Please use this identifier to cite or link to this item: http://ahro.austin.org.au/austinjspui/handle/1/22679
Title: Big Data Approaches to Phenotyping Acute Ischemic Stroke Using Automated Lesion Segmentation of Multi-Center Magnetic Resonance Imaging Data.
Authors: Wu, Ona;Winzeck, Stefan;Giese, Anne-Katrin;Hancock, Brandon L;Etherton, Mark R;Bouts, Mark J R J;Donahue, Kathleen;Schirmer, Markus D;Irie, Robert E;Mocking, Steven J T;McIntosh, Elissa C;Bezerra, Raquel;Kamnitsas, Konstantinos;Frid, Petrea;Wasselius, Johan;Cole, John W;Xu, Huichun;Holmegaard, Lukas;Jiménez-Conde, Jordi;Lemmens, Robin;Lorentzen, Eric;McArdle, Patrick F;Meschia, James F;Roquer, Jaume;Rundek, Tatjana;Sacco, Ralph L;Schmidt, Reinhold;Sharma, Pankaj;Slowik, Agnieszka;Stanne, Tara M;Thijs, Vincent N;Vagal, Achala;Woo, Daniel;Bevan, Stephen;Kittner, Steven J;Mitchell, Braxton D;Rosand, Jonathan;Worrall, Bradford B;Jern, Christina;Lindgren, Arne G;Maguire, Jane;Rost, Natalia S
Affiliation: Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown
Division of Anaesthesia, Department of Medicine, University of Cambridge, United Kingdom
Division of Endocrinology, Diabetes and Nutrition, Department of Medicine, University of Maryland School of Medicine, Baltimore, MD
Geriatrics Research and Education Clinical Center, Baltimore Veterans Administration Medical Center, MD
Stroke Division, Florey Institute of Neuroscience and Mental Health, HDB, Australia
Department of Clinical Sciences Lund, Lund University, Sweden
Department of Neurology and Rehabilitation Medicine, Neurology, Skåne University Hospital, Lund, Sweden..
Institute of Cardiovascular Research, Royal Holloway University of London (ICR2UL), Egham, United Kingdom
Ashford and St Peter's Hospital, United Kingdom
Department of Neurosciences, Experimental Neurology, KU Leuven-University of Leuven
VIB-Center for Brain & Disease Research
Department of Neurology, University Hospitals Leuven, Belgium
Department of Radiology, Skåne University Hospital, Lund, Sweden
Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown
Department of Neurology, JP Kistler Stroke Research Center, MGH, Boston, MA
From Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown
Department of Computing, Imperial College London, United Kingdom
Department of Clinical Sciences Lund, Lund University, Sweden
Department of Neurology, University of Maryland School of Medicine and Veterans Affairs Maryland Health Care System, Baltimore, MD
Division of Endocrinology, Diabetes and Nutrition, Department of Medicine, University of Maryland School of Medicine, Baltimore, MD
Institute of Neuroscience and Physiology, the Sahlgrenska Academy at University of Gothenburg, Sweden
Department of Neurology, Neurovascular Research Group (NEUVAS), IMIM-Hospital del Mar (Institut Hospital del Mar d'Investigacions Mèdiques), Universitat Autonoma de Barcelona, Spain
Department of Laboratory Medicine, Institute of Biomedicine, the Sahlgrenska Academy at University of Gothenburg, Sweden
Division of Endocrinology, Diabetes and Nutrition, Department of Medicine, University of Maryland School of Medicine, Baltimore, MD
Department of Neurology, Mayo Clinic, Jacksonville, FL
Department of Neurology, Neurovascular Research Group (NEUVAS), IMIM-Hospital del Mar (Institut Hospital del Mar d'Investigacions Mèdiques), Universitat Autonoma de Barcelona, Spain
Department of Neurology, Miller School of Medicine, University of Miami, The Evelyn F. McKnight Brain Institute, FL
Clinical Division of Neurogeriatrics, Department of Neurology, Medical University Graz, Austria
Department of Neurology, Jagiellonian University Medical College, Krakow, Poland
Department of Laboratory Medicine, Institute of Biomedicine, the Sahlgrenska Academy at University of Gothenburg, Sweden
Department of Radiology, University of Cincinnati College of Medicine, OH
Department of Neurology and Rehabilitation Medicine, University of Cincinnati College of Medicine, OH
School of Life Science, University of Lincoln, United Kingdom
Department of Neurology, University of Maryland School of Medicine and Veterans Affairs Maryland Health Care System, Baltimore, MD
Henry and Allison McCance Center for Brain Health Massachusetts General Hospital, Boston
Departments of Neurology and Public Health Sciences, University of Virginia, Charlottesville
Department of Laboratory Medicine, Institute of Biomedicine, the Sahlgrenska Academy at University of Gothenburg, Sweden
University of Technology Sydney, Australia
Department of Neurology, JP Kistler Stroke Research Center, MGH, Boston, MA
Department of Neurology, Austin Health, Heidelberg, Victoria, Australia
Issue Date: Jul-2019
EDate: 2019-07
Citation: Stroke 2019; 50(7): 1734-1741
Abstract: Background and Purpose- 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. Methods- 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. Results- 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; P<0.0001). Median (interquartile range) diffusion-weighted MRI lesion volumes from 2770 patients were 3.7 cm3 (0.9-16.6 cm3). Patients with small artery occlusion stroke subtype had smaller lesion volumes ( P<0.0001) and different topography compared with other stroke subtypes. Conclusions- 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.
URI: http://ahro.austin.org.au/austinjspui/handle/1/22679
DOI: 10.1161/STROKEAHA.119.025373
ORCID: 0000-0002-6614-8417
PubMed URL: 31177973
Type: Journal Article
Multicenter Study
Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't
Research Support, U.S. Gov't, Non-P.H.S.
Subjects: diffusion magnetic resonance imaging
machine learning
phenotype
risk factors
stroke
Appears in Collections:Journal articles

Files in This Item:
There are no files associated with this item.


Items in AHRO are protected by copyright, with all rights reserved, unless otherwise indicated.