Selecting the most relevant brain regions to discriminate Alzheimer's disease patients from healthy controls using multiple kernel learning: A comparison across functional and structural imaging modalities and atlases

Abstract

BackgroundMachine learning techniques such as support vector machine (SVM) have been applied recently in order to accurately classify individuals with neuropsychiatric disorders such as Alzheimer's disease (AD) based on neuroimaging data. However, the multivariate nature of the SVM approach often precludes the identification of the brain regions that contribute most to classification accuracy. Multiple kernel learning (MKL) is a sparse machine learning method that allows the identification of the most relevant sources for the classification. By parcelating the brain into regions of interest (ROI) it is possible to use each ROI as a source to MKL (ROI-MKL).MethodsWe applied MKL to multimodal neuroimaging data in order to: 1) compare the diagnostic performance of ROI-MKL and whole-brain SVM in discriminating patients with AD from demographically matched healthy controls and 2) identify the most relevant brain regions to the classification. We used two atlases (AAL and Brodmann's) to parcelate the brain into ROIs and applied ROI-MKL to structural (T1) MRI, 18F-FDG-PET and regional cerebral blood flow SPECT (rCBF-SPECT) data acquired from the same subjects (20 patients with early AD and 18 controls). In ROI-MKL, each ROI received a weight (ROI-weight) that indicated the region's relevance to the classification. For each ROI, we also calculated whether there was a predominance of voxels indicating decreased or increased regional activity (for 18F-FDG-PET and rCBF-SPECT) or volume (for T1-MRI) in AD patients.ResultsCompared to whole-brain SVM, the ROI-MKL approach resulted in better accuracies (with either atlas) for classification using 18F-FDG-PET (92.5% accuracy for ROI-MKL versus 84% for whole-brain), but not when using rCBF-SPECT or T1-MRI. Although several cortical and subcortical regions contributed to discrimination, high ROI-weights and predominance of hypometabolism and atrophy were identified specially in medial parietal and temporo-limbic cortical regions. Also, the weight of discrimination due to a pattern of increased voxel-weight values in AD individuals was surprisingly high (ranging from approximately 20% to 40% depending on the imaging modality), located mainly in primary sensorimotor and visual cortices and subcortical nuclei.ConclusionThe MKL-ROI approach highlights the high discriminative weight of a subset of brain regions of known relevance to AD, the selection of which contributes to increased classification accuracy when applied to 18F-FDG-PET data. Moreover, the MKL-ROI approach demonstrates that brain regions typically spared in mild stages of AD also contribute substantially in the individual discrimination of AD patients from controls.

Quantifying Neurodegenerative Progression With DeepSymNet, an End-to-End Data-Driven Approach

“…The pallidum region was found to have a significant difference in beta-amyloid burden between early and late-onset AD (Youn et al, 2017) and differences in RNA binding protein TDP-43 deposits (Josephs et al, 2016). Finally, researchers using different imaging modalities to discriminate AD patients found the pallidum and putamen to be consistently important (Rondina et al, 2018). Various frontal regions were relevant for the model decision; this might represent an advanced disease stage in the selected population.…”

Modified ketogenic diet is associated with improved cerebrospinal fluid biomarker profile, cerebral perfusion, and cerebral ketone body uptake in older adults at risk for Alzheimer’s disease: a pilot study

“…Two CSF analytes (tau and neurofilament light) had non-normal distributions and were log-transformed prior to analysis. For MR gray matter volume values, a meta-ROI shown to discriminate between AD and controls was constructed as the average of the following regions: inferior temporal gyrus, caudate, paracentral lobule, superior temporal pole, posterior cingulate, amygdala, hippocampus, entorhinal cortex, angular gyrus, and mid-temporal pole (Rondina et al, 2018). Main effects of diet, cognitive group, and the interaction of diet by cognitive group on PCASL perfusion were examined utilizing a voxel-wise multiple regression in SPM12.…”

Diagnosis of Alzheimer’s Disease via Multi-Modality 3D Convolutional Neural Network

“…Additionally, Zhang et al (2011) conducted a classification based on MRI, PET, and CSF biomarkers. Moreover, other imaging modalities or PET tracers can be considered, as Rondina et al (2018) used T1-MRI, 18 F-FDG-PET and rCBF-SPECT as the imaging modalities while Wang et al (2016) used 18 F-FDG and 18 F-florbetapir as tracers of PET.…”