Reduced neuronal activity in Alzheimer's disease and mild cognitive impairment measured by resting state fMRI: Neuroimaging / Optimal neuroimaging measures for early detection. (7th December 2020)

• Record Type: Journal Article
• Authors: Haddad, Seyyed Mohammad Hassan; Scott, Christopher J.M.; Arnott, Stephen R.; Ozzoude, Miracle; Strother, Stephen C.; Black, Sandra E.; Borrie, Michael; Finger, Elizabeth; Tartaglia, Maria Carmela; Kwan, Donna; Beaton, Derek; Symons, Sean; Soddu, Andrea; Menon, Ravi S.; Montero‐Odasso, Manuel; Bartha, Robert
• Is Part Of: Alzheimer's & dementia. Volume 16(2020)Supplement 5
• Page Start: n/a
• Page End: n/a
• Publication Date: 2020-12-07
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.046067
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: Early pathological changes in mild cognitive impairment (MCI) and Alzheimer's disease (AD) gradually decrease neuronal metabolism and function measured by PET and functional MRI (fMRI). These changes are often associated with cognitive decline and can help in the diagnosis of AD. However more sensitive indicators of the earliest stages of disease must be developed to detect disease prior to cognitive impairment. Recently, a novel neuronal activity (NA) metric was introduced based on resting‐state fMRI (rs‐fMRI) signal oscillations, which showed dramatically lower NA in AD. Here we introduced a more sophisticated and potentially more sensitive NA metric to identify subtle NA differences between three groups: normal elderly controls (NEC), MCI, and AD. Method: The rs‐fMRI scans (TR=∼2400, ∼10 minute acquisition) were acquired at 3T from 8 MRI scanners in the Ontario Neurodegenerative Disease Research Initiative (AD group, N=40, aged 71.8 ± 8.1, 42% female; and MCI groups, N=80, aged 70.3 ± 8.3, 46% female), and the Gait and Brain Study (NEC group, N=30, aged 71.3 ± 6.0, 23.33% female). The rs‐fMRI signal was pre‐processed using FMRIB Software Library (FSL) and decomposed into independent components (ICs) using IC analysis. The ICs were classified into neuronal and non‐neuronal using a support vector machine classifier. The proposed voxel‐wise NA metric was defined based on similarity ( cross covariance ) between rs‐fMRI signal and neuronal ICs. This metric was utilized to create resting‐state NA map in each subject, which were compared between groups. Result: Single subject and group average NA maps are provided in Fig. 1. The % difference maps between each group are shown in Fig. 2. The average voxelwise % difference in NA between groups was 38% between NEC and MCI, 38% between NEC and AD, and 10% between MCI and AD. Average NA in whole gray matter, hippocampus, posterior cingulate cortex, and precuneus, in each group are also provided in Fig. 3. Conclusion: Lower NA was clearly detected in AD and MCI groups compared to NEC. The large decrease in NA detected in MCI subjects compared to NEC suggests this metric is highly sensitive to early changes in neuronal function.