miami, The Evelyn F. McKnight Brain Institute, FL (T.R., R.L.S.)
-
- Ralph L. Sacco
- Department of Neurology, Miller School of Medicine, University of Miami, The Evelyn F. McKnight Brain Institute, FL (T.R., R.L.S.)
-
- Reinhold Schmidt
- Clinical Division of Neurogeriatrics, Department of Neurology, Medical University Graz, Austria (R.S.)
-
- Pankaj Sharma
- Institute of Cardiovascular Research, Royal Holloway University of London (ICR2UL), Egham, United Kingdom (P.S.)
-
- Agnieszka Slowik
- Department of Neurology, Jagiellonian University Medical College, Krakow, Poland (A.S.)
-
- Tara M. Stanne
- Department of Laboratory Medicine, Institute of Biomedicine, the Sahlgrenska Academy at University of Gothenburg, Sweden (E.L., T.M.S., C.J.)
-
- Vincent Thijs
- Stroke Division, Florey Institute of Neuroscience and Mental Health, HDB, Australia (V.T.)
-
- Achala Vagal
- Department of Radiology (A.V.), University of Cincinnati College of Medicine, OH
-
- Daniel Woo
- Department of Neurology and Rehabilitation Medicine (D.W.), University of Cincinnati College of Medicine, OH
-
- Stephen Bevan
- School of Life Science, University of Lincoln, United Kingdom (S.B.)
-
- Steven J. Kittner
- Department of Neurology, University of Maryland School of Medicine and Veterans Affairs Maryland Health Care System, Baltimore, MD (J.W.C., S.J.K.)
-
- Braxton D. Mitchell
- Division of Endocrinology, Diabetes and Nutrition, Department of Medicine, University of Maryland School of Medicine, Baltimore, MD (H.X., P.F.M., B.D.M.)
-
- Jonathan Rosand
- Henry and Allison McCance Center for Brain Health Massachusetts General Hospital, Boston (J.R.)
-
- Bradford B. Worrall
- Departments of Neurology and Public Health Sciences, University of Virginia, Charlottesville (B.B.W.)
-
- Christina Jern
- Department of Laboratory Medicine, Institute of Biomedicine, the Sahlgrenska Academy at University of Gothenburg, Sweden (E.L., T.M.S., C.J.)
-
- Arne G. Lindgren
- Department of Clinical Sciences Lund, Lund University, Sweden (P.F., J.W., A.G.L.)
-
- Jane Maguire
- University of Technology Sydney, Australia (J.M.).
-
- Natalia S. Rost
- Department of Neurology, JP Kistler Stroke Research Center, MGH, Boston, MA (A.-K.G., M.R.E., K.D., M.D.S., N.S.R.)
抄録
<jats:sec> <jats:title>Background and Purpose—</jats:title> <jats:p>We evaluated deep learning algorithms’ segmentation of acute ischemic lesions on heterogeneous multi-center clinical diffusion-weighted magnetic resonance imaging (MRI) data sets and explored the potential role of this tool for phenotyping acute ischemic stroke.</jats:p> </jats:sec> <jats:sec> <jats:title>Methods—</jats:title> <jats:p>Ischemic stroke data sets from the MRI-GENIE (MRI-Genetics Interface Exploration) repository consisting of 12 international genetic research centers were retrospectively analyzed using an automated deep learning segmentation algorithm consisting of an ensemble of 3-dimensional convolutional neural networks. Three ensembles were trained using data from the following: (1) 267 patients from an independent single-center cohort, (2) 267 patients from MRI-GENIE, and (3) mixture of (1) and (2). The algorithms’ performances were compared against manual outlines from a separate 383 patient subset from MRI-GENIE. Univariable and multivariable logistic regression with respect to demographics, stroke subtypes, and vascular risk factors were performed to identify phenotypes associated with large acute diffusion-weighted MRI volumes and greater stroke severity in 2770 MRI-GENIE patients. Stroke topography was investigated.</jats:p> </jats:sec> <jats:sec> <jats:title>Results—</jats:title> <jats:p> The ensemble consisting of a mixture of MRI-GENIE and single-center convolutional neural networks performed best. Subset analysis comparing automated and manual lesion volumes in 383 patients found excellent correlation (ρ=0.92; <jats:italic>P</jats:italic> <0.0001). Median (interquartile range) diffusion-weighted MRI lesion volumes from 2770 patients were 3.7 cm <jats:sup>3</jats:sup> (0.9–16.6 cm <jats:sup>3</jats:sup> ). Patients with small artery occlusion stroke subtype had smaller lesion volumes ( <jats:italic>P</jats:italic> <0.0001) and different topography compared with other stroke subtypes. </jats:p> </jats:sec> <jats:sec> <jats:title>Conclusions—</jats:title> <jats:p>Automated accurate clinical diffusion-weighted MRI lesion segmentation using deep learning algorithms trained with multi-center and diverse data is feasible. Both lesion volume and topography can provide insight into stroke subtypes with sufficient sample size from big heterogeneous multi-center clinical imaging phenotype data sets.</jats:p> </jats:sec>
収録刊行物
-
- Stroke
-
Stroke 50 (7), 1734-1741, 2019-07
Ovid Technologies (Wolters Kluwer Health)