Biomarker Extraction Based on Subspace Learning for the Prediction of Mild Cognitive Impairment Conversion

Accurate recognition of progressive mild cognitive impairment (MCI) is helpful to reduce the risk of developing Alzheimer’s disease (AD). However, it is still challenging to extract effective biomarkers from multivariate brain structural magnetic resonance imaging (MRI) features to accurately differentiate the progressive MCI from stable MCI. We develop novel biomarkers by combining subspace learning methods with the information of AD as well as normal control (NC) subjects for the prediction of MCI conversion using multivariate structural MRI data. Specifically, we first learn two projection matrices to map multivariate structural MRI data into a common label subspace for AD and NC subjects, where the original data structure and the one-to-one correspondence between multiple variables are kept as much as possible. Afterwards, the multivariate structural MRI features of MCI subjects are mapped into a common subspace according to the projection matrices. We then perform the self-weighted operation and weighted fusion on the features in common subspace to extract the novel biomarkers for MCI subjects. The proposed biomarkers are tested on Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset. Experimental results indicate that our proposed biomarkers outperform the competing biomarkers on the discrimination between progressive MCI and stable MCI. And the improvement from the proposed biomarkers is not limited to a particular classifier. Moreover, the results also confirm that the information of AD and NC subjects is conducive to predicting conversion from MCI to AD. In conclusion, we find a good representation of brain features from high-dimensional MRI data, which exhibits promising performance for predicting conversion from MCI to AD.

Download Full-text

A Genetic Algorithm for the selection of structural MRI features for classification of Mild Cognitive Impairment and Alzheimer's Disease

2015 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) ◽

10.1109/bibm.2015.7359909 ◽

2015 ◽

Cited By ~ 3

Author(s):

Alexander Luke Spedding ◽

Giuseppe Di Fatta ◽

Mario Cannataro

Keyword(s):

Alzheimer’S Disease ◽

Genetic Algorithm ◽

Alzheimer's Disease ◽

Cognitive Impairment ◽

Mild Cognitive Impairment ◽

Structural Mri ◽

Mri Features ◽

Selection Of

Download Full-text

A comparison of resting state EEG and structural MRI for classifying Alzheimer’s disease and mild cognitive impairment

NeuroImage ◽

10.1016/j.neuroimage.2020.116795 ◽

2020 ◽

Vol 215 ◽

pp. 116795 ◽

Cited By ~ 2

Author(s):

F.R. Farina ◽

D.D. Emek-Savaş ◽

L. Rueda-Delgado ◽

R. Boyle ◽

H. Kiiski ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Cognitive Impairment ◽

Mild Cognitive Impairment ◽

Resting State ◽

Structural Mri

Download Full-text

A comparison of resting state EEG and structural MRI for classifying Alzheimer’s disease and mild cognitive impairment

10.1101/711465 ◽

2019 ◽

Author(s):

FR Farina ◽

DD Emek-Savaş ◽

L Rueda-Delgado ◽

R Boyle ◽

H Kiiski ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Cognitive Impairment ◽

Mild Cognitive Impairment ◽

Resting State ◽

Healthy Aging ◽

Neurodegenerative Disorder ◽

Low Cost ◽

Structural Mri ◽

Lower Accuracy

AbstractAlzheimer’s disease (AD) is a neurodegenerative disorder characterised by severe cognitive decline and loss of autonomy. AD is the leading cause of dementia. AD is preceded by mild cognitive impairment (MCI). By 2050, 68% of new dementia cases will occur in low- and middle-income countries. In the absence of objective biomarkers, psychological assessments are typically used to diagnose MCI and AD. However, these require specialist training and rely on subjective judgements. The need for low-cost, accessible and objective tools to aid AD and MCI diagnosis is therefore crucial. Electroencephalography (EEG) has potential as one such tool: it is relatively inexpensive (cf. magnetic resonance imaging; MRI) and is portable. In this study, we collected resting state EEG, structural MRI and rich neuropsychological data from older adults (55+ years) with AD, with MCI and from healthy controls (n~60 per group). Our goal was to evaluate the utility of EEG, relative to MRI, for the classification of MCI and AD. We also assessed the performance of combined EEG and behavioural (Mini-Mental State Examination; MMSE) and structural MRI classification models. Resting state EEG classified AD and HC participants with moderate accuracy (AROC=0.76), with lower accuracy when distinguishing MCI from HC participants (AROC=0.67). The addition of EEG data to MMSE scores had no additional value compared to MMSE alone. Structural MRI out-performed EEG (AD vs HC, AD vs MCI: AROCs=1.00; HC vs MCI: AROC=0.73). Resting state EEG does not appear to be a suitable tool for classifying AD. However, EEG classification accuracy was comparable to structural MRI when distinguishing MCI from healthy aging, although neither were sufficiently accurate to have clinical utility. This is the first direct comparison of EEG and MRI as classification tools in AD and MCI participants.

Download Full-text

Progressive Volume Atrophy in Hippocampal Subfields and the Correlation with Cognition in Alzheimer’s Disease and Mild Cognitive Impairment

10.21203/rs.3.rs-304948/v1 ◽

2021 ◽

Author(s):

Guixia Kang ◽

Peiqi Luo ◽

Xin Xu ◽

Ying Han ◽

Xuemei Li ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Cognitive Impairment ◽

Mild Cognitive Impairment ◽

Reference Value ◽

Structural Mri ◽

Volume Changes ◽

Brain Scans ◽

Left Hippocampus ◽

The Right

Abstract Objective: To assess the progression of volume changes in hippocampus and its subfields of patients with Alzheimer's disease (AD) and mild cognitive impairment (MCI), and to explore the association of the hippocampus and its subfields volumes with cognitive function.Methods: Five groups of participants including 35 normal controls (NC) persons, 30 MCI patients, 30 Mild AD patients, 30 Moderate AD patients and 8 Severe AD patients received structural MRI brain scans. Freesurfer6.0 was used for automatically segmentation of MRI, and the left and right hippocampus were respectively divided into 12 subfields. By statistical analysis, the volumes of hippocampus and its subfields were compared between the five groups, and the correlation of the volumes with Mini-mental State Examination (MMSE) score was analyzed.Result & Conclusion: In the disease, each hippocampal subfield shows an uneven atrophy trajectory; The volumes of the subiculum and presubiculum are significantly different between Mild AD and MCI, which can contribute to the early diagnosis of AD; Parasubiculum is the least sensitive subfield for volume atrophy of AD, while subiculum, presubiculum, CA1, molecular_layer_HP and fimbria show much more significant volume changes. Meanwhile the volumes of these five subfields are positively correlated with MMSE, which may help in stage division of AD; Compared with the right hippocampus, the volume atrophy on the left side is more significantly, and the volumes are more significantly correlated with MMSE, So the left hippocampus and its subfields may provide a higher reference value for the clinical evaluation of AD than the right side.

Download Full-text