Machine learning of neuroimaging for assisted diagnosis of cognitive impairment and dementia: A systematic review

Машинное обучение нейровизуализации для вспомогательной диагностики когнитивных нарушений и деменции: систематический обзор
Joanna Marguerite Wardlaw, Emanuele Trucco, Tom MacGillivray, Dominic E. Job, María del C. Valdés Hernández, Samuel Danso, Susana Muñoz Maniega, Cyril R Pernet, Víctor González‐Castro, Francesca M. Chappell, Enrico Pellegrini, Lucia Ballerini, Devasuda Anblagan, Grant Mair
2018-01-01

Alzheimer's diseasedementia diagnosismachine learningmild cognitive impairmentneuroimaging
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.
1
A systematic review identified 111 neuroimaging machine-learning studies published from 2006 to late 2016 on differentiating healthy aging and dementia.
2
Accuracy was highest for distinguishing Alzheimer’s disease from healthy controls but poor for separating healthy controls, mild cognitive impairment, and Alzheimer’s disease.
3
Machine-learning accuracy improved when multiple data types were combined, but not with data source, sample size, or machine-learning method.
4
Most studies focused on Alzheimer’s disease versus healthy controls, predominantly using Alzheimer’s Disease Neuroimaging Initiative data, support vector machines, and T1-weighted imaging.
5
The reviewed methods did not yet reliably differentiate clinically relevant disease categories; progress requires more diverse datasets, multimodal data, and closer clinical integration.

Neuroimaging data of healthy aging, mild cognitive impairment, and dementia, particularly Alzheimer’s disease

The diagnostic accuracy and clinical discriminative performance of machine-learning models across cognitive-impairment and dementia categories

Publication Details
Publication Date
2018-01-01
Journal
Publisher
ISSN
Cited by
310
Access Type
Author Information
Authors
Joanna Marguerite Wardlaw
Emanuele Trucco
Tom MacGillivray
Dominic E. Job
María del C. Valdés Hernández
Samuel Danso
Susana Muñoz Maniega
Cyril R Pernet
Víctor González‐Castro
Francesca M. Chappell
Enrico Pellegrini
Lucia Ballerini
Devasuda Anblagan
Grant Mair
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%