Multimodal Classification of Alzheimer’s Disease and Mild Cognitive Impairment

By Daoqiang Zhang, Yaping Wang, Luping Zhou, Hong Yuan, Dinggang Shen. NeuroImage, 2011

1,289 citations recorded by the source corpus.

Effective and accurate diagnosis of Alzheimer’s disease (AD), as well as its prodromal stage (i.e., mild cognitive impairment (MCI)), has attracted more and more attentions recently. So far, multiple biomarkers have been shown sensitive to the diagnosis of AD and MCI, i.e., structural MR imaging (MRI) for brain atrophy measurement, functional imaging (e.g., FDG-PET) for hypometabolism quantification, and cerebrospinal fluid (CSF) for quantification of specific proteins. However, most existing research focuses on only a single modality of biomarkers for diagnosis of AD and MCI, although recent studies have shown that different biomarkers may provide complementary information for diagnosis of AD and MCI. In this paper, we propose to combine three modalities of biomarkers, i.e., MRI, FDG-PET, and CSF biomarkers, to discriminate between AD (or MCI) and healthy controls, using a kernel combination method. Specifically, ADNI baseline MRI, FDG-PET, and CSF data from 51 AD patients, 99 MCI patients (including 43 MCI converters who had converted to AD within 18 months and 56 MCI non-converters who had not converted to AD within 18 months), and 52 healthy controls are used for development and validation of our proposed multimodal classification method. In particular, for each MR or FDG-PET image, 93 volumetric features are extracted from the 93 regions of interest (ROIs), automatically labeled by an atlas warping algorithm. For CSF biomarkers, their original values are directly used as features. Then, a linear support vector machine (SVM) is adopted to evaluate the classification accuracy, using a 10-fold cross-validation. As a result, for classifying AD from healthy controls, we achieve a classification accuracy of 93.2% (with a sensitivity of 93% and a specificity of 93.3%) when combining all three modalities of biomarkers, and only 86.5% when using even the best individual modality of biomarkers. Similarly, for classifying MCI from healthy controls, we achieve a classification accuracy of 76.4% (with a sensitivity of 81.8% and a specificity of 66%) for our combined method, and only 72% even using the best individual modality of biomarkers. Further analysis on MCI sensitivity of our combined method indicates that 91.5% of MCI converters and 73.4% of MCI non-converters are correctly classified. Moreover, we also evaluate the classification performance when employing a feature selection method to select the most discriminative MR and FDG-PET features. Again, our combined method shows considerably better performance, compared to the case of using an individual modality of biomarkers.

Subfields: Artificial Intelligence, Neurology, Psychiatry and Mental health

View Paper: https://www.semanticscholar.org/paper/3cf57015668a61e93003fdf469e3437f0aef8f7d

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Neurology

In Psychiatry and Mental health

Citations in the corpus

Convolutional Neural Networks-Based MRI Image Analysis for the Alzheimer’s Disease Prediction From Mild Cognitive Impairment (2018, methodology): “…used neuroimaging biomarkers to classify AD patients at diﬀerent disease stages or to predict the MCI-to-AD conversion (Cuingnet et al., 2011; Zhang et al., 2011; Tong et al., 2013, 2017; Guerrero et al., 2014; Suk et al., 2014; Cheng et al., 2015; Eskildsen et al., 2015; Li et al., 2015;…”
Random forest-based similarity measures for multi-modal classification of Alzheimer's disease (2013, positive): “This finding is in agreement with other recent studies (Hinrichs et al., 2011; Zhang et al., 2011), and supports previous suggestions that there is some complementary information between modalities which can be exploited to produce more powerful combined classifiers (Landau et al.”
Ensemble Sparse Classification of Alzheimer’s Disease (2012, methodology): “Among them, support vector machine (SVM) may be the mostwidely used classifier, because of its high performance for classification of highdimensional data (Davatzikos et al., 2008; Fan et al., 2007; Kloppel et al., 2008, epub; Magnin et al., 2009; Zhang et al., 2011).”
Imaging based enrichment criteria using deep learning algorithms for efficient clinical trials in MCI (2015, background): “To this end, several recent studies have used support vector machines (SVMs) and other machine learning models to design such multimodal markers [8,13–16].”
Brain MRI analysis for Alzheimer’s disease diagnosis using an ensemble system of deep convolutional neural networks (2018, positive): “Moreover, information from multiple modalities has been combined to improve the diagnosis performance [36–47].”
Multi-Modal Neuroimaging Feature Learning for Multi-Class Diagnosis of Alzheimer’s Disease (2014, methodology): “In the framework proposed by Suk et al. [39], one setting of stacked autoencoders (SAEs) was trained for each image modality; then, the learnt high-level features were further fused with a multikernel support vector machine (MKSVM).”
Multimodal Neuroimaging Feature Learning With Multimodal Stacked Deep Polynomial Networks for Diagnosis of Alzheimer's Disease (2018, methodology): “The learned features from the second stage SDPN and the individual learned features from MRI and PET in the first stage SDPNs can be regarded as multi-view data for MKL.”
Does feature selection improve classification accuracy? Impact of sample size and feature selection on classification using anatomical magnetic resonance images (2012, methodology): “The imaging data came from the Alzheimer's disease Neuroimaging Initiative (ADNI), which have been used by many studies of AD classification (Filipovych and Davatzikos, 2011; Gerardin et al., 2009; PelaezCoca et al., 2011; Stonnington et al., 2010; Yang et al., 2011; Zhang et al., 2011).”
A review on neuroimaging‐based classification studies and associated feature extraction methods for Alzheimer's disease and its prodromal stages (2017, neutral): “(Zhang et al., 2011) 51 99 – – 52 SFS Multi-kernel SVM ADNI 93.”
Integrating Multimodal Priors in Predictive Models for the Functional Characterization of Alzheimer's Disease (2015, methodology): “For instance [14] used a kernel to combine FDGPET, anatomical MRI and CSF biomarker measures to obtain a high AD detection accuracy.”