학술논문
Machine learning of neuroimaging for assisted diagnosis of cognitive impairment and dementia: A systematic review.
Document Type
Academic Journal
Author
Pellegrini E; Division of Neuroimaging, Centre for Clinical Brain Sciences and Edinburgh Imaging, University of Edinburgh, Scotland, UK.; Ballerini L; Division of Neuroimaging, Centre for Clinical Brain Sciences and Edinburgh Imaging, University of Edinburgh, Scotland, UK.; Hernandez MDCV; Division of Neuroimaging, Centre for Clinical Brain Sciences and Edinburgh Imaging, University of Edinburgh, Scotland, UK.; Chappell FM; Division of Neuroimaging, Centre for Clinical Brain Sciences and Edinburgh Imaging, University of Edinburgh, Scotland, UK.; González-Castro V; Department of Electrical, Systems and Automatics Engineering, Universidad de León, León, Spain.; Anblagan D; Division of Neuroimaging, Centre for Clinical Brain Sciences and Edinburgh Imaging, University of Edinburgh, Scotland, UK.; Danso S; Division of Neuroimaging, Centre for Clinical Brain Sciences and Edinburgh Imaging, University of Edinburgh, Scotland, UK.; Muñoz-Maniega S; Division of Neuroimaging, Centre for Clinical Brain Sciences and Edinburgh Imaging, University of Edinburgh, Scotland, UK.; Job D; Division of Neuroimaging, Centre for Clinical Brain Sciences and Edinburgh Imaging, University of Edinburgh, Scotland, UK.; Pernet C; Division of Neuroimaging, Centre for Clinical Brain Sciences and Edinburgh Imaging, University of Edinburgh, Scotland, UK.; Mair G; Division of Neuroimaging, Centre for Clinical Brain Sciences and Edinburgh Imaging, University of Edinburgh, Scotland, UK.; MacGillivray TJ; Division of Neuroimaging, Centre for Clinical Brain Sciences and Edinburgh Imaging, University of Edinburgh, Scotland, UK.; VAMPIRE project, University of Edinburgh, Scotland, UK.; Trucco E; VAMPIRE project, Computing, School of Science and Engineering, University of Dundee, Dundee, UK.; Wardlaw JM; Division of Neuroimaging, Centre for Clinical Brain Sciences and Edinburgh Imaging, University of Edinburgh, Scotland, UK.; UK Dementia Institute, University of Edinburgh, Scotland, UK.
Source
Publisher: Wiley on behalf of the Alzheimer's Association Country of Publication: United States NLM ID: 101654604 Publication Model: eCollection Cited Medium: Print ISSN: 2352-8729 (Print) Linking ISSN: 23528729 NLM ISO Abbreviation: Alzheimers Dement (Amst) Subsets: PubMed not MEDLINE
Subject
Language
English
ISSN
2352-8729
Abstract
Introduction: Advanced machine learning methods might help to identify dementia risk from neuroimaging, but their accuracy to date is unclear.
Methods: We systematically reviewed the literature, 2006 to late 2016, for machine learning studies differentiating healthy aging from dementia of various types, assessing study quality, and comparing accuracy at different disease boundaries.
Results: Of 111 relevant studies, most assessed Alzheimer's disease versus healthy controls, using AD Neuroimaging Initiative data, support vector machines, and only T1-weighted sequences. Accuracy was highest for differentiating Alzheimer's disease from healthy controls and poor for differentiating healthy controls versus mild cognitive impairment versus Alzheimer's disease or mild cognitive impairment converters versus nonconverters. Accuracy increased using combined data types, but not by data source, sample size, or machine learning method.
Discussion: Machine learning does not differentiate clinically relevant disease categories yet. More diverse data sets, combinations of different types of data, and close clinical integration of machine learning would help to advance the field.
Methods: We systematically reviewed the literature, 2006 to late 2016, for machine learning studies differentiating healthy aging from dementia of various types, assessing study quality, and comparing accuracy at different disease boundaries.
Results: Of 111 relevant studies, most assessed Alzheimer's disease versus healthy controls, using AD Neuroimaging Initiative data, support vector machines, and only T1-weighted sequences. Accuracy was highest for differentiating Alzheimer's disease from healthy controls and poor for differentiating healthy controls versus mild cognitive impairment versus Alzheimer's disease or mild cognitive impairment converters versus nonconverters. Accuracy increased using combined data types, but not by data source, sample size, or machine learning method.
Discussion: Machine learning does not differentiate clinically relevant disease categories yet. More diverse data sets, combinations of different types of data, and close clinical integration of machine learning would help to advance the field.